ERMA New Zealand Evaluation & Review Report
Application for approval to import for release a moth, the boneseed leafroller (Tortrix s.l. sp. “chrysanthemoides”), that is not currently present in New Zealand for the purpose of biological control of the weed pest boneseed Chrysanthemoides monilifera subsp. monilifera (L.) Norlindh (Family: Asteraceae)
Application code: NOR03001
Prepared for the Environmental Risk Management Authority
Key Issues
The project team notes that no New Zealand natives belong to the same tribe as Chrysanthemoides monilifera subsp. monilifera.
The project team considers that the applicant has followed internationally accepted host-specificity testing protocols.
External experts and submitters questioned the choice of New Zealand natives used in the host-specificity and the project team notes that the selection criteria were not well explained.
The source of plant material was not consistent between the Australian and New Zealand tests which may impact on the interpretation of results.
The use of results from experiments done in environments unlike those experienced in New Zealand may impact on applicability of conclusions to the current application.
Boneseed has been declared an unwanted organism by the Department of Conservation and the project team notes that in the application DoC was identified as a potential participant in the release programme. It is noted however, that DoC have made a submission opposing the approval of the application.
The project team considers that there would be value in discussing these matters further at the hearing.
Page 2 of 84 Table of Contents ERMA New Zealand Evaluation & Review Report 1 1 Introduction 5 1.1 Application Summary 5 1.2 Project team 7 2 Application Process Review 7 2.1 Timeline 7 2.2 Agencies notified 7 3 Information provided 8 3.1 Application 8 3.2 Submissions 8 3.3 External expertise 9 3.4 Additional information 9 4 The decision-making process 9 4.1 General 9 4.2 Analysis and terminology 11 5 Identification of the Organism (including scope) 13 6 Organism Characteristics 14 7 Efficacy of the biological agent 15 8 International Obligations 20 9 Identification and assessment of potentially significant adverse and beneficial effects (risks, costs and benefits) 21 9.2 The environment 21 9.3 Human health 35 9.4 Māori culture 36 9.5 Society and community 44 9.6 The market economy 46 10 Overall Evaluation 47 10.1 Summary of information 47 10.2 Summary of adverse and beneficial effects 48 10.3 Evaluation against statutory criteria 49 10.4 Previous consideration of similar applications 52 11 References 53 Appendix 1: Summary of submissions received 58 Appendix 2: Dr Barbara Barratt’s Report 63 Appendix 3: Dr Ilse Breitwieser’s Report 69 Appendix 4: Decision Pathway 76 Appendix 5: Qualitative scales for risk assessment 79
Page 3 of 84 Appendix 6: Submissions received by ERMA New Zealand 84
Page 4 of 84 1 Introduction
1.1 Application Summary
Purpose of Application
1.1.1 Environment Canterbury seek approval to import for release Tortrix Linnaeus, 1758 s.l. sp. "chrysanthemoides" under section 34(1)(a) of the Hazardous Substances and New Organisms Act 1996 (HSNO).
Biological control programme
1.1.2 The project team notes that the boneseed genus Chrysanthemoides is one of six genera in the tribe Calenduleae (Family: Asteraceae) that are native to South Africa (Weiss, 1986). The genus Chrysanthemoides comprises only two species, C. monilifera and C. incana. However, six subspecies of C. monilifera have been described. Chrysanthemoides monilifera has been introduced into a number of countries including Australia, Italy, St Helena, and France (Weiss, 1986; Scott, 1996). Only one subspecies C. monilifera subsp. monilifera is in New Zealand (Winks & Fowler, 2000).
1.1.3 The applicant states that in New Zealand C. monilifera subsp. monilifera (boneseed) has the potential to cause substantial environmental damage (Syrett 1999). It invades plant communities on coastal cliffs and dune- lands, and inland grassland and shrub areas. Dense thickets of boneseed can form which displace native vegetation and shade out native seedlings. The long-term impact of boneseed on native biodiversity may be much more serious than that of a weed such as gorse because there is little native seedling recruitment under a pure stand of boneseed (McAlpine & Timmins 2002).
1.1.4 In 2001 the Department of Conservation (DoC) declared boneseed an unwanted organism under section 164(c) of the Biosecurity Act, 1993. This was on the basis that it was capable of forming self sustaining populations in New Zealand, taking into account the ease of eradication; and displacing or reducing any native species or any introduced species for which the Department of Conservation is responsible; or causing the alteration or deterioration of natural habitats; or causing adverse effects to New Zealand’s indigenous biological diversity.
1.1.5 Boneseed is located mainly in coastal habitats throughout the North Island, and in Nelson/Marlborough, around Banks Peninsula, and Anderson's Bay in Dunedin and an indicative map of boneseed distribution is on page 6 of the application. Several regional councils and DoC implement active control programmes for boneseed in their area.
1.1.6 The boneseed shrub is a South African native, generally 2-3 m high, with thick, leathery leaves and masses of bright yellow daisy-like flowers, followed by clusters of black fruit that contain a single bone coloured
Page 5 of 84 seed. This hardy weed can grow in a range of environments and is a prolific seed producer (Weiss, 1986).
1.1.7 One submitter (6330) requested some specific information about the weed including lifespan of individuals and stands, does succession occur, plant form in South Africa, and site of colonisation (eg disturbed soil), followed by succession or stability, seed production (numbers and viability and spread). The project team considers that this issue is adequately addressed in sections 7.1.14-7.1.15 and section 2.2 of the application.
1.1.8 One submitter (6330) specifically questioned what information was available to prove that boneseed was spreading and indicated the potential magnitude of the problem and to what extent was displacement of other species occurring. The project team considers that this issue is addressed in section 1.1.5 and comments by submitters 6321, 6322, 6325, 6326 and 6332.
1.1.9 Biological control using boneseed leafroller (the biological characteristics of which are described in Section 6 of this report) in a number of areas is considered the only option for managing this weed pest. This is because herbicide use is often impractical due to damage to native plants and/or inaccessible terrain. In certain areas, control may be possible by stock grazing on small plants but in other areas of steep terrain this is considered neither feasible nor desirable.
1.1.10 The applicant proposes to establish a rearing colony of the boneseed leafroller in South Africa, based on specimens collected from the same area as those used for host range testing. Larvae (5th and 6th instar) will be imported into New Zealand and held in quarantine until clearance is granted by the MAF (as discussed in section 9.2.33. Mass rearing will then be done at Landcare Research, Lincoln, followed by release at sites to be determined collaboratively by Landcare Research, regional councils, the Department of Conservation, iwi throughout New Zealand, and colleagues in South Africa and Australia. The project team notes that the submission by DoC did not support this application so is unclear about their potential involvement in any release programme. Potential release sites are indicated in a map on page 6 of the application in relation to the distribution of boneseed. The project team notes that the printed version of the map does not show the potential release sites clearly but implies that initial releases are likely to exclude coastal Bay of Plenty, Southern Wellington, Marlborough and Otago. It is also noted that the applicant has proposed to do post-release monitoring to assess the establishment and impact of the release biocontrol agent as detailed in section 4.2 of the application.
Page 6 of 84 1.2 Project team
1.2.1 The project team consists of the following ERMA New Zealand staff members:
Dr Sonia Whiteman Project Leader Dr Abdul Moeed Senior Scientific Advisor Dr Fleur François Scientific Advisor Linda Robinson Manager Māori Zack Bishara Māori Advisor Janet Gough Senior Analyst
1.2.2 This report was reviewed by Shaun Slattery (Programme Manager), Dr Geoff Ridley (Science Manager) and Dr Bas Walker (Chief Executive) and reviewed and signed-out by Dr Libby Harrison (Manager, New Organisms) on the 24th of November 2004.
2 Application Process Review
2.1 Timeline
2.1.1 The application was formally received on 19 August 2004 and publicly notified on 2 September 2004. Public submissions closed on 14th October 2004. Additional information was sought under section 58(1) of the HSNO Act as described in section 3.3.1. As a result the hearing was postponed by 11 working days under section 58(3) of the HSNO Act. This Evaluation & Review (E&R) report was circulated 13 days prior to the date of the hearing, which was held on the 13th of December 2004.
2.2 Agencies notified
2.2.1 Various government departments and agencies were notified of the receipt of the application (in accordance with section 53(4) of the HSNO Act), and provided with an opportunity to comment or make a public submissions on the application.
2.2.2 Of the government departments and agencies notified, only two made submissions the Department of Conservation (DoC) and the Ministry of Agriculture and Forestry.
Page 7 of 84 3 Information provided
3.1 Application
3.1.1 The following documents were supplied by the applicant at the time the application was received:
Application NOR03001 to Import for Release or Release from Containment any new organism including a genetically modified organism but excluding conditional release and rapid assessment [Form NOR] Copies of cited literature, as listed in the application. 3.1.2 No confidential information was provided as part of the application or supporting material.
3.2 Submissions
3.2.1 The application was publicly notified on the ERMA New Zealand website, and notices placed in The New Zealand Herald, The Dominion Post, The Christchurch Press and The Otago Daily Times on 8th September 2004. In addition, interested parties identified by ERMA New Zealand were notified of the receipt of the application by letter, and provided with an opportunity to make a public submission on the application. Interested parties were contacted on the basis of their listing on ERMA New Zealand’s database, and they broadly included environmental, scientific and primary production industry groups, as well as individuals who had previously expressed interest in new organism release applications.
3.2.2 A total of 13 submissions were received on this application, of which six submitters indicated that they wished to be heard in support of their submission (refer to Table 1 below). Copies of submissions are provided in Appendix 6 of this report.
Table 1: Summary of submissions received
No. Submitter Contact Hearing Stance 6321 Greater Wellington Dave Bayly No Support Regional Council 6322 Waikato Regional Council David Stephens No Support 6323 Te Runanga O Te Rarawa Anahera Herbert- Yes Conditional support Graves 6324 Wairewa Rūnanga Iaean Cranwell No Oppose 6325 RR Scott RR Scott No Support 6326 EW Coop & Sons RM Coop No Support 6327 Ngati Whatua o Orakei Bernadette Papa No Conditional support Corp. Ltd. 6328 Te Atiawa Manawhenua Ki Mick Park Yes Oppose Te Tau Ihu Trust 6329 Te Mana Taiao o Raukawa Steven Wilson Yes Not stated 6330 Cliff Mason Cliff Mason Yes Oppose without
Page 8 of 84 further information 6331 Department of Leanne Perry- Yes Oppose without Conservation Meyer further information 6332 Te Runanga o Ngai Tahu Jason Holland No Neutral 6333 Ministry of Agriculture and Christina Yes Neutral Forestry Vieglais
3.2.3 A summary of issues raised in submissions is presented in Appendix 1 and these issues are addressed within the report.
3.3 External expertise
3.3.1 Additional information was requested and supplied by Drs Barbara Barratt, of AgResearch, and Dr Ilse Breitweiser, of Landcare Research under section 58(1) of the HSNO Act in the form of comments on the application, submissions and a draft of the E&R report. This additional information was supplied to the decision making committee in the form of the final E&R report. Dr Barratt is a recognised expert in insect biological control agents and Dr Breitweiser is a recognised expert on the phylogeny of the Asteraceae family.
3.3.2 In addition both Dr Barratt and Dr Breitweiser will be attending the hearing.
3.4 Additional information
3.4.1 Additional information was sought from the external experts as described in section 3.3.1 above under section 58 of the HSNO Act. The reports of the two external experts, Drs Barratt and Breitweiser, are attached in Appendix 2 and 3 respectively and noted in this report as appropriate.
4 The decision-making process
4.1 General
4.1.1 The application was made for a full release and is to be considered via the decision path for application to import for release or release a new organism from containment (application made under Section 34 of the Act and determined under Section 38 of the Act). A copy of the relevant decision path is included in Appendix 4 of this report. The source of this decision path is the ERMA New Zealand Protocol Policy Series: Protocol 2: Figure 5.
4.1.2 Under section 38B of the HSNO Act, when an application for full release is made, in circumstances where deemed appropriate the Authority may offer the applicant the option of changing the application to a conditional release. The decision to treat a release application as a conditional release
Page 9 of 84 will in most circumstances be made prior to public notification of the application.
4.1.3 The option of a conditional release was presented to the applicant, however as the objective of the application is to establish self-sustaining populations in all areas of New Zealand affected by boneseed, the mobility of the boneseed leafroller would prevent recovery after release even if it were released only in certain restricted areas. The applicant considered that no controls can be envisaged that would mitigate any risks associated with the release of the organism or enhance the possibility of benefits being realized. The project team agrees that conditions would not add anything to the management of risks. Therefore, the project team agrees with the applicant that full release is an appropriate application path.
4.1.4 The Authority has published a policy position on processing and considering applications for conditional release.
4.1.5 This policy addresses switching a release application to a conditional release application. This position is that the Authority will allow an application for release to be changed so that it will be considered as a conditional release up to the point of public notification. Thus if the Authority forms the view that a conditional release application is more appropriate than an unconditional release application prior to public notification, it will seek agreement for the change from the applicant.
4.1.6 Once an application has been publicly notified as a full release the Authority cannot request the applicant to change the application to a conditional release. However, the applicant may choose to request a change at any time. If this occurs prior to the commencement of the public hearing, the Authority will agree to the consideration of conditions, but will normally require the application to be re-notified as a result to provide a fair opportunity for submitters to re-submit accordingly. If the request is lodged after the start of the public hearing, the Authority is more likely to ask for the application to be withdrawn, and for a new application to be made.
4.1.7 One submitter (6323) requested that if the Authority approved this application that controls requiring that any user of the approval consult with local groups prior to doing so. This issue will addressed further in section 9.4.41. Another submitter (6331) suggested that the applicant should initially release a small number of leafrollers in a few isolated areas to allow monitoring for adverse effects to be done, such that in the event of such effects the organism could be recalled. The project team is of the opinion that based on the biology of the organism eradication would be difficult. The project team notes that controls cannot be applied
4.1.8 One submitter (6325) requested that post-release monitoring be a requirement of the approval. Submitter 6331 noted concerns that unless monitoring was made a condition of the release as a legal requirement, it would not occur. The project team again notes that controls can not be
Page 10 of 84 applied to full release approval. This is discussed further in section 4.1.1- 4.1.4 of this report.
4.1.9 Two submitters (6330 and 6331) also requested further details of any monitoring programme. The project team notes that should the Authority approve this application they may wish to recommend to the applicant to complete and undertake the proposed monitoring programme.
4.1.10 In response to these questions about the proposed monitoring regime the applicant provided the following information. It is important that biological control programmes are effectively monitored after the release of agents and if approval is given for the release of the boneseed leafroller in New Zealand a detailed monitoring programme will be devised to assess its establishment and impact. When choosing initial release sites for the boneseed leafroller, consideration will be given to the monitoring programme. Releases will be made at selected sites, and other matched sites will be randomly allocated as “control” sites. All boneseed leafroller release sites will be visited at least once during the first two years after release to record the number of boneseed leafrollers found (adults, larvae, or pupae) and the distances that these were found from the point of release. More intensive monitoring to gain quantitative information on the impact that the boneseed leafroller is having on boneseed will be carried out at selected sites. The number of sites selected for intensive monitoring will be finalised once it has been confirmed what resources will be available for post release monitoring. Monitoring will be done by Landcare Research staff in collaboration with staff from regional councils, the Department of Conservation, and iwi throughout New Zealand. Two biological control agents have recently been released in New Zealand for mistflower. A white smut fungus, Entyloma ageratinae, was released in 1998, and a gall fly, Procecidochares alani, was released in 2001. Considerable effort has been put into a monitoring programme for these two agents and results of the monitoring, to June 2003, are reported in a Landcare Research contract report (Barton et al. 2003). If approval is given for the release of the boneseed leafroller in New Zealand a monitoring programme of a similar or greater scale is envisaged.
4.2 Analysis and terminology
4.2.1 The HSNO Act and the HSNO Methodology Order 1998 (the Methodology) require the Authority to consider all adverse and beneficial effects associated with the organism and the application.
4.2.2 The identification and assessment of adverse and beneficial effects is structured such that the consideration of adverse effects includes discussion of risks and costs, while the consideration of beneficial effects includes benefits. Following the Methodology, risk means the combination of the magnitude of an adverse effect and the probability of its occurrence. Cost is described as “the value of a particular adverse effect expressed in monetary or non-monetary terms”, and benefit as “the
Page 11 of 84 value of a particular positive effect expressed in monetary or non- monetary terms”. Each risk will have an associated cost.
4.2.3 Effects are not certain. All effects (risks, costs and benefits) will be considered in terms of the likelihood of their occurrence as well as the magnitude of their effect if they should occur. Where possible and relevant uncertainty is discussed.
Identifying adverse and beneficial effects
4.2.4 Identifying adverse and beneficial effects requires identifying the sources of effect, the areas of impact (what might happen), and the exposure pathways (how the source and area of impact might intersect). Identifying the source of effect (risk, cost or benefit) requires analysing the potential hazard or benefit, and examining the incidents or triggers that may release this intrinsic potential.
4.2.5 All identified effects are discussed in terms of their area of impact. The relevant areas of impact, based on the requirements of the HSNO Act, are effects on the environment, human health, Māori culture, society and community and the economy
Assessing adverse and beneficial effects
4.2.6 The project team has evaluated the available information on the potential adverse and beneficial effects of the application and addressed them in terms of the magnitude of occurrence of the effect should it eventuate (ie a measure of the size and distribution of the effect) and the likelihood of the effect occurring.
4.2.7 The likelihood of the effect being realised has been measured on a seven point qualitative scale: highly improbable, improbable, very unlikely, unlikely, likely, very likely and extremely likely. The magnitude of the effect has been measured on a five point qualitative scale: minimal, minor, moderate, major and massive. The level of risk (combined likelihood and magnitude) can then be calculated. The six levels (or categories) of risk are: A, B, C, D, E, and F where A may be deemed to be equivalent to “insignificant”. These scales are for comparative purposes only, and only relate to this application. Appendix 5 of this report contains more complete description of these qualitative scales. Qualitative measures have been used because there is insufficient quantitative data to generate useful quantitative measures.
Approach to risk
4.2.8 Clause 33 of the Methodology lists the risk characteristics that the Authority will have regard to in determining its approach to risk. These risks are whether exposure to the risk is involuntary, the risk will persist over time, the risk is subject to uncontrollable spread and is likely to extend its effects beyond the immediate location of incidence, the potential adverse effects are irreversible and the risk is not known or
Page 12 of 84 understood by the general public and there is little experience or understanding of possible measures for managing the potential adverse effects. Where risks are assessed as being significant the project team has discussed these characteristics.
5 Identification of the Organism (including scope) 5.1.1 According to the application, the species of the boneseed leafroller has not yet been formally described. It was initially identified as belonging to the genus Tortrix but more recent examination of material by a tortricid expert (Dr Marianne Horak, CSIRO, Australia) has indicated that it almost certainly does not belong to the genus Tortrix (Edwards, 1993). As a result the species is currently being referred to as Tortrix s.l. sp. "chrysanthemoides", s.l. meaning “sensu lato” or “in the wider sense”. The use of “chrysanthemoides” identifies the organism at the same level as a species that is associated only with the plant genus Chrysanthemoides. This species has also been referred to as “Tortrix” sp. in the literature (Scott & Adair, 1990; Scott & Brown, 1992).
5.1.2 Dr Marianne Horak is satisfied that the boneseed leafroller is a specific entity. The applicant contends that while it would be possible to give this species a formal name now, the opinion from taxonomists is that this would not be appropriate. Dr Horak is of the view that, rather than rush to give this species a name, it would be better to wait for a revision of the African Archipini (the tribe to which the boneseed leafroller belongs) to establish the generic placement and formally describe the species. Jozef Razowski, a lepidopterist in Krakow, Poland, is currently working on a revision of the African Archipini but this is likely to take quite a few years to complete.
5.1.3 According to the applicant, preserved boneseed leafroller specimens have been deposited in the New Zealand Arthropod Collection (Landcare Research) and these specimens have been examined by John Dugdale, a tortricid (leafroller) expert as being distinguishable from other tortricids (see Appendix 7 of the application for John Dugdale comments). Further the applicant notes that that Mr John Dugdale (ex Landcare Research, now retired), a tortricid expert in New Zealand, agrees with Dr Horak that there is no scientific point in manufacturing a name, especially as nomenclatural changes will undoubtedly occur later. Providing a name now will only reiterate a status that is already encapsulated in the name “boneseed leafroller”. The entity of choice here, that feeds only on the shrub genus Chrysanthemoides, can be defined by its host plant and there is clearly sufficient evidence from thorough ecological studies that this entity is distinct from other related species.
5.1.4 The applicant states that given the way boneseed is spreading in New Zealand, the regional councils wish to proceed with a biological control programme without further delay and contends that this is a sensible
Page 13 of 84 decision, and that the proposed biocontrol agent is a valid species despite the lack of a formal genus and species name.
5.1.5 It should be noted than none of the 13 submitters raised any concerns about the lack of a formal binomial, in fact one submitter (6325) considered “the biological material proposed for introduction comprises a good species”.
5.1.6 The project team concurs with the applicant and John Dugdale and notes that boneseed leafroller Tortrix s.l. sp. "chrysanthemoides" uniquely identifies the organism to be imported for release. Should the Committee decide to approve this application this unique identification would be used in ERMA New Zealand’s register of organisms.
6 Organism Characteristics 6.1.1 Boneseed leafroller Tortrix s.l. sp. "chrysanthemoides" is a moth of the family Tortricidae. New Zealand has over 200 species in the Tortricidae family and many of them are endemic (Dugdale, 1988). Tortricids are often called bell moths because when at rest the wings of many species form a neat bell shape. The caterpillars are fast movers and can move in reverse. Some tortricid caterpillars tunnel into fruit, seeds, or soft stems (Crowe, 2002). Most leafroller caterpillars make themselves a safe home by rolling or folding up leaves and tying them with silken webbing that they produce from their mouths. Formerly, Māori looked out for these folded leaves, to collect and eat the caterpillars, naming them moka or tikopa (from kopa meaning bent or folded).
6.1.2 Boneseed leafroller is native to South Africa. Previously, the boneseed leafroller was identified as Tortrix capensana (Walker), a pest of fruit crops but morphological and host specificity studies have clearly demonstrated that they are two different species (Scott & Brown, 1992).
6.1.3 Morphologically, adult forewing colour is variable but usually mottled earthy brown with a darker narrow zigzag band extending at around 60 degrees to the costa. Mean forewing length 9.5 mm (range 7.9–10.6 mm) (n = 30). Forewings are narrowly triangular with an acute apex (wing tip). Forms with uniformly mottled forewings also frequently occur. The hindwing is bone coloured with weakly marked eye spots towards the outer margin.
6.1.4 When first laid, eggs are pale yellow turning to orange, flat and oval (length 0.7– 1.4 mm, width 0.6–1.1 mm) with reticulate surface ridges. The area between the ridges is covered in dense scurf of crystal-like scales, a distinctive feature of boneseed leafroller that sets it apart from other species. The egg batch consists of rows of eggs side by side, forming an irregular shape, and is covered with a transparent film secreted by the female during egg laying. The number of eggs laid per female is about 200 (Scott and Brown, 1992).
Page 14 of 84 6.1.5 First and second instar larvae are olive green with indistinct tuberculae (raised, hardened skin patches appearing as white spots) on all abdominal segments (Scott & Brown, 1992). In later instars (third to sixth) the head, thoracic shield, and anal plate are black, and the abdomen is dark olive brown to black, paler on the underside, with white tuburculae in parallel rows along the length.
6.1.6 Pupae are light to dark brown, average length 10.6 mm (range 8.3–12.0 mm) (n = 96), with a row of small spines on the upper side of each segment of the abdomen, and hooked spines on the anal segment.
6.1.7 The boneseed leafroller is multivoltine with three generations per year. Development time from hatching to the emergence of adults takes about 6 weeks at 20-22ºC and adults live for 10-12 days.
6.1.8 In South Africa, high rates of parasitism have been recorded in eggs and larvae of the boneseed leafroller (Edwards, 1993). The potential impact of parasitism on efficacy of the organism as a biological control agent will be discussed in sections 7.1.17-7.1.19 of this report.
6.1.9 The boneseed leafroller has the capacity to defoliate the whole plant by webbing together older leaves down the stems (Edwards, 1993).
6.1.10 In South Africa, the boneseed leafroller is restricted to all six subspecies of C. monilifera and the related C. incana (Scott & Brown, 1992). Data on host specificity of this insect will be assessed in sections 9.2.16-9.2.21 of this report
7 Efficacy of the biological agent 7.1.1 The project team identified the following matters.
Establishment and efficacy prediction
7.1.2 The project team notes that important factors in making good predictions on establishment and efficacy of a biological control agent include the agent’s previous history of establishment overseas and/or assessment of related biological control programmes, and a comprehensive knowledge of the biology and climatic requirements of biological control agent. It is given that climate to some extent is different between the boneseed leafroller’s native habitat and its proposed introduction into New Zealand. These differences are likely to have an effect on its establishment and on the number of generations per year but this is unlikely to have an effect on its efficacy as biological control agent.
7.1.3 One submitter (6331) noted that an overall assessment of likelihood of establishment had not been done and concluded that on the basis of the Australian results establishment would be poor and considered that this conclusion should be considered carefully in the overall cost/benefit
Page 15 of 84 analysis. The project team notes that the extent of establishment will not be known until the organism is released.
7.1.4 The applicant notes that the intended release programme involves large numbers of the boneseed leafroller being released over a wide geographical range in New Zealand and it is likely that they will establish. The project team notes that it is difficult to predict at this stage how successfully the boneseed leafroller will establish. The issue of establishment is likely to become clearer after post release monitoring as proposed by the applicant in section 4.2 of the application.
7.1.5 With respect to efficacy, the application notes that although it is not possible to predict with certainty the efficacy of a specific biocontrol agent, some insight into its potential performance can be gained from experience overseas, both in its country of origin and in countries to which the agent has been introduced. The boneseed leafroller is considered to be the most damaging insect found on the two species of Chrysanthemoides (C. monilifera and C. incana) in South Africa, and has a significant influence on their population dynamics (Scott & Adair, 1990). Boneseed leafroller outbreaks occur periodically on Chrysanthemoides in South Africa, and can lead to complete defoliation and death of plants (Edwards, 1993). Persistent defoliation by the boneseed leafroller is expected to reduce the vigour of boneseed in New Zealand by lowering seed production and shoot and root growth rates, and sometimes killing entire bushes, as occurs in South Africa. The reduced "fitness" and lowered canopy cover of boneseed are expected to facilitate the establishment of co-occuring native plants in New Zealand that are presently being out-competed by boneseed.
7.1.6 The boneseed leafroller was released in Australia in April 2002. It is difficult to assess the efficacy of the organism in this release because it was released too recently and post-release monitoring is still in its early stages (Edwards & Adair, 1999; Ireson et al., 2002).
7.1.7 There is some evidence of predation of the boneseed leafroller by Argentine Ants occurring at some release sites (Ireson et al., 2002). The potential impact of predation on the efficacy of the biological control agent in New Zealand is discussed further in sections 7.1.20-7.1.24 of this report.
7.1.8 One submitter (6328) did not support the release of boneseed leafroller due to its apparent failure in Australia. While another submitter (6330) requested a statement of the success of the Australian release and if it had been or was being considered for release elsewhere. The project team notes that the Australian release is still in its infancy so conclusions about efficacy are premature.
7.1.9 The project team notes that determination of efficacy of the biological control agent is relevant to the consideration and evaluation of all effects. The project team also notes that the ultimate measure of a successful biological programme is that, following establishment of the biological
Page 16 of 84 control agent, the weed no longer presents any economic or environmental risk and that the weed remains under sustained control (see Syrett, et al., 2000).
7.1.10 The project team also notes that the evidence suggests (see Scott & Brown, 1992; Edwards, 1993) that provided successful establishment of a significant population of boneseed leafroller occurs it is likely to have an adverse effect on the boneseed in New Zealand. The extent of this adverse effect will only be known after the event.
7.1.11 The project team also notes that the applicant in not anticipating 100% success but is expecting a significant reduction in the density of boneseed which will reduce the overall cost of control.
Effectiveness of the biological control agent in comparison with other methods
7.1.12 Boneseed (Chrysanthemoides monilifera subsp. monilifera) is a vigorous perennial that is tolerant to frost and salinity and can grow in the open or in the shade of other plants (Weiss, 1986). Boneseed can be controlled with herbicides such as glyphosate, Tordon 50D (picloram plus 2,4-D), 2,4-D amine, bromoxynil, triclopyr and metsulfuron methyl. However none are selective for boneseed (Weiss, 1986; Edwards, 1999). Destruction of boneseed requires repeated applications of herbicide. Some areas of infestation are inaccessible to herbicide application and concerns about risks to non-target plants and human health restrict the use of herbicides to relatively small infestations (Edwards, 1999). Therefore, the project team considers that herbicide treatment alone will not be more effective than the potential success of a biological agent such as boneseed leafroller in controlling boneseed infestations over the long- term in New Zealand. The implications of this application for reducing herbicide use are discussed in section 9.2.54-9.2.57 of this report.
7.1.13 One submitter noted a preference for the use of herbicide over the introduction of biological control agents (6324) and another submitter (6330) questioned what quantitative ecological data was available to assess the impact of herbicide use for boneseed control. Four submitters (6321, 6322, 6325 and 6326) noted that herbicide control was not a practical option in their areas so supported the importation and release of the leafroller on the basis that it was the only option.
7.1.14 Boneseed can be killed by fire but fire also facilitates germination of the plant from soil-stored seed (Scott, 1996). Therefore, frequent burning is required to kill the regenerating seedlings, and is either detrimental to neighbouring native plants or infeasible in many areas where boneseed is a problem in New Zealand. Hand-pulling is an effective control method for boneseed which has a shallow and weakly developed root system (Edwards, 1999). However, it requires a high labour input, is only applicable to small infestations and is impossible in some areas such as on steep cliffs. Therefore, the project team consider that fire or hand- pulling are only effective for small infestations of boneseed and will not prove more effective for the long-term control of boneseed (especially in
Page 17 of 84 inaccessible areas) than the potential success of a biological control agent.
7.1.15 In Australia, viable seed densities of 800 – 2500 m-2 have been recorded and boneseed can form a seed bank with 13% of the seed viable after three years (Scott, 1996). Seed production is seasonal which implies that seed predators would require an alternative host or aestivating or hibernating phase in their life cycle. The project team considers that based on the extent of the boneseed seed bank and requirement for alternative hosts, use of seed predators such as Mesoclanis polana (previously released in Australia) are unlikely to be more effective as single biological control agents in New Zealand than a defoliating agent such as the boneseed leafroller.
Resistance of New Zealand boneseed to boneseed leafroller
7.1.16 The project team notes that according to the available information, boneseed leafroller exerts a degree of control on boneseed in their native South Africa (Edwards, 1993). If boneseed was to develop resistance to the leafroller, this would have already occurred in its native range. The project team agrees with the assessment in the application that there is a hypothetical but remote possibility that boneseed could develop resistance to the boneseed leafroller, but the agent is also capable of evolving so as to track changes in the host organism. Evolution of new resistant forms of weeds has not been recorded in weed biological control programmes. The risk is thus considered to be insignificant and therefore this issue has not been discussed further.
Impact of parasitism or predation
7.1.17 The applicant notes that during a survey of invertebrate fauna associated with boneseed, parasitoids were reared from a variety of caterpillar species collected from boneseed, including tortricid species (Winks & Fowler, 2000). Some of these parasitoids are likely to affect the boneseed leafroller if it is released in New Zealand.
7.1.18 In South Africa the boneseed leafroller is subject to high levels of parasitism by an egg parasitoid (Trichogramma sp.), by 11 species of larval parasitoids, and a pupal parasitoid, Arctia sp. (Tachinidae) (Scott & Brown, 1992). Egg parasitism by Trichogramma sp. may reach levels greater than 90% in the second and third generations of the boneseed leafroller each year, and larval parasitism can range from 30 to 70% over most of the year. It is considered that a temporary escape from the parasite load contributes to the observed outbreaks in South Africa (Edwards & Adair, 1999). Quarantine containment procedures as described in section 9.2.33 of this report will ensure that these parasitoids are not co-introduced with the boneseed leafroller into New Zealand.
Page 18 of 84 7.1.19 The project team notes that existing parasitoids in New Zealand may attack the boneseed leafroller but the impact is difficult to predict, but it is likely to be similar to attacks on other caterpillars. Although the extent of parasitism of the existing caterpillars in New Zealand is not known, it is unlikely to be as much as to completely destroy their existence. It is noted that the regulation of pest populations by natural enemies is based on the ecological concept of density-dependence, whereby the predator populations react to pest population levels so that a fluctuating but stable equilibrium is reached. The density-dependent nature of the parasitoid- host interaction means that parasitoids were not expected to cause extinction of host populations. It is concluded that if parasitism of the boneseed leafroller does occur it is unlikely to result in total ineffectiveness of the biological control agent.
7.1.20 The boneseed leafroller spends large parts of its life cycle exposed on foliage of boneseed which makes them vulnerable to predation. Field observations and natural enemy exclusion studies have demonstrated that predation by ants, mites and spiders have a significant effect on the survival and establishment of the boneseed leafroller in Australia (Ireson, et al., 2002). In particular, predation by Argentine ants, has been identified as a key factor limiting the establishment of the boneseed leafroller in the Australian biocontrol programme.
7.1.21 Argentine ants, Linepithema humile, were recorded from three sites during the survey of the invertebrates associated with boneseed in New Zealand. It was noted that at the three sites, where large numbers of Argentine ants were present, scale insects (Coccidae) were particularly common. Argentine ants are known to farm sap-feeders, such as scale insects, by providing protection from predators and parasites, in order to gain access to their carbohydrate-rich secretions (honeydew) (Gullan, 1997). Argentine ants can reach very high densities in New Zealand, and they are able to out compete most other ant species (Harris, 2001). As well as feeding on carbohydrate, such as honeydew, Argentine ants feed on protein, which could take the form of a wide range of invertebrates. It was noted that where there were large numbers of Argentine ants, the numbers and diversity of other invertebrates (apart from scale insects) was greatly reduced.
7.1.22 Judging by experiences in Australia with ants, the future distribution and abundance of Argentine ants in New Zealand could be a factor in the establishment of the boneseed leafroller as an effective biological control agent. However, the extent of this effect is difficult to predict prior to release. Only post-release monitoring proposed by the applicant would elucidate this issue. It is concluded that Argentine ants are unlikely to result in total ineffectiveness on the boneseed leafroller
7.1.23 Don (1974) in studies of New Zealand ants considered 10 species out of a total of 33 established species to be native and distributed widely on the main islands and many offshore islands. Earlier, Brown (1958) observed 10 native species and suggested that a cooler climate may be responsible for the lack of species diversity of ants in New Zealand. Moeed & Meads
Page 19 of 84 (1985, 1986) found six native ant species in native forests near Wellington, five in the lowland forest and only a single species at above 700 meters above sea level, and concluded that if the lowland forest is the most suitable habitat for ants then the small number of ant species in New Zealand could be a reflection of the cooler temperatures of our forests. In comparison, habitats in lowland forests are warmer than in higher altitude forests. The project team concludes that it is very unlikely that any of the native ants would contribute to the decline in numbers of the bone seed leafroller or would benefit appreciably from it to increase their population densities.
7.1.24 Other generalist insect predator species such as spiders, ladybirds, predatory shield bugs, earwigs, wasps and praying mantids, have been found on boneseed in New Zealand (Winks & Fowler, 2000). The project team considers that the combined effect of generalist predators could potentially inhibit the effectiveness of the boneseed leafroller but again the extent of this effect is difficult to predict. However, it is considered unlikely to result in total ineffectiveness of the biological control agent.
7.1.25 The applicant noted that an introduced plant-feeding species might affect existing food webs if its population increased to the extent that it became an important alternative host for parasitoids, predators, or diseases that might in turn outbreak, and have an abnormally high impact on their usual hosts. However, the introduced species could also possibly enhance the populations of rare native parasitoids and predators. It is likely that any effects on existing food webs would be minimal as the boneseed leafroller is likely to make up only a small proportion of the total available prey.
7.1.26 This issue was also of concern to three submitters (6323, 6331 and 6333) one of whom who felt that further information on predation was required before approval could be given. Two submitters noted the potential impact of Argentine ants (6331 and 6333) on the establishment of boneseed leafroller and one submitter (6331) had concerns that the provision of a new food source would increase the Argentine ant population. This issue is addressed in sections 7.1.20-7.1.26 above.
Competition and complementary action between biological control agents
7.1.27 The project team notes that boneseed leafroller Tortrix s.l. sp. “chrysanthemoides” is the first biological control agent proposed for release to control boneseed in New Zealand therefore competition and complementary action between biological control agents is not an issue.
8 International Obligations 8.1.1 The project team is note aware of any international obligations that affect the importation for release of the boneseed leafroller.
Page 20 of 84 9 Identification and assessment of potentially significant adverse and beneficial effects (risks, costs and benefits) 9.1.1 The project team identified potential effects related to the application, based on the requirements of the HSNO Act and the Methodology, in a brainstorming session and by reviewing the application, submissions and expert advice sought by ERMA New Zealand.
9.1.2 The potential for adverse or beneficial effects to be realised is based on an assumption of establishment of a self-sustaining population of the boneseed leafroller.
9.1.3 The applicant identified effects associated with the proposed introduction of boneseed leafroller (see Tables 2-9 of the application). The information provided is considered by the project team to be adequate and inclusive.
9.1.4 Where identified effects are deemed to be potentially significant, these have been assessed in terms of the likelihood of their occurrence and the magnitude the effect if it should occur (see section 4.2 of this report).
9.2 The environment
Adverse effects
Identification - Potential impacts on non-target plants
9.2.1 The most important risk identified in the application is potential damage to non-target plants. As stated by the applicant, testing of every possible species against a proposed biocontrol agent is not possible and an internationally recognised procedure has been devised to determine whether the proposed new organism is safe to release. Potential control agents were tested with a range of plant species selected according to criteria established by Zwölfer & Harris (1971) and Wapshere (1974, 1975), and revised by Forno & Heard (1997). The predictability of this procedure is well established and unexpected effects have been minimal. Over 1,150 releases of more than 365 species of insects and pathogens for biological control of 133 weeds in 75 countries have been made (Julien & Griffiths, 1998), and where the protocols have been applied correctly, impact on non-target plants has been negligible (McFadyen, 1998). Dr Barratt noted that while this procedure had been used in selecting the plants used in the Australian tests the applicant had not applied this procedure to testing done on New Zealand plants.
9.2.2 The boneseed leafroller was selected as a potential biocontrol agent for boneseed in New Zealand because it is regarded by the applicant as being specific to the two species of the genus Chrysanthemoides and the risk to other biota in New Zealand is considered to be negligible. This high specificity was determined firstly by thorough ecological studies. Extensive quantitative searches were undertaken by the Australian
Page 21 of 84 CSIRO Biological Control Unit based at the University of Cape Town, South Africa, for the Australian biological control of boneseed programme. Searches were undertaken in South Africa over a 2-year period on 126 plant species growing in close proximity to C. monilifera and C. incana plants infested with the boneseed leafroller. It was found that the leafroller did not attack any other plant species, even in situations where numbers were high enough to severely defoliate or kill the nearby Chrysanthemoides hosts (Edwards & Adair, 1999).
9.2.3 Extensive host-range testing has been conducted on the boneseed leafroller for the Australian biological control of boneseed programme, including laboratory larval development tests (choice and no-choice) and open-field testing in South Africa (Edwards & Adair, 1999). Ninety-six plant species from 31 families were tested, including 50 species from the family Asteraceae to which boneseed belongs (Edwards & Adair, 1999).
9.2.4 In laboratory tests, feeding occurred on a number of non-target test- plants, a result that was considered inconsistent with the high level of specificity observed in the field in South Africa. This necessitated the use of extensive open-field testing in South Africa to examine the oviposition behaviour of adult females and the host choice behaviour of larvae (Edwards & Adair, 1999). The conclusion drawn from these field tests was that the boneseed leafroller is specific to the two species of the genus Chrysanthemoides (C. monilifera and C. incana). Non-target oviposition and very limited survival of larvae may occur on Calendula (marigolds) in the field but only in situations where these plants grow in very close proximity to C. monilifera (Note: Calendula is in the same tribe (Calenduleae) as C. monilifera).
9.2.5 In the field, insects go through two stages of host selection, choice of oviposition site by females, and choice of feeding site by larvae. Extended host ranges under laboratory conditions are not unusual in the evaluation of biocontrol agents (Cullen, 1990; Wapshere, 1989). Laboratory testing conditions can remove essential cues required by the agent for host selection and produce a physiological feeding range that exceeds the natural range. Laboratory tests are generally considered ‘worst case’ while data from field tests are more realistic.
9.2.6 Open-field testing provides a sensitive and powerful tool for host specificity determination and is often the only means of clarifying ambiguous laboratory results (Clement & Cristofaro, 1995; Briese, 1999). However, open-field tests are not commonly undertaken because of the much higher expense involved compared with laboratory testing procedures. In the case of the boneseed leafroller, open-field testing was the only method of ascertaining the true feeding range of this species.
9.2.7 The host specificity of the boneseed leafroller was further tested during 2001, using an oviposition field test, to determine its suitability as a biocontrol agent for boneseed in New Zealand (Winks et al., 2002). Landcare Research of New Zealand carried out this work in South Africa, in conjunction with the CSIRO Biological Control Unit based at the
Page 22 of 84 University of Cape Town, South Africa. Ten test-plant species plus boneseed were used in the host range testing for New Zealand (refer to Table 1 of the application).
9.2.8 Potential impacts on non-target plants was also an issue raised by several submitters. Three submitters (6324, 6327 and 6331) opposed the approval of the application on this basis, particularly due to concerns about the extrapolation of data obtained overseas to New Zealand conditions. In response to these comments Dr Barratt responded that there is a risk that host range tests carried out overseas will not be indicative of results that might be obtained in New Zealand because of any number of differences in biological, ecological and physical conditions. However, we are clearly unable to carry out tests in the field in NZ pre-release, so it is frequently argued that tests in the field in the country of origin of the proposed biological control agent are likely to give more realistic data on potential host range than tests in the laboratory in New Zealand.
9.2.9 One submitter (6332) while remaining neutral on the application noted concerns about non-target effects. One submitter (6329) felt the evidence presented did not confirm that non-target effects would not occur and had concerns about the feeding habits of the leafroller following natural genetic mutation. Submitters 6331 and 6333 noted that a no choice field test was required to determine if adaptation to these plants would occur in the absence of the host and requested further testing (see section 9.2.11 of this report). Submitter 6331 noted that this should be done based on the criteria described by Wapshere and Withers choice and non-choice methods in a field trial. The same submitter noted that the Australian testing was “more extensive and was of a higher scientific standard than the Landcare” research. Finally this submitter also noted concerns about the ability of the leafroller to have non-target effects when in the absence of the host and felt that longterm this could cause selection pressure that would result in a non-target favouring population of boneseed leafrollers. One submitter (6325) noted that it was their opinion that the data presented indicated that the leafroller would be highly host-specific.
9.2.10 In response to these issues raised by submitters Dr Barratt noted that no- choice tests are more conservative than choice tests because they determine the range of species upon which a herbivore is able to develop physiologically. These tests are usually carried out initially on a range of test plants chosen using the centrifugal phylogenetic testing protocol to establish a profile of potential host species. Choice tests are then often used to add more information on host preferences, especially using test plant species that were attacked in the no-choice tests. Oviposition tests are often considered to be more indicative of ecological host range than larval feeding tests. In this case no-choice tests were carried out in the laboratory by the Australian but not the New Zealand researchers. The former showed that full development of boneseed leafroller occurred on 36 species of plants in 11 families. The choice-testing carried out indicated quite clearly that boneseed leafroller has a strong preference for boneseed when that species is present. However, choice tests also showed
Page 23 of 84 that oviposition would occur occasionally on other species or even non- plant material in the presence of boneseed. Similarly choice tests showed that larvae were able to develop fully on alternative species. These results in combination indicate that there must remain a degree of uncertainty in predicting post-release host specificity of boneseed leafroller in New Zealand.
9.2.11 For a response on whether the absence of the host plant induce selection pressure for the boneseed leafroller to attack non-target plants see comments from Dr Barratt in Appendix 2. The project team notes that the scenario will be difficult to test in the laboratory and can only be tested post release in view of the fact that conditional release is not a practical option taking into account the difficulties associated with containing boneseed leafroller in a given area outside of a containment structure.
9.2.12 One submitter (6328) requested that further testing be done based on plant species selected by iwi groups (this is discussed further in section 9.4.28). One submitter (6330) requested that further testing be done on members of the Myrtaceae family and the Leptinella, Olearia, Senecio and Linum genera due to feeding/life-cycle completion observed on related non-target hosts in the laboratory. One submitter (6333) noted that there had been “insufficient host specificity testing on New Zealand native plant species” and provided a list of candidates. One submitter (6331) felt that there were deficiencies in the quantity and quality of the host testing on New Zealand’s indigenous flora and suggested a further 29 species. Phylogeny of bone seed and its relationship is commented upon by Dr Breitwieser in her report (Appendix 3), which concludes that there are no native plants in the same tribe to which boneseed belongs. For additional comment on this issue please refer to section 9.2.13 of this report.
9.2.13 In response to questions about the choice of test plants the applicant responded that the list was developed in consultation with Landcare Research botanist Dr I. Breitwieser. The Asteraceae family is divided into three subfamilies, the Barnadesoideae, the Asteriodeae, and the Cichorioideae. There are no representatives of the subfamily Barnadesoideae in New Zealand. For an insect such as the boneseed leafroller, where a high degree of host specificity is expected, testing would normally concentrate on plants of importance to New Zealand (as natives, crops, or ornamentals) from the same subfamily (Asteroideae) and tribe (Calenduleae) as the target weed. There are no native New Zealand representatives of the tribe Calenduleae, and the two most commonly grown ornamentals from the tribe, Calendula officinalis (marigolds) and Osteospermum fruticosum (dimorpotheca), were both tested for the Australian biological control programme. Representative genera were selected from other tribes in the Asteraceae for the New Zealand host specificity testing, and particular emphasis was placed on the subfamily Asteroideae to which boneseed belongs. The results of this field test (Winks et al. 2002 – Appendix 1 of the application) add to the evidence already available from field observations in South Africa, and
Page 24 of 84 host specificity testing conducted for Australia, that the boneseed leafroller is expected to feed only on boneseed in New Zealand.
9.2.14 The project team notes that Dr Breitwieser was employed by ERMA New Zealand to comment on the choice of New Zealand plants used in the host-specificity testing, so it may appear from the applicants comments that she is critiquing her own work. However, Dr Breitwieser contends that she only has a vague recollection of the consultation referred to by the applicant and that it was limited to a discussion about the Asteraceae family and tribal relationships for a range of Asteraceae species. Dr Breitwieser did not specifically comment on the individual species that were subsequently selected for testing.
9.2.15 The project team is not aware if there are ecotype differences in the boneseed leafroller populations that may affect host range of the insects to be released and therefore the issue of ecotype has not been discussed further.
Assessment - Potential impacts on non-target plants
9.2.16 The project team assessed the adequacy of the testing regime based on internationally acceptable regime and considers that the use of open field testing in South Africa was worthwhile to further illustrate some of the anomalies that can arise in testing carried out in containment. According to Marohasy (1998) some weed biological control practitioners consider that host specificity tests in the laboratory often over-estimate host range. This may lead to the rejection of candidate agents that would be adequately specific under field conditions (see Dunn, 1978; Marohasy, 1996; Wapshere, 1989; Cullen, 1990). In contrast, other biologists are becoming increasingly concerned about safety and potential negative environmental impact of biological control agents (see Harris, 1990; Howarth, 1991; Simberloff & Stiling, 1996). However, in their real world environment, phytophagous insects employ a complex behavioural sequence to find and then determine the suitability of potential host plants. Yet an analysis of the mechanisms underlying the process of host plant finding and acceptance is not usually included in the design or interpretation of host specificity tests for weed biological control (Marohasy, 1998).
9.2.17 The project team notes as also observed by both the external reviewers (Drs Barratt and Breitwieser, see their reports in Appendix 2 and 3) that the applicant did not explain the criteria for selection of plants for host specificity testing. As noted by Dr Breitwieser it is assumed that testing was done on representatives of at least one genus of each tribe of Asteroideae that were available in South Africa. On this basis the project team agrees with the applicant that host range testing of the boneseed leafroller has demonstrated that, given the current host-range, attack on non-target species is very unlikely.
9.2.18 According to Marohasy (1996) many biologists perceive organisms as constantly evolving and therefore consider the host plant range of
Page 25 of 84 biological control agents as labile. Host plant ranges are thus likely to undergo adaptive change should environmental conditions change, for example following successful biological control. As a consequence, introduction of biological control agents against weeds is considered by many to be an inherently unsafe practice with non-target plants at risk. However, despite the introduction of over 600 insect species from one geographic region to another for biological control during the century, there are relatively few documented cases of changes in host plant range.
9.2.19 In terms of the future, evolutionary change in host range may occur, but it is likely to be gradual and happen in very small steps. The time frames for any evolutionary change would be very long term and would probably only occur under heavy selection pressure, and probably only if the weed became very rare. The nature of co-evolution between plants and their specialised herbivores means that any such changes affect relationships with very closely related species rather than more distantly related species. Thus any future change to the boneseed leafroller is more likely to affect its relationship with other species of the tribe Calenduleae than more distantly related species. There are no native members of this tribe in New Zealand, and few introduced species of economic importance in the tribe. Thus the possibility of the boneseed leafroller through the process of evolution, switching from feeding on boneseed to cause significant damage to non-target plants is considered to be improbable with a minimal to minor magnitude of effect.
9.2.20 The project team considers that uncertainty in this regard would remain as this is a complex phenomenon and therefore it is difficult to elaborate on this issue. According to Sheppard et al. (2003) non target effects from biological control agents are a concern internationally because:
(i) previous releases (<1%) are having significant negative impacts on native species;
(ii) introduced organisms are a recognized global threat to sustainable agriculture and biodiversity;
(iii) biological control is increasingly being used in complex natural ecosystems where indirect effects are harder to predict.
As a result, governments are adopting a more risk-averse attitude to biological control as they assess such releases from an environmental and economic standpoint.
9.2.21 However, Sheppard et al. (2003) also note that currently only the New Zealand approach closely matches a full ecological risk-benefit-cost analysis of biological control releases with a precautionary approach, open consultation, a broad hazard/benefit definition in the release application and a judicial basis to the decision, but it comes at a high cost.
Conclusion - Potential impacts on non-target plants
Page 26 of 84 9.2.22 The project team notes that experimental design for the host specificity testing in New Zealand and Australia was different and therefore the project team agrees with the external reviewer’s comment that we have no data from tests with New Zealand plants equivalent to those where non-target oviposition occurred in the Australian tests (see also comments by Dr Barratt in Appendix 2). In her comments Dr Barratt notes that further testing to determine physiological host range in containment in New Zealand would provide the basis for further field testing outside New Zealand. The project team while agreeing with the reviewer notes the inherent difficulties in maintaining the same line of plant populations to be tested in different jurisdictions and would appreciate comments on the applicability of this suggestion from the applicant.
9.2.23 The project team notes that the applicant has followed the internationally accepted host specificity testing protocol. Paynter et al. (2004) presented the results of surveys of non-target feeding by 14 weed bio-control agents that demonstrate the reliability of these techniques. The surveys showed that where good host-specificity testing was done prior to release non- target effects were not realised. In fact these techniques tended to produce false positive predictions of non-target effects, that is predicted feeding did not occur. However, as concerns have been raised by submitters and the external experts on the rigour of the testing regime the Authority may wish to discuss this issue at the hearing.
9.2.24 Irrespective of the above uncertainty, the project team notes that there are no native plants in the same tribe in New Zealand to which boneseed belongs. The project team concludes that given the likelihood and magnitude of the risk of impacts on non-target plants and the low degree of uncertainty surrounding the assessment described in sections 9.2.16- 9.2.21 this risk is considered to be not significant.
Identification – Potential displacement of native invertebrate species
9.2.25 According to the application, boneseed is not known to provide a valuable habitat for native invertebrate species in New Zealand. A gradual decline in boneseed populations due to biological control, and consequent recolonisation by other plants means native fauna currently associated with boneseed would be very unlikely to be disadvantaged. Competition with, or displacement of, existing natural enemies of boneseed in New Zealand is unlikely to cause any significant adverse effect. The invertebrate fauna associated with boneseed in New Zealand was surveyed in 1999–2000 (Winks & Fowler, 2000). The survey showed that boneseed is utilised by a wide range of native and exotic invertebrates in New Zealand but damage is usually not severe. Most of the damage observed on boneseed could be attributed to non-native invertebrates generally regarded as “pest” species (e.g. the garden weevil, Phlyctinus callosus (native to South Africa), and Fuller's rose weevil, Asynonychus cervinus (native to South America). These species have alternative hosts and are therefore unlikely to be disadvantaged by a decline in boneseed numbers.
Page 27 of 84 9.2.26 One submitter (6325) was of the opinion after reading the application that there was virtually no potential for displacement of native species.
9.2.27 One submitter (6329) was particularly concerned about this issue due to genetic mutation resulting in the development of a new species. The project team considers this issue in sections 9.2.17-9.2.18 of this report.
9.2.28 One submitter (6330) also considered impacts of food webs hadn’t been addressed. The project team considers that this issue is addressed throughout section 7 of this report.
9.2.29 One submitter (6331) noted concerns about increases in Argentine ant populations due to the presence of boneseed leafroller as a food source. The project team considers that this issue is addressed in section 7.1.21- 7.1.22 of this report.
Assessment – Potential displacement of native invertebrate species
9.2.30 The project team notes that in any predator prey guild, environmental selective pressures along with the interactive elements generally reach equilibrium where both predator and prey may exist albeit at low densities. It is also noted that if Argentine ants are already established in amongst the boneseed infestations then the ants will be established as a result of existing food sources. Introduction of the boneseed leafroller is likely to add to an additional food source for Argentine ants but any advantage that these ants may draw would depend upon the abundance of the host specific boneseed leafroller whose abundance is likely to decline as the vigour of the boneseed declines. As a result, no selective advantage for the ants is expected because they are not as closely associated with the boneseed leafroller as the boneseed leafroller is associated with the boneseed.
9.2.31 The project teams agrees with the assessment in the application and considers it improbable that any adverse effects on the native fauna will occur as a result of the release of the boneseed leafroller. The magnitude of any such effects are considered to be minimal to minor.
Conclusion – Potential displacement of invertebrate native species
9.2.32 The project team notes the weedy nature of boneseed and the species reported to occur on it. The project team concludes that given the low degree uncertainty surrounding the assessment of likelihood and magnitude described in 9.2.30-9.2.31, the risk is considered to be not significant.
Identification – Potential introduction of associated parasites
9.2.33 The application notes that any parasitoid, predator, or pathogen is most unlikely to pass undetected through the quarantine containment procedures required by the Ministry of Agriculture and Forestry (MAF) in accordance with an Import Health Standard issued under the
Page 28 of 84 Biosecurity Act 1993. The Import Health Standard will specify a period during which the species will be reared in quarantine containment.
9.2.34 Voucher specimens will be lodged in the New Zealand Arthropod Collection (NZAC) and taxonomic identification by a specialist lepidopteran taxonomist will be conducted while the insects are still in containment.
9.2.35 A sample of specimens will be tested for the presence of pathogens and provided results of the pathogen testing are negative, and the insects are confirmed to be the correct species, application will be made to MAF for permission to release the boneseed leafroller from quarantine containment.
9.2.36 The boneseed leafroller could conceivably become a vector for plant pathogens. However, since it is essentially host-specific, it is unlikely that it would transmit disease to non-target plants.
9.2.37 Three submitters (6323, 6325 and 6330) raised concerns about the need to ensure that the leafroller being released were free of associated organisms. One submitter felt that the potential adverse effects of accidental release of associated organisms had not been sufficiently addressed and felt that approval should not be granted until this assessment had been completed.
9.2.38 The project team notes that the quarantine procedures required by MAF would ensure that parasites or pathogens are not associated with the boneseed leafroller and that the possibility of quarantine measures failing is extremely and as a result the risk is not considered further.
Identification – Potential deterioration of natural habitats
9.2.39 The project team notes that introduced species with invasive potential remain the second most significant threat to biodiversity after habitat destruction, threatening significant percentages of listed rare endangered native plant species (Pimental, 2002).
9.2.40 One submitter (6325) noted that rather than causing complete eradication the leafroller would serve to limit growth and reproduction resulting in a gradual decline. However, another submitter (6330) was concerned about the establishment of other weeds following boneseed reduction.
9.2.41 The application notes that rapid and uniform destruction of boneseed infestations over a short period could lead to an increase in erosion on some erosion-prone slopes currently infested by boneseed. There is also the possibility of an increased rate of water runoff, and decline in water quality. However, boneseed is a woody perennial, and it is unlikely that successful biological control would lead to rapid, wholesale death of mature plants over a wide area. It is more likely that heavy foliage damage would lead to reduced growth, shoot dieback, and the death of a small proportion of mature plants annually, leading to a slow decline in
Page 29 of 84 plant density. Even if an agent killed an entire plant it would not immediately remove it, and the dead material would continue to hold the soil surface for a period. On most erosion-prone slopes, reduction in the density of boneseed as a result of biological control will result in complementary invasion by other plant species.
9.2.42 The project team agrees with the assessment in the application and conclude that reduction in the boneseed infestation is likely to be gradual and over time and this process is likely to allow the opportunity for other plant communities to establish. In such a scenario destruction of the current habitats occupied by boneseed is unlikely to occur. There is uncertainty whether these latter effects would be adverse or beneficial to the environment of New Zealand as a result this risk is not assessed further.
Identification – Potential loss of valuable sources of food for vertebrates
9.2.43 According to the application, decline of boneseed may lead to a loss of food for vertebrates. The project team is not aware of any native vertebrates that consume boneseed. Most vertebrates tend to be dietary generalists and, as such, do not often rely exclusively on a particular weed species as a food source (Stanley & Fowler, 2004; the applicant also refers to Stanley, M. C., Fowler, S. V. 2004. Conflicts of interest associated with the biological control of weeds. In: Proceedings of the XI International Symposium on Biological Control of Weeds (eds Cullen, J.M., Briese, D.T., Kriticos, D.J., Lonsdale, W.M., Morin, L. and Scott, J.K.) In press. CSIRO Entomology, Canberra, Australia.). The species benefiting from weed invasions are usually common generalists that would not be driven to extinction by reduction or eradication of the weed (Schiffman, 1997). Although particular species may be found to benefit from weed species, the overall biodiversity of systems usually decreases rapidly as a result of the weed invasion (Griffin et al., 1989; Braithewaite et al., 1989; Samways et al., 1996; Ekert & Bucher, 1999).
9.2.44 In most cases, weeds are not essential resources, and use of weeds by native vertebrate species is usually the result of availability or preference rather than dependence. Exotic weed species may provide sub-optimal food and habitat for native fauna relative to native plant species (Williams & Karl, 2002).
9.2.45 Weed exploitation by native species may also have a number of cascade effects on ecosystems: 1) increase the spread of weeds in natural areas through fruit consumption and seed dispersal (Williams & Karl, 1996); 2) native bird species may no longer fulfil their function as pollinators and dispersers of native plant species (Pooley, 1993; Scott-Shaw, 1999); 3) extra resources provided by weed species could result in unnaturally high population numbers of some vertebrate species, which could in turn result in detrimental economic (orchard pests) and ecological (predation, competition with native species) consequences (Major et al., 1996; Bomford & Sinclair, 2002).
Page 30 of 84 9.2.46 The benefits of removing boneseed, in terms of biodiversity gains and increases in native bird populations, outweigh any negligible benefits boneseed might have as a food source for birds. The project team considers that any decline in boneseed infestation as a result of the biological control agent is not expected to have any significant effect on animals that feed on it as many of these animals would have had access to other food sources prior to boneseed infestation in these habitats. The magnitude of any effect is considered to be extremely low and therefore this risk is not assessed further.
Identification – Potential for incorrect taxonomic identification
9.2.47 The project team notes that it is very unlikely that a species other than boneseed leafroller (Tortrix s.l. sp. “chrysanthemoides”) will be imported (see section 9.2.34). As a result this risk is not assessed further.
Beneficial effects (benefits)
Identification – Potential for enhancement of native biodiversity
9.2.48 The project team notes that in 2001 the Department of Conservation declared boneseed as an unwanted organism (see section 1.1.4). As supplied in the application (Appendix 4) DoC commented on the invasive nature of the weed. “Boneseed is considered to be an invasive weed and is controlled by the Department of Conservation in certain parts of New Zealand. About 15 Area Offices (out of 51) control boneseed as part of their weed control programmes. Some of the offices control boneseed across their entire Area as part of eradication or containment programmes while others control only priority sites. For example in the Nelson/ Marlborough Conservancy boneseed control is undertaken jointly by DOC, Tasman District Council and Marlborough District Council. Within Nelson City, boneseed is not controlled as the infestation is too large and inaccessible and control is too expensive. Outside this area, all known sites of boneseed are controlled either by councils or DOC. Boneseed is mainly regarded as a threat in low-growing coastal vegetation and in regenerating coastal forest. Although it is susceptible to a range of control techniques, boneseed often grows in inaccessible areas such as cliffs, making control difficult and expensive. Sites with boneseed are often weedy and control programmes often include more species than just boneseed. In addition, boneseed is included on the surveillance lists of 11 Area Offices, which means that boneseed is at low levels in that Area and staff are encouraged to look for sightings so it can be controlled before a big expensive problem develops”.
9.2.49 A number of submissions were made in support of approval of this application due to concerns they had about the presence of boneseeds in their local communities. The submitters included regional councils (6321 and 6322) and private individuals (6325 and 6326). While remaining neutral on the application one submitter 6332) did note the serious threat boneseed posed to indigenous species.
Page 31 of 84 9.2.50 The applicant notes that the long-term impact of boneseed on native biodiversity may be much more serious than that of a weed such as gorse because there is little native seedling recruitment under a pure stand of boneseed (McAlpine & Timmins, 2002). This is because a pure stand of boneseed restricts seedling growth and recruitment of other plant communities, particularly natives.
9.2.51 The project team notes that in the absence of fire mature gorse infestations are regarded as nurseries of native plants. It is noted that in the absence of boneseed it is likely that native plants may establish if there are stands of native plants close by.
9.2.52 One submitter (6325) in support of the application noted that they had observed boneseed to form dense thickets that displace native vegetation and shade native seedlings. Another submitter (6330) noted that for this beneficial effect to be realised a programme of planting native seedlings would need to be implemented.
9.2.53 The native biodiversity could be enhanced by the removal of boneseed but the extent of this can only be determined over time. Due to this uncertainty this beneficial effect is not assessed further.
Identification – Potential for reduced herbicide use
9.2.54 The applicant notes that reduced herbicide use would lead to reduced adverse impacts on non-target organisms (collateral damage) and reduced pollution of soil, air, and water. Also, as many boneseed infestations are on very steep slopes, current control methods could lead to injury or death if a weed controller slipped and fell. Details of the types of herbicides used, their advantages and their disadvantages are outlined in Appendix 6 of the application.
9.2.55 In contrast, submitter 6324 considers that increased herbicide use would be preferable to release of the biological control agent. The applicant responded to this by stating that there are several herbicides that can be used to successfully control boneseed. They detailed in the application in Appendix 6.
9.2.56 The project team notes that biological control is considered a preferred option for managing the weed as herbicide use is often impractical due to damage to native plants and/or inaccessible terrain. In certain areas, control may be possible by stock grazing on small plants but in other areas of steep terrain this would be impractical.
Assessment and Conclusion – Potential for reduced herbicide use
9.2.57 The project team considers that it is likely that the use of herbicides will decrease if the biological control agent is successful and the magnitude of this beneficial environmental effect would be moderate to major since the selective application of herbicides in these circumstances is difficult. Thus the benefit is considered to be significant. Uncertainty surrounding
Page 32 of 84 realisation of this beneficial effect is related to the efficacy of the biological control agent.
Identification – Potential for increased conservation values
9.2.58 The application contains a case study of the expected conservation benefits of the release of the biocontrol agent expressing the increase in conservation value in monetary terms for Banks Peninsula (see Appendix 3). This case study is based on a cost/benefit analysis of the introduction of biological control agents for boneseed by Harris consulting on behalf of Environment Canterbury.
9.2.59 The case study analyses two alternative futures (a) ‘with’ boneseed leafroller present, using a mixed control regime which includes a combination of the biocontrol agent and conventional control methods (biocontrol + conventional), and (b) ‘without’ the biocontrol agent based on conventional methods only (conventional).
9.2.60 The ‘conventional’ alternative is based on a ten year control programme with costs ranging between $50,000 and $85,000 per annum. The expected outcome from this scenario is prevention of further spread of boneseed, and a reduction in the current extent of 20%. Over this ten year period expenditure would be applied to areas of lowest density infestation (<1%).
9.2.61 Three scenarios are presented for the ‘conventional plus biocontrol’ alternative. The ‘base case’ assumes an 80% reduction in boneseed density over the ten year time frame, with low and high impact scenarios assuming reductions of 50% and 90% respectively.
9.2.62 The key assumptions in the case study and the project team’s evaluation of these assumptions are as follows - (a) By itself the biocontrol agent will not reduce the total area affected by boneseed, but will reduce its density. The project team considers that this assumption is valid. (b) The value for conservation gain is equal to the average benefit resulting from a reduction in boneseed. The project team considers that this assumption is valid. The applicant notes that as boneseed becomes less extensive the marginal benefit from its removal may change and the average benefit applied may overstate the conservation value. The project team agrees that this approach to valuing benefits is appropriate, and whilst acknowledging the potential for change over time, considers that the average value is appropriate for the ten year period of the case study. (c) Over a ten year timeframe the base case for introduction of the biocontrol agent in conjunction with conventional control will result in an 80% reduction in boneseed density (no impact in years one through five, and straightline density reduction over years six through ten). Low and high impact density reductions are 50% and 90%.
Page 33 of 84 The project team acknowledges that these assumptions are derived from verifiable Australian studies, but considers that 80% reduction appears high, and that the conservative (low impact) scenario is probably the most appropriate for the New Zealand circumstance.
Assessment – Potential for increased conservation values
9.2.63 Tables 8, 9, and 10 in Appendix 3 of the application compare the benefits and costs of the three scenarios for the introduction of the biocontrol agent, and the alternative future without the biocontrol agent. As there was a miscalculation in the Tables presented in the application, this information has been recalculated and is summarised below. All of the figures are NPVs calculated over ten years using a discount rate of eight percent. Benefits have been calculated using information from a recent DoC study (direct costs), and costs are Environment Canterbury planned costs for conventional control and costs of introduction of the biocontrol agent ($40,000 per year for 10 years).
Scenario/Alternative future Benefit Cost Net Benefit
Conventional only (ie no $400,000 $430,000 $-30,000 biocontrol) Biocontrol + Conventional $770,000 $680,000 $90,000 Low impact (50% reduction) Biocontrol + Conventional $1,020,000 $680,000 $340,000 Base Case (80% reduction) Biocontrol + Conventional $1,110,000 $680,000 $430,000 High Impact (90% reduction)
9.2.64 The project team considers that it is likely that the low impact scenario for the biocontrol + conventional alternative future will be realised. The magnitude for the Banks Peninsula case study is moderate. Therefore the beneficial effect is significant. The project team considers that there is some uncertainty associated with this effect, associated with the efficacy of the biocontrol agent.
9.2.65 The project team’s assessment has adopted a more conservative perspective than that adopted in the case study. If the magnitude of the benefit were assessed on the basis of the proposed base case, then it would be deemed to be major. However, the project team considers that the likelihood of the base case being realised is lower than that for the low impact scenario, with greater uncertainty and has therefore chosen to adopt this conservative approach.
Conclusion – Potential for increased conservation values
9.2.66 Based on the assessment of likelihood and magnitude described in section 9.2.63-9.2.65 and noting the low uncertainty which is related to the efficacy of the biocontrol agent, the project team considers the potential for increased conservation values to be significant.
Page 34 of 84 9.3 Human health
Adverse effects
Identification – Potential for allergic reaction
9.3.1 The applicant noted that moth scales sometimes cause allergies, but there is no evidence that scales from these organisms would be any more likely to cause allergies than other similar species already in New Zealand (page 37 of the application). The project team concurs with the assessment in the application that the release of the boneseed leafroller would not have any additional consequences than those already existing. Any marginal increase is not considered to be important therefore this risk is not assessed further.
Identification – Potential for attacks, bites, stings, or other human health effects
9.3.2 It is highly unlikely that the boneseed leaf roller could represent a threat to human or animal health and safety by attacking, biting or stinging because moths are not known to do so and it is also very unlikely that it would form swarms of nuisance value. There is no evidence to suggest that the organisms could become vectors of insect-borne pathogens of humans or animals. As a result this risk is not assessed further.
Beneficial effects (benefits)
Identification – Potential for reduced herbicide use
9.3.3 The applicant notes that there is a potential for a decrease in the use of herbicides leading to potential health benefits which should apply to both spray operators and the general public.
9.3.4 The economic case study provided with the application postulates continuation of current chemical and mechanical control alongside the biological control agent, therefore any reduction in use of herbicide would not be expected to take effect until 5 or 10 years beyond the release. No evidence of the scale of such benefits has been provided, and it would be very difficult to identify specific adverse health effects resulting from herbicide use specifically with respect to control of boneseed. This benefit is not assessed further.
Identification – Potential for a reduction in worker injuries
9.3.5 Boneseed often grows on steep banks. Submitter 6321 has noted that difficult access to boneseed infested sites limits current control. No evidence of worker injury has been provided by the applicant and therefore there are no identified potentially significant benefits from reduction in worker safety resulting from a reduced need for chemical and mechanical control.
Page 35 of 84 Identification – Potential for a reduction in allergic reactions
9.3.6 Boneseed pollen is not known to be a cause of human allergies, therefore no potentially significant beneficial effects are identified from the biological control of boneseed.
9.4 Māori culture
Introduction
9.4.1 The project team considered the potential cultural effects in accordance with the HSNO Methodology Order 1998: Information Used by the Authority Methodology clauses 9(b)(i)and 9(c)(iv) and Sections 6(d) and 8 of the HSNO Act 1996.
9.4.2 In addition, the project team used the assessment framework contained in the ERMA New Zealand user guide “Working with Māori under the HSNO Act 1996”, and the ERMA New Zealand Revised Protocol “Incorporating Māori perspectives in Part V Decision-making”, as guides in assessing the information contained in the application and submissions.
Consultation
9.4.3 Consultation with Māori occurs in order to give effect to section 6(d) of the HSNO Act where the applicant is required to take into account the relationship of Māori and their culture, traditions and other taonga.
9.4.4 The applicant used the guidelines contained in the ERMA New Zealand User Guide “Dealing with Māori under the HSNO Act 1996” to assess the application against issues of significance to Māori in relation to the Treaty of Waitangi/Te Tiriti o Waitangi, Environmental, Cultural and Health and Wellbeing outcomes.
9.4.5 The project team notes that the onus was placed on the applicant by ERMA New Zealand to consult nationally in accordance with the User Guide and ERMA New Zealand national consultation policy.
9.4.6 The applicant was provided with advice and guidance from the Māori Unit in order to appropriately satisfy the national consultation requirements. Consequently, the applicant distributed letters outlining the details of the proposal to release the boneseed leafroller to 64 iwi throughout New Zealand on 21 January 2004. The iwi were invited to respond by 15 February 2004 utilising feedback forms and freepost envelopes provided by the applicant. A further reminder by phone, fax or email was issued to those who did not respond by that date.
9.4.7 In total, twelve responses were received:
Ngāti Rangitihi, Ngāti Kahungunu ki Wairarapa, Te Rūnanga ō Ngāti Tahu Ngāti Whaoa Inc, Te Rūnanga ō Turanganui ā Kiwa, Ngāti Paoa
Page 36 of 84 Whanau Trust and Te Arawa Trust Board indicated that they had no issues, raised queries and/or asked to be kept informed;
The Muaupoko Tribal Authority and Te Rūnanga ō Ngāti Hauiti noted their inability to comment on national issues due to a lack of resource and capability;
Te Rūnanga ō Whaingaroa opposed the application seeking reassurance that a contingency was in place should unanticipated effects become apparent;
Te Rūnanga ō Ngāi Tahu noted their position on previous application NOR02001 opposing the release of C.japonicus due to the uncertainty relating to potential non-target species effects1.
Ngāti Te Ata Awaroa ki Manukau and Ngā Rauru Iwi Authority indicated possible issues with the proposal and requested more discussion and a meeting.
9.4.8 At the time of preparing this report, the two groups requesting further discussion and a meeting had not responded to several contacts made by the applicant to confirm arrangements to meet. However the applicant did provide further information to these two groups.
9.4.9 Although the level of feedback was low, ERMA New Zealand’s experience indicates that often responses are not forthcoming due to resource capability and capacity issues amongst iwi/Māori organisations. In addition iwi have expressed their dissatisfaction with this style of consultative approach, and Te Atiawa Manawhenua ki Te Tau Ihu Trust in their submission (6328) noted that in their opinion the consultation was insufficient and did not address issues raised.
9.4.10 The Raukawa Trust Board (submitter 6329) requested further consultation if they could not be provided with an opportunity to have the application independently evaluated at the applicants expense. In addition, in their submission (6333) the Ministry of Agriculture and Forestry (MAF) noted that consultation did not appear to have been clear or full enough and should have involved other relevant authorities including Māori managers of Regional Councils, Federation of Māori Authorities, Māori liaison personnel of the District Health Boards and/or regional directors from Te Puni Kokiri.
9.4.11 In response to these concerns the applicant noted that it is very difficult to consult with every individual who may or may not have a view on a particular application. In the case of this application, guidance on the appropriate type of consultation was sought from ERMA New Zealand. As a result, consultation was undertaken in accordance with the User Guide (Dealing with Māori under the HSNO Act 1996) and ERMA New
1 Te Rūnanga ō Ngāi Tahu submission on application NOR02001, 7 April 2004.
Page 37 of 84 Zealand’s national consultation policy. The ERMA database for Te Tau Ihu O Te Waka includes the Ngāti Rarua Iwi Trust, the Ngati Rarua Atiawa Iwi Trust and the Ngati Koata no Rangitoto ki te Tonga Trust. Details of the application where forwarded to all three. The applicant also felt that the concerns raised by iwi were all sufficiently covered in the application.
9.4.12 Having noted these concerns and the response of the applicant the project team considers that the applicant entered into a consultation process designed to identify and assess risks to Māori nationally in accordance with ERMA New Zealand policy and guidelines. Additional to these requirements, the applicant made further effort to personally follow up with those who did not respond.
Treaty of Waitangi
9.4.13 Section 8 of the HSNO Act requires that when considering applications the Authority shall take into account the principles of the Treaty of Waitangi.
9.4.14 The project team considers that applications that are unlikely to adversely affect native or valued flora and fauna species, and tikanga Māori (specifically the ability of Māori to maintain their traditional knowledge and role as kaitiaki) are likely to be consistent with the principles of the Treaty of Waitangi.
9.4.15 Of particular relevance to this application is the principle of active protection affirmed by the Court of Appeal in the Lands case (1987). It refers to the Crown’s obligation to take positive steps to ensure that Māori interests are protected, and the courts have primarily considered it in association with the interests guaranteed to Māori in Article II of the Treaty. Specifically the Court noted that “… the duty of the Crown is not merely passive but extends to active protection of Māori people in the use of their lands and waters to the fullest extent practicable”2.
9.4.16 The Waitangi Tribunal have also noted that “the Treaty of Waitangi obliges the Crown not only to recognise the Māori interests specified in the Treaty, but actively to protect them … It follows that the omission to provide that protection is as much a breach of the Treaty as an act that removes those rights”3.
9.4.17 A pending claim to the Waitangi Tribunal known as WAI262 the Flora and Fauna Claim, is of notable significance in decision making for the introduction of a new organism. The claim concerns the “protection, control, conservation, management, treatment, propagation, sale, dispersal, utilisation, and restriction on the use of and transmission of the knowledge of New Zealand indigenous flora and fauna and the genetic
2 Judgement of Cooke P in the Lands Court of Appeal Case (1987), p37. 3 Manukau Report, Waitangi Tribunal, p95.
Page 38 of 84 resources contained therein”4. It refers specifically to the protection of indigenous biodiversity, and notes that the Hazardous Substances and New Organisms Act is one example of Crown legislation that adversely affects Māori rights in this regard.
9.4.18 In their submission, Te Atiawa Manawhenua ki Te Tau Ihu Trust (6328) refers to the WAI262 Claim noting that the application does not provide a legal opinion. As this claim has not been given full consideration by the Waitangi Tribunal to date, its does not have a legal status per se. However, the intent of the claim is consistent with the principle of active protection (amongst others) and this particular principle is recognised in the Authority’s protocol Incorporating Māori Perspectives in Part V Decision-making. The project team notes that Authority decision-making does not substitute for the Waitangi Tribunal process, though it should take account of relevant issues raised by claimants in accordance with their responsibilities under section 8 of the HSNO Act.
9.4.19 In the applicants response to this issue, it was considered that a legal interpretation of the WAI262 Claim is an unnecessary prerequisite to this application because the Claim is not sufficiently advanced to merit such scrutiny. In addition, the claim was not identified as an issue during the consultation process.
9.4.20 Taking into account the principle of active protection, would require the applicant to have provided sufficient evidence to show that the application poses no risk of adverse effect to native/endemic and or other taonga species, ecosystems and traditional Māori values and practices.
9.4.21 The project team considers that any level of uncertainty regarding the potential for adverse effects on taonga species and traditional Māori values and practices, may be viewed as inconsistent with the principle of active protection.
Adverse effects
Introduction
9.4.22 The applicant noted that there were no significant risks or costs in reference to sections 6(d) and 8 of the HSNO Act arising from the application. However the project team considers that although no specifically cultural issues were raised during consultation, the proposed release of the boneseed leafroller poses potential direct and indirect adverse effects on taonga flora and fauna species, and to tikanga Māori.
9.4.23 Therefore the release has potentially significant impacts on the relationship of Māori and their culture and traditions with their ancestral lands, water, sites, wāhi tapu, valued flora and fauna and other taonga.
4 Statement of Claim, 9 October 1991, Wai 262 record of inquiry, doc#1.1, Waitangi Tribunal.
Page 39 of 84 Identification – Potential impacts on taonga species
9.4.24 The diversity of iwi perspective on the issue of biological control makes it difficult to ascertain a national Māori opinion and perspective on their use and potential impact on taonga Māori.
9.4.25 Taonga has been variously defined as things deemed to be of value to Māori. General acknowledgement is made that native flora and fauna, and specific introduced flora and fauna in New Zealand are considered taonga to Māori (taonga koiora and taonga tuku iho respectively). Taonga in terms of whakapapa (the shared genealogy of Māori with native flora and fauna species), mahinga kai (sources of traditional and/or valued food), and rongoa (sources of traditional medicines).
9.4.26 References to taonga of this nature are included in Iwi Settlement Acts (such as the Ngāi Tahu Claims Settlement Act 1998) and pending claims before the Waitangi Tribunal (such as WAI262 the Flora and Fauna Claim). Also in numerous government publications of organisations such as the Ministry for the Environment, Department of Conservation, Ministry for Agriculture and Forestry, and Parliamentary Commissioner for the Environment.
9.4.27 It is noted in the environmental effects section of this report and in responses and submissions received from iwi, there is a low level of uncertainty about the potential adverse effects to non-target species, namely native and valued flora and fauna, from the introduction of the boneseed leafroller. The environmental assessment will not be duplicated here, only to note that some uncertainty remains. Although the current host range testing suggests that effects on non-target species are very unlikely, the possibility that the boneseed leafroller may evolve within a new environmental context could result in the potential (however unlikely) for the future host range of the biological control agent to change.
9.4.28 Iwi have repeatedly requested assurance, without uncertainty, that the release of biological control agents poses no threat to taonga species and native ecosystems in the short, medium or long term. In reference to this application, Ngāti Raukawa noted in their submission (6329) that there is a lack of “conclusive or unconditional evidence” that the moth will not impact on non-target indigenous species. The submission by Te Atiawa Manawhenua ki Te Tau Ihu Trust (6328) considers that further host specificity testing be undertaken by the applicant on a range of native species selected by iwi. The applicant responded to these submissions by stating that the results of the field test (Winks et al. 2002 – Appendix 1 of the application) add to the evidence already available from field observations in South Africa, and host specificity testing conducted for Australia, that the boneseed leafroller is expected to feed only on boneseed in New Zealand.
9.4.29 Te Rūnanga ō Ngāi Tahu, in their submission on application NOR02001, noted concern about the potential for unanticipated adaptation or
Page 40 of 84 interbreeding to impact on mahinga kai and other taonga tuku iho over time. The project team notes that this issue is considered more fully in section 9.2.9.
9.4.30 In addition Te Rūnanga ō Te Rarawa in their submission on the application (6323) raise concern about the potential impact to taonga species of parasitoids, predators or pathogens associated with the boneseed leafroller. They also noted the need to further investigate the impact on native/endemic and existing introduced parasitoids and predators (such as the Argentine ants) that may become associated with the leafroller, and the subsequent effect to existing populations and food chains.
9.4.31 The project team notes that these issues would also raise potential adverse effects to mahinga kai, rongoa and the sustainability of taonga species and ecosystems. In this regard, the submission from MAF (6333) noted that consideration needed to be given to protection of the rongoa gene stock by the applicant.
Assessment and Conclusion – Potential impacts on taonga species
9.4.32 In regard to associated organisms, the project team has noted previously that the quarantine procedures required by MAF in accordance with an Import Health Standard issued under the Biosecurity Act 1993 adequately address the issue. In addition, the host specificity of the organism will mitigate any possibility of the leafroller becoming a vector for plant pathogens, or lead to a significant change in the existing food chains of native/endemic parasitoids or predators.
9.4.33 Overall, the project team considers that in relation to potential adverse effects to taonga species, there is a high level of reliance on the robustness of the host specificity testing. Given the nature and extent of the testing, and international experience, the likelihood of significant adverse effect to taonga species is considered to be improbable. However, if adverse effects were to occur in relation to taonga species and ecosystems either through effects to non-target species or through unanticipated effects from associated organisms, the magnitude of effect could range from moderate to major in terms of potential effect to both taonga koiora and taonga tuku iho. Exposure to this risk is involuntary, the risk may persist over time and may extend its effects beyond the immediate location. The potential adverse effects are irreversible.
Identification – Potential impacts on kaitiakitanga
9.4.34 The role of Māori as kaitiaki has been formally recognised, including in legislation such as the Resource Management Act 1991, as guardians and/or stewards of New Zealand’s natural resources. Kaitiakitanga is the undertaking of duties and obligations inherited from the atua (spiritual guardians) over the realms of those atua in accordance with tikanga
Page 41 of 84 Māori5. This application poses two potential sources of adverse effect to kaitiakitanga through unanticipated effects to the mauri of native and valued species and ecosystems (taonga koiora), and to the traditional values and practices of Māori (taonga tuku iho) in relation to those taonga.
9.4.35 Mauri is a key element of kaitiakitanga. It is a vital form of energy providing balance and enabling regeneration6 amongst the many children of Papatuanuku and Ranginui. Of relevance to this application, is the understanding that mauri is vulnerable to unnatural changes to the environment7. Within a Māori worldview it is the responsibility of Iwi/Māori, within their specific regions, to exercise kaitiakitanga to protect the mauri of significant resources to ensure their sustainability and availability for generations to come.
9.4.36 As environmental changes influence mauri, any assessment of the potential effect to the mauri of taonga flora and fauna species resulting from the release of a biocontrol agent is likely to be closely aligned with the assessment of environmental effects. Although no adverse effect to the mauri of taonga species and ecosystems was specifically referred to by the applicant or those iwi consulted, the project team considers the potential for adverse environmental effect outlined in section 9.2.1-9.2.47 to have potential subsequent effects on the mauri of those species and ecosystems. This assumption is consistent with the holistic nature of mātauranga Māori (Māori knowledge and knowledge systems), i.e. that all things within a Māori worldview consist of physical and metaphysical elements. If the physical well-being of taonga flora and fauna species is altered, so too is its metaphysical well-being or mauri.
9.4.37 As metaphysical or intangible effects are difficult to measure, any identification, assessment and evaluation of the likelihood or magnitude of effect to mauri from an iwi perspective, will be challenging to obtain. Mauri is a concept whose whole is more than the sum of its parts. Therefore, although the biophysical risks posed by the introduction of a new organism may be considered low, quite another perception of risk may exist for iwi as a function of the unique relationship they have with their taonga.
9.4.38 The responsibility of kaitiakitanga is a challenging one as iwi deal with managing the effects to mauri in a rapidly changing environment. Their ability to continue to adequately fulfil this responsibility and to maintain practices and traditions (taonga tuku iho), which are embodied within tikanga Māori, is placed at risk by decision making that does not adequately reflect their concerns or interests.
5 ERMA New Zealand (2003) Draft Protocol ‘Incorporating Māori Perspectives in Part V Decision-Making’, page 9. 6 Ibid, page 16. 7 Massey Universty Lecture Notes (1990) New Zealand’s Natural Heritage. Paper 21.103. Case Study No.3 Manukau Harbour Claim.
Page 42 of 84 9.4.39 Should adverse effect result to mauri or the ability of iwi to maintain rongoa stocks, collect food, or maintain waahi taonga, the costs associated with fulfilling the responsibility of kaitiakitanga will be borne by Māori. This will not be limited to financial cost.
Assessment and Conclusion – Potential impacts on kaitiakitanga
9.4.40 The project team considers that although difficult to evaluate because of the issues noted above, the magnitude of adverse effects on the ability of Māori to continue to fulfil their role as kaitiaki would be major should adverse effects eventuate. However the project team consider the risk of adverse effects to be improbable. Exposure to this risk is involuntary, the risk may persist over time and may extend its effects beyond the immediate location. The potential adverse effects are irreversible.
9.4.41 Further, the project team notes that the applicant asserts in section 4.1 (page 17) of the application that the locations, numbers, timing, methods of release and post release monitoring will be determined and undertaken collaboratively by staff from Landcare Research, regional councils, the Department of Conservation, and iwi throughout New Zealand. Te Rūnanga ō Te Rarawa in their submission (6326) have requested that a condition be placed on any proposed approval that reinforces the need to consult environmental stakeholders prior to any release of the boneseed leafroller in their region. The project team considers this request to be consistent with the requirements of kaitiakitanga (as codified within tikanga Māori) and endorses the importance of iwi/Māori involvement, as appropriate, at the implementation stage. However, it is noted that controls cannot be placed on full release approvals.
9.4.42 The applicant has responded to this issue by noting that Regional Councils are mandated by the Biosecurity Act 1993 to undertake pest control within the ambit of a pest management strategy. There are specific requirements under section 73(c) to undertake consultation with “the tangata whenua of the area who may be so affected, through iwi authorities and tribal runanga” during the preparation of a pest management strategy. Strategies must be publicly notified and submissions called for. It is considered that appropriate release protocols with iwi and runanga should be undertaken and they are best developed through the strategy formation process.
Beneficial effects (benefits)
Identification – potential impacts on taonga, eco-systems and traditional Maori values and practices
9.4.43 The applicant considers that Māori may experience beneficial cultural effects from the release and success of the boneseed leafroller biological control agent (table 6, page 40). These benefits are well described in previous sections of this report and include:
Page 43 of 84 Restoration and enhancement of indigenous and valued flora and fauna and ecosystems from the reduction in adverse effect caused by the boneseed plant; and Improvement to the purity of water, land, air and human health and well-being through a reduction in herbicide use. 9.4.44 Te Rūnanga ō Ngāi Tahu noted in their submission (6332) that the boneseed weed posed a serious threat to indigenous coastal species and the associated cultural values. This issue is also raised in section 9.2.48- 9.2.53 dealing with environmental benefits.
9.4.45 The project team also notes that beneficial cultural effects may include the enhancement of the mauri of taonga species and ecosystems through the decline in boneseed population and the reduction in use of herbicide. This enhancement would have subsequent beneficial effects on the ability of iwi/Māori to maintain and improve their role as kaitiaki limiting the need for intervention in the management of mauri. In addition, the potential enhancement of taonga species may provide more opportunity for iwi/Māori to maintain and improve their cultural practices and knowledge.
Assessment and Conclusion – potential impacts on taonga, eco-systems and traditional Maori values and practices
9.4.46 The project team agrees that if successful, the magnitude of beneficial effect to taonga species, ecosystems, and traditional Māori values and practices may be major. The likelihood is more difficult to ascertain due to the variable success of the boneseed leafroller overseas, and due to the advent of pre-existing competitive exotic weed species that could inhabit the ecosystem made available by the eradication of the boneseed plant, more quickly than native species. This uncertainty, although identified as low from a biological perspective, is considered likely to have unknown effects from a cultural and particularly metaphysical perspective. For example, there may be beneficial effects to the ability of iwi/Māori to appropriately fulfil their role as kaitiaki in regard to taonga, resulting from the release of the boneseed leafroller, as noted above. However, there will remain a level of responsibility to continue to monitor and manage the long term effects (both physical and metaphysical) of the introduction of the weed and the leafroller in terms of their overall relationship to taonga. The shape and nature of that ongoing responsibility remains unknown.
9.5 Society and community
Adverse effects
Identification – Potential damage to a valued home garden plant
9.5.1 Boneseed is not a garden plant or valued species. Potential adverse effects to non-target home garden plants are addressed in section 9.2.22- 9.2.24 of this report.
Page 44 of 84 Identification – Potential for loss of amenity values
9.5.2 Defoliation resulting from infestation may result in unsightly bushes in areas where boneseed is particularly prevalent. However, boneseed infestation is likely to be gradual over time, thus allowing other plant to communities to establish and replace boneseed. Any adverse amenity effect from defoliation is not considered to be potentially significant and is not assessed further.
9.5.3 At this stage, while it is not known what plant species would be likely to replace boneseed, it is possible that in some circumstances other weed species would invade. This is not deemed to be relevant to this application since it is not possible to generalise as to which weed species might invade, and it is likely that there would be a wide variation over the whole of New Zealand.
9.5.4 Given the current favoured location of boneseed, the introduction of the biocontrol agent might lead to short term increased risk of coastal erosion on clay cliffs. However, as stated in section 9.5.2 of this application the most likely result of introduction of the biocontrol agent is gradual die back and replacement by other species (page 28 of the application). Any adverse amenity effect from defoliation is not considered to be potentially significant and is not assessed further.
Identification – Potential for nuisance effects
9.5.5 Nuisance effects such as stinging and biting have been considered in Human Health effects (9.3.2). There is no evidence that scales from these organisms would be any more likely to cause allergies or nuisance than other similar species already in New Zealand (page 37), and the marginal increase is not considered to be potentially significant and is not assessed further.
Beneficial effects
Identification – Potential for reduction in herbicide use
9.5.6 Information provided by the applicant and submitters confirms the physical difficulties in attempting to control boneseed using herbicides and mechanical control. While one submitter (6324) indicated a preference for herbicide control, other submitters (including 6326) believed that herbicide control is not effective and has potential adverse social effects (see also 7.1.13).
9.5.7 Current control methods for boneseed involve applicators using herbicides on rugged coastal areas and cliffs (see 9.3.5). While there may be social benefits associated with the health and safety benefits from reduced herbicide use the project team does not consider these to be potentially significant and are not assessed further.
Page 45 of 84 Identification – Potential for reduction in fire risk
9.5.8 Boneseed infestations are highly flammable. The project team considers that removal of boneseed infestations from inaccessible coastal areas may reduce the associated potential fire risk and have beneficial effects if replaced by less flammable species. While the project team acknowledges this beneficial effect, it is not considered to be significant and therefore it is not assessed further.
9.6 The market economy
Adverse effects
Identification – Potential for loss of employment
9.6.1 One submitter (6324) indicated a preference for herbicide control as it had the added advantage of providing employment. The project team notes that one of the reasons why current control is not effective is the difficulty of access to areas infested. Submitter 6326 notes that boneseed infestations around Lyttelton Harbour and Port Levy are in areas where “spraying is impractical”. Thus, declining the application would not be expected to lead to an increase in employment. As submitter 6325 notes, “herbicide use is not an option to control the weed on cost, access and environmental grounds”.
9.6.2 Introduction of the biocontrol agent is expected to result in a reduction of the need for mechanical control of boneseed. This would in turn result in a reduced requirement for labour. Given that any such effect would be very gradual and uncertain (since control of boneseed is conducted alongside other weed control programmes) it is not considered to be potentially significant in terms of the economy and therefore is not assessed further.
Beneficial effects (benefits)
Identification – Potential for reduction in cost of control of boneseed
9.6.3 If the boneseed leafroller were to be effective in the medium to long term then there would be a reduction in cost of control of boneseed, including the cost of the herbicide, costs of application of herbicide, and cost of other control methods such as cutting.
9.6.4 The economic case study provided by the applicant for the Banks Peninsula region provides for conventional control methods to continue for the next ten years. The same level of funding would be applied over that period, but would be used to address areas of increasing density of infestation. The intention is therefore to use the biocontrol to supplement conventional control in the short to medium term.
Page 46 of 84 9.6.5 No information is available about actual control costs for other regions, and it is not known whether conventional control methods would be continued in other regions or not.
9.6.6 Two submitters (6322 and 6325) noted their support for use of boneseed leafroller as a cost effective means of controlling the weed. One submitter (6330) stated the analysis provided by the applicant was “inadmissible” as the assumptions used were “possibly wrong” and had not been “tested against empirical data”. The project team notes that any case study or modelling approach must make assumptions. The validity of the assumptions has been addressed in 9.2.62 and 9.2.63.
Assessment and conclusion – Potential for a reduction in the cost of control of boneseed
9.6.7 Given the information provided by the applicant about the approach expected to be adopted by ECAN for Banks Peninsula, the project team considers that in the longer term it is likely that there will be a reduction in the cost of control of boneseed, and given the current control costs that the magnitude of this effect is minor to moderate. Thus this effect is considered to be significant. Any uncertainty associated with this effect is associated with the efficacy of the biocontrol agent.
Identification – Potential for a reduction in costs related to worker injuries
9.6.8 Current control methods for boneseed involve applicators using herbicides and other control methods on rugged coastal areas and cliffs (see 9.3.5 and 9.5.6). The applicant has identified potential economic benefits associated with health and social benefits resulting from a reduction in boneseed and subsequent reduction in the need for worker activity in these areas. However, at the present time no information is available about the nature, severity or extent of injuries currently sustained by workers. The project team acknowledges that potential benefits may exist, but in the absence of information is unable to assess them further. However, given the expected gradual nature of decline of boneseed and Environment Canterbury's intention to continue with conventional control methods for at least a ten year period, any such benefit in the Canterbury region would be very small. Extrapolating to the whole country the project team does not consider that this benefit would be potentially significant and therefore it is not assessed further.
10 Overall Evaluation
10.1 Summary of information
10.1.1 The biological control agent (boneseed leafroller Tortrix s.l. sp. “chrysanthemoides”) has not yet been described but the description boneseed leafroller Tortrix s.l. sp. “chrysanthemoides” uniquely identifies the organism to be imported for release.
Page 47 of 84 10.1.2 Chrysanthemoides monilifera subsp. monilifera (boneseed) is a recognised weed (Roy et al., 1998) that has been declared an unwanted organism by the Department of Conservation in 2001. This bushy shrub that grows 2-3 metres has a number of features that make it a competitive weed. It can establish in a variety of soils, is frost-resistant and tolerant of semi-shade, drought, and salty environments and is a prolific producer of seeds. Boneseed is a weed of coastal areas where it can displace native vegetation and shade out native seedlings. This location also makes it difficult to control through traditional methods of herbicide application, hand-pulling, fire and or stock grazing.
10.1.3 The boneseed leafroller which helps to naturally control boneseed in its native habitat in South Africa is proposed for release as a biological control agent in New Zealand. Boneseed leafroller caterpillars web leaves together to form a shelter, and feed on leaf material from within this shelter. It is expected that the establishment of the boneseed leafroller in New Zealand will limit its vigour and make it less competitive in the New Zealand environment.
10.1.4 It is noted that depending upon the effectiveness of the boneseed leafroller Tortrix s.l. sp. “chrysanthemoides”, further biological control agents may in future be considered for release in New Zealand.
10.2 Summary of adverse and beneficial effects
10.2.1 A summary of all the major effects, the likelihood of the effect being realised, the magnitude of that effect should it occur, the uncertainty regarding that effect and its associated ranking with cross references and comments is provided in Table 4.
10.2.2 The potential for adverse or beneficial effects to be realised is based on an assumption of establishment of a self-sustaining population of the boneseed leafroller.
10.2.3 The project team has assessed all the potential adverse and beneficial effects to the sustainability of flora and fauna; public health; the economy; social, cultural and ethical issues; the intrinsic value of ecosystems; Maori culture and values; international obligations and the reasonable foreseeable needs of future generations.
10.2.4 The project team’s evaluation concluded that four adverse effects are not insignificant. In general the degree of uncertainty associated with adverse effects is low. The project team notes that the risk characteristics that pertain to the approach to risk have been discussed under the assessment of the individual adverse effects where the risks were considered to be significant. The project team notes that the Authority’s approach to risk must be documented in the decision.
10.2.5 The project team’s evaluation concluded that five beneficial effects are not insignificant. In general the degree of uncertainty associated with beneficial effects is low.
Page 48 of 84 10.3 Evaluation against statutory criteria
10.3.1 The project team considered the ability of the organism to cause any significant effects to the following matters as detailed in section 36 of the HSNO Act.
10.3.2 The project team considered that based on the information presented in sections 9.2.25-9.2.32 the organism is not likely to cause any significant displacement of any native species within its natural habitat (section 36(a) of the HSNO Act).
10.3.3 The project team considered that based on the information presented in sections 9.2.39-9.2.42 the organism is not likely to cause any significant deterioration of natural habitats (section 36(b) of the HSNO Act).
10.3.4 The project team considered that based on the information presented in section 9.3 the organism is not likely to cause any significant adverse effects on human health and safety (section (36(c) of the HSNO).
10.3.5 The project team considered that based on the information presented in sections 9.2.1-9.2.24 and 9.2.43-9.2.46 the organism is not likely to cause any significant adverse effect to New Zealand’s inherent genetic diversity (section 36(d) of the HSNO Act).
10.3.6 The project team considered that based on the information presented in sections 9.2.37-9.2.36 the organism is not likely to cause disease, be parasitic, or become a vector for human, animal, or plant disease (section 36(e) of the HSNO Act).
10.3.7 In conclusion the project team is of the opinion that the boneseed leafroller meets the minimum standards outlined in section 36 of the HSNO Act.
Page 49 of 84 Table 2: Summary of overall evaluation
Adverse effects (risks and costs)
Description Likelihood Magnitude Level of risk/cost Uncertainty Comments Section The environment Potential impacts on non- Improbable Minimal - minor A-B Low There is uncertainty surrounding this adverse 9.2.24 target plants effect due to the limitations of host-specificity testing. Displacement of native Improbable Minimal - minor A-B Low 9.2.32 invertebrate species Māori culture and values Potential impacts on Improbable Moderate - major C-D - Dependent on the robustness of the host 9.4.33 taonga species specificity testing. The use of risk terminology here is provided as a guide only, particularly given determination of uncertainty is difficult for intangible effects. Potential impacts on Improbable Major D - The use of risk terminology here is provided 9.4.40 kaitiakitanga as a guide only, particularly given determination of uncertainty is difficult for intangible effects.
Page 50 of 84 Beneficial effects (benefits) Description Likelihood Magnitude Level of risk/cost Uncertainty Comments Section The environment Reduced herbicide use Likely Moderate - major F - Realisation of benefit dependant on efficacy 9.2.57 of biological control agent Increased conservation Likely Moderate E Low 9.2.66 values Maori culture and values - Major - - It is difficult to assess beneficial cultural 9.4.46 effects using this type of risk assessment model. The use of risk terminology here is provided as a guide only. Society and community Reduction in herbicide Likely Minor E Low 9.5.6 use The market economy Reduction in cost of Likely Minor – E Low 9.6.7 control of boneseed moderate
Page 51 of 84 10.4 Previous consideration of similar applications
10.4.1 The Authority has to date approved two other applications for release of new organisms under section 34 of the HSNO Act which are relevant to this application. The project team notes that while the biology of the organisms are very different to those of the current application, the basic premise is the same in that they were applications for full release of insect biological control agents to control weedy pests.
10.4.2 Application NOR99004 was submitted by Auckland Regional Council to release from containment the biological control agent Procecidochares alani (gall fly) for the control of Ageratina riparia (mist flower). This biological control agent had previously been released in two countries where no adverse effects had been noted, although they hadn’t been specifically monitored for. This information in combination with the fact that there are no indigenous members of the target plant’s tribe resulted in a conclusion that the biological control agent would be host-specific. Some concerns were raised about the impact of parasitism on efficacy, however, the Committee was of the opinion that P. alani would still exert some degree of control over mistflower. Concerns were also raised about the presence of P. alani resulting in a population explosion of parasites that could impact on native tephritids, however the Committee considered that any population changes would be gradual and unlikely to be a significant risk. The importance of correct identification of the agent and good quarantine procedures were noted. It was noted that the key risk of approving this application was non-target effects and concerns by Maori to that effect. However, the Committee felt that the beneficial effects of reduced impact of mist flower on native flora would outweigh this risk.
10.4.3 Application NOR00001 was submitted by the Hieracium Control Trust to import for release three insects species (Macrolabis pilosellae, Cheilosia urbana and C. psilophthalma) for the biological control of weedy Hieracium species. Again host specificity testing and field observations in the country of origin suggested the biological control agents would not have non-target effects. Concerns about ecotype variability were overcome by importing organisms from the same localities as those used in testing. The importance of correct identification of the agent and good quarantine procedures were noted. Maori also raised concerns about biological controls in general. However, the Committee felt that the beneficial effects of reduced impact of Hieracium on native flora would outweigh this risk.
Page 52 of 84 11 References Bomford, M.; Sinclair, R. 2002: Australian research on bird pests: impact, management and future directions. Emu 102, 29-45, 2002.
Braithewaite, R.W., Lonsdale, W.M., Estbergs, J.A. 1989: Alien vegetation and native biota in tropical Australia: the impact of Mimosa pigra. Biological Conservation 48, 189-210.
Briese, D. 1999: Open field host-specificity tests: is “natural” good enough for risk assessment? In: Withers, T.M.; Brown, B.; Stanley J. ed. Host specificity testing in Australasia: towards improved assays for biological controls. CRC for Tropical Pest Management, University of Queensland. Brisbane, Australia.
Brown, W.L. 1958: A review of the ants of New Zealand (Hymenoptera). Acta Hymenopterologica 1, 1050.
Clement, S.L.; Cristofaro, M. 1995: Open-field tests in host-specificity deternination of insects for biological control of weeds. Biocontrol Science and Technology 5, 395–406.
Crowe, A. 2002: Which New Zealand insect. Penguin Books, Auckland, New Zealand. 127 pp.
Cullen, J.M. 1990: Current problems in host-specificity screening. In: Delfosse, E.S. ed. Proceedings of the VII International Symposium on Biological Control of Weeds, 6–11 March 1988, Rome, Italy. Ist. Sper. Patol. Veg. (MAF), pp. 27–36.
Don, A.W. 1974: Insects: ants and termites. New Zealand’s Natural Heritage 3, 909-913.
Dugdale, J.S. 1988: Lepidoptera- annotated catalogue, and keys to family- group taxa. Fauna of New Zealand 14. Science Information Publishing Centre, DSIR, Wellington.
Dunn, P.H. 1978: Shortcomings in the classic tests of candidate insects for the biocontrol of weeds. In: Freeman, T.E. (ed) Proceedings of the IV International Symposium on Biological Control of Weeds. Gainesville, Florida, 1976. Gainsville; University of Florida, pp. 51-56.
Edwards, P.B. 1993: Research in South Africa on biological control agents for Chrysanthemoides monilifera. Proceedings of a National Workshop on Chrysanthemoides monilifera. Port Macquarie, Australia. Pp. 16–19.
Edwards, P.B.; Adair, R.J. 1999: Application for the release from quarantine of “Tortrix” sp., a potential biological control agent for Chrysanthemoides monilifera. Unpublished report: Keith Turnbull Research Institute, Victoria, Australia; and CSIRO, Canberra, Australia. 40 pp.
Page 53 of 84 Ekert, P.A., Bucher, D.J. 1999: Winter use of large-leafed privet Ligustrum lucidum (Family: Oleaceae) by birds in suburban Lismore, New South Wales. Proceedings of the Linnean Society of New South Wales 121, 29-38.
Forno, W.; Heard, T. 1997: Compiling a plant list for testing the host range of insects. In: Biological control of weeds: theory and practical application. Australian Centre for International Agricultural Research Monograph Series 49, 71–75.
Harris, P. 1990: Environmental impact of introduced biological control agents. In: Mackauer, M.; Ehler, L.E.; Roland, J. (eds) Critical issues in biological control. Andover, UK; Intercept, pp. 289-300.
Howarth, F.G. 1991: Environmental impacts of classical biological control. Annual Review of Entomology 36, 485-509.
Ireson, J.E.; Davies, J.T.; Chatterton, W.S.; Holloway, R.J. 2002: Attempts to establish biological control agents for boneseed in Tasmania. Proceedings of the Thirteenth Australian Weeds Conference. Perth, Australia. Pp. 407–411.
Julien M.H.; Griffiths, M. 1998: Biological control of weeds – a world catalogue of agents and their target weeds. 4th edition, CABI Publishing, Wallingford, UK. 223 pp.
McAlpine, K.; Timmins, S. M. 2002: The effect of fire on bone-seed and gorse germination. Department of Conservation Science poster no. 56 (edition 1 September 2002).
McFadyen, R.C. 1998: Biological control of weeds. Annual Review of Entomology 43, 369–393.
Major, R.E.; Gowing, G.; Kendal, C.E. 1996: Nest predation in Australian urban environments and the role of the pied currawong, Strepera graculina. Australian Journal of Ecology 21, 399-409.
Moeed, A.; Meads, M.J. 1985: Seasonality of pitfall trapped invertebrates in three types of native forest, Orongorongo Valley, New Zealand. New Zealand Journal of Zoology 12, 17-53.
Moeed, A.; Meads, M.J. 1986: Seasonality of litter- inhabiting invertebrates in two native forest communities of Orongorongo Valley, New Zealand. New Zealand Journal of Zoology 13, 45-63.
Marohasy, J. 1996: Host shifts in biological weed control: real problems, semantic difficulties or poor science? International Journal of Pest Management 42, 71-75.
Marohasy, J. 1998: The design and interpretation of host specificity tests for weed biological control with particular reference to insect behaviour. Biocontrol New and Information 19, 13–20.
Page 54 of 84 Paynter, Q.E., Fowler, S.V., Gourlay, A.H., Haines, M.L., Harman, H.M., Hona, S.R., Peterson, P.G., Smith, L.A., Wilson-Davey, J.R.A., Winks, C.J., Withers, T.M. 2004: Safety in New Zealand weed biocontrol: A nationwide survey for impacts on non-target plants. New Zealand Plant Protection 27, 102-107.
Pimental, D. (ed) 2002: Biological invasions: economic and environmental costs of alien plant, animal, and microbe species. Boca Raton, FL, USA; CRC Press, 384 pp.
Pooley, E. 1993: The Complete Field Guide to Trees of Natal, Zululand and Transkei. Natal Flora Publications Trust, Durban.
Roy, B., Popay, I., Champion, P., James, T., Rahman, A. 1998: An Illustrated Guide to the Common weeds of New Zealand. New Zealand Plant Protection Society. 282 pp.
Samways, M.J., Caldwell, P.M., Osborn, R. 1996: Ground-living invertebrate assemblages in native, planted and invasive vegetation in South Africa. Agriculture. Ecosystems and Environment 59, 19-32.
Schiffman, P.M. 1997: Animal-mediated dispersal and disturbance: driving forces behind alien plant naturalization. In: “Assessment and Management of Plant Invasions” (Eds. Luken, J.O. and Thieret, J.W.). Springer-Verlag. New York, USA. pp 87-94.
Scott, J. K. 1996: Population ecology of Chrysanthemoides monilifera in South Africa: implications for its control in Australia. Journal of Applied Ecology 33, 496–508.
Scott, J.K.; Adair, R.J. 1990: The commencement of biological control of bitou bush and boneseed (Chrysanthemoides monilifera). Proceedings of the 9th Australian Weeds Conference, Adelaide, South Australia. pp. 126–129.
Scott, J. K.; Brown, E. M. 1992: Preliminary observations on the biology and host plants of “Tortrix” sp. (Lepidoptera: Tortricidae), a possible biological control agent for Chrysanthemoides spp. (Asteraceae). Journal of the Entomological Society of South Africa 55: 245–253.
Scott-Shaw.C.R. 1999: Rare and Threatened Plants of KwaZulu-Natal and Neighbouring Regions. KwaZulu-Natal Nature Conservation Service, Pietermaritzburg.
Sheppard, A.W.; Hill, R.; DeClerck-Floate, R.A.; McClay, A.; Olckers, T.; Quimby Jr., P.C.; Zimmermann, H.G. 2003: Biocontrol News and Information 24, 91-108.
Simberloff, D.; Stiling, P. 1996: How risky is biological control? Ecology 77, 1965-1974.
Page 55 of 84 Stanley, M. C., Fowler, S. V. 2004: Conflicts of interest associated with the biological control of weeds. In: Proceedings of the XI International Symposium on Biological Control of Weeds (eds Cullen, J.M., Briese, D.T., Kriticos, D.J., Lonsdale, W.M., Morin, L. and Scott, J.K.) In press. CSIRO Entomology, Canberra, Australia.
Syrett, P. 1999: Prospects for biological control of bone-seed, Chrysanthemoides monilifera spp. monilifera (Asteraceae: Calenduleae). Unpublished report, Landcare Research, Lincoln, New Zealand. 18 p.
Syrett, P.; Briese, D. T.; Hoffman, J. H. 2000: Success in biological control of terrestrial weeds by arthropods. In: Gurr, G.; Wratten, S. (eds), Biological control: measures of success. Kluwer, Dordrecht, The Netherlands. pp. 189– 230.
Wapshere, A.J. 1974: A strategy for evaluating the safety of organisms for biological weed control. Annals of Applied Biology 77, 201–211.
Wapshere, A.J. 1975: A protocol for programmes for biological control of weeds. PANS 21, 295–303.
Wapshere, A.J. 1989: A testing sequence for reducing the rejection of potential biological control agents for weeds. Annals of Applied Biology 114, 515–526.
Webb, C. J.; Sykes, W. R.; Garnock-Jones, P. J. 1988: Flora of New Zealand, Volume IV, Naturalised Pteridophytes, Gymnosperms, Dicotelydons. Christchurch, Department of Scientific and Industrial Research. 1365 pp.
Weiss P.W. 1986: The biology of Australian weeds 14 Chrysanthemoides monilifera (L.) T. Norl. Journal of the Australian Institute of Agricultural Science 52, 127–134.
Williams, P.A., and Karl, B.J. 1996: Fleshy fruits of indigenous and adventive plants in the diet of birds in forest remnants, Nelson, New Zealand. New Zealand Journal of Ecology 20, 127-145.
Williams, P.A. and Karl, B.J. 2002: Birds and small mammals in kanuka (Kunzea ericoides ) and gorse (Ulex europaeus ) scrub and the resulting seed rain and seedling dynamics. New Zealand Journal of Ecology 26, 31-41.
Winks, C. J.; Fowler, S. V. 2000: The invertebrate fauna of bone-seed, Chrysanthemoides monilifera spp. monilifera, in New Zealand. Unpublished report, Landcare Research, Auckland, New Zealand. 24 pp.
Winks C. J.; Syrett P.; Muller P.A. 2002: Testing the host specificity of the Chrysanthemoides Leafroller (“Tortrix” sp.) for New Zealand. Unpublished report, Landcare Research, Auckland, New Zealand. 17 p. (Note: this report is included in this application as Appendix 1).
Page 56 of 84 Zwölfer, H.; Harris, P. 1971: Host specificity determination of insects for biological control of weeds. Annual Review of Entomology 16: 159–178.
Page 57 of 84 Appendix 1: Summary of submissions received Submission #6321 by Greater Wellington Regional Council The submitter noted that there were “significant infestations of boneseed” in coastal areas of the Wellington Region, control of which was impossible due to issues of access. The submitter supported the introduction of boneseed leafroller as it was viewed as the only control option.
Submission #6322 by Waikato Regional Council The submitter noted that there were “significant infestations of boneseed” at Thames and Raglan within the Waikato Region and concerns about spread into other areas were raised. The submitter supported the release of boneseed leafroller and they considered that it would “constitute an effective and cost-efficient mechanism for controlling boneseed”.
Submission #6323 by Te Runanga O Te Rarawa The submitter noted concerns that regional authorities were under no obligations to consult with environmental stakeholders prior to release of the leafroller. The submitter also noted that they were “cautious about the introduction of any new organisms” and had concerns about the high mobility of leafroller. The submitter also felt that the issue of impacts of associated organisms on the leafroller’s efficacy and native fauna were “not well-covered”. The submitter requested that the Authority obtain this information prior to making it’s decision and that should approval be granted a condition be established that requires anyone releasing the boneseed leafroller to “consult with their environmental stakeholders” before doing so.
Submission #6324 by Wairewa Rūnanga The submitter very clearly stated their opposition to any further importation and release of any new organisms due to the concerns about non-target feeding on native flora. These concerns were borne out of the extrapolation of field data to New Zealand conditions. A preference for the use of herbicides was noted.
Submission #6325 by RR Scott The submitter noted that they had observed the increasing spread of boneseed in the Canterbury region in recent years, particularly in the Sumner area and the resulting shading effects on native seedlings. The submitter noted that herbicide was not a practical option due to costs, issues of access and adverse environmental effects. The submitter, a qualified ecologist, stated that it was his opinion that rather than complete eradication of boneseed the leaf roller would be a self-sustaining method of limited growth and reproduction. The submitter considered that the lack of a formal binomial was not of consequence as the data demonstrated that the leaf-roller is highly host- specific with “virtually no potential to displace endemic New Zealand species”. The submitter supported the approval of this application provided that quarantine regulations that would avoid the accidental introduction of associated organisms were adhered to and requested that the Authority require post-release monitoring.
Submission #6326 by RM Coop The submitter noted that they had spent considerable time removing young boneseed plants from their farm in Port Levy which was adjacent to an existing infestation. The submitter noted that the use of herbicides in existing infestations in their community is
Page 58 of 84 impracticable due to location including steep country and housing areas. The submitter supported the importation and release of the leafroller.
Submission #6327 by Bernadette Papa The submitter requested that further testing been done in New Zealand due to concerns about the extrapolation of overseas data in determining potential adverse effects of the leafroller on native flora and fauna.
Submission #6328 by Te Atiawa Manawhenua Ki Te Tau Ihu Trust The submitter opposed the introduction of the boneseed leafroller until further host- specificity as determined by iwi is completed. The submitter raised concerns about the consultation process as they felt they hadn’t been consulted and the concerns of those iwi that had, had not been addressed. The submitter also noted that the issue of the wai 262 claim had not explored. The submitter was of the opinion that the boneseed leafroller had not been effective in Australia and so should not be released in New Zealand unless as a vector (of what was not specified).
Submission #6329 by Raukawa Trust Board The submitter was of the opinion that there was not “conclusive, unconditional evidence” that there would be non-target effects on both plants (indigenous or exotics) or indigenous moths (due to competition for resources). Concerns were also raised about the ability for natural mutation to result in the development of a new species with unknown risks to the environment. The submitter requested that they could have the application independently assessed by a person of their choosing at the applicants cost and should this not occur for the applicant to pay for mediated consultation with Raukawa.
Submission #6330 by Cliff Mason The submitter was of the opinion that there was a lack on information about a number of issues (listed in brief below) which meant that the application should be declined. Information regarding boneseed: Lifespan of individuals and stands and does succession occur Form in South Africa and site of colonisation (eg disturbed soil), followed by succession or stability Seed production (numbers and viability and spread) The recent change to classification as a pest – is this reversible Evidence of continued spread and resulting implications (particularly for natives) Quantative ecological data on the impact of current control measures. Information on the leafroller Issues relating to efficacy of outbreaks (frequency, associated conditions, impacts on boneseed, geographical spread and Australian results) Introduced or considered for introduction elsewhere? Success (efficacy and host-specificity) of other leafrollers Non-target impacts on members of the Myrtaceae family and the Leptinella, Olearia, Senecio and Linum genera. Information on effects on the broader environment
Page 59 of 84 Species likely to replace boneseed and resulting effects on food webs etc. Information on the overall Programme Resources for multiple releases and associated agents Monitoring unspecified Planting required to realise ecological benefits Information on other matters Impact on future biological control agents (eg for Tortricid pests) Concerns about the accuracy of the Cost-Benefit analysis by Harris consulting Would the applicant have undertaken the project had a lot of the research already not been completed for the Australian release? Evidence for the long-term efficacy of biological control in general
Submission #6331 by the Department of Conservation The submitter identified a number of significant concerns with the application as detailed under the following headings and on the basis of these concerns sought a decision to decline the application unless further host-testing was completed. Host testing Noted deficiencies in the quantity and quality of host testing on New Zealand’s indigenous flora in particular that only 11 were tested in total and only two were endemic (29 further species for testing were recommended) Testing was not done in New Zealand – raising concerns about regional/continental influences on foliage quality The Australian work was “more extensive and was of a higher scientific standard than the Landcare” work Caged no choice test showed complete development on a range of non- host species suggested boneseed leafroller is not specific, though this was the case in field trials, though a no-choice field trial was not done Lack of no choice raised concerns about non-target effects when the boneseed leafroller was under “pressure” (ie in the absence of the host) which resulting selection pressures for non-target favouring boneseed leafrollers. Recommended use of Wapshere in combination with Withers choice and non-choice field tests Environmental monitoring programme Strongly advocated the development of a post-release monitoring programme to assess impacts on non-target plant species and other insects with a greater detail than currently supplied Suggested an initial release into a small number of isolated areas so monitoring could be done with a view to recalling the organisms should adverse effects be realised Noted that unless monitoring was a condition of release there was no legal requirement for it to be done and that resources had not been confirmed Success of the release Overall assessment of likelihood of establishment not done
Page 60 of 84 Concerns about effects of parasitoids (particularly indigenous insects), predators (Argentine ants in particular) and drought conditions Based on Australian results concluded establishment would be poor and should be taken into account in the cost/benefit analysis
Submission #6332 by Te Runanaga o Ngai Tahu This submitter acknowledged that “boneseed constitutes a serious threat to indigenous coastal species, and the cultural values associated with those species and their ecosystems”. However due to concerns about non-target effects on native species they decided to remain neutral on their decision to support or oppose the application until further information was available.
Submission #6333 by Ministry of Agriculture and Forestry This submitter noted several concerns with the application and stated that if the application was approved they recommended a second stage of host-specificity testing using a wider range of native plants (a list was provided) and including a non-choice field test to investigate issues of selection pressure. The submitter also noted concerns about predation of boneseed leafroller by Argentine ants in relation to the efficacy of the biological control agent and potential increases in the Argentine ant population due to the presence of a new food source. The submitter noted that consideration needed to be given to protection of the rongoä gene stock by the applicant. The submitter also noted that consultation did not appear to have been clear or full enough and may missed some relevant authorities including Maori managers of Regional Councils, Federation of Maori Authorities, Mari liaison personnel of the District Health Boards and/or regional directs of Te Puni Kokiri. The submitter also queried if approval was required under the Resource Management Act.
Miscellaneous It should be noted that the project team has attempted to address all the issues raised by submitters in the general body of this report. However, while a number of points made by the submitters raised valid issues, the project team did not consider them material to the consideration of this application as so they are noted below and have not been addressed in the main body of this report. If research hadn’t already been completed for the Australian release would the applicant have initiated their own programme based on the size of the problem (6330). Are sufficient resources available to fund multiple releases (6330)? What is the impact of the leafroller between outbreaks and how often and the extent to which they occurred and what conditions were required for them to occur (6330). General queries about biological control, including the release of other leafroller biological control agents (in relation to their efficacy and adverse effects) and the long-term sustainable effectiveness of this approach (6330). What is the potential impact of this introduction for future releases of biological control agents, in particular for the control of Tortricid pests (6330)? Submitter (6330) noted that in their opinion single releases were often ineffective and questioned if resources were available for further releases against boneseed.
Page 61 of 84 What resources are available for replanting natives should boneseed be eradicated (6330)? Submitter 6333 queried if RMA approval was required. Submitter 6330 questioned the reason for this recent elevation to being declared an unwanted organism and whether it was a temporary change in population dynamics. The same submitter also questioned what quantitative ecological data this decision was made on.
Page 62 of 84 Appendix 2: Dr Barbara Barratt’s Report
Application for approval to import for release a moth, the boneseed leafroller (Tortrix s.l. sp. “chrysanthemoides”), that is not currently present in New Zealand for the purpose of biological control of the weed pest boneseed Chrysanthemoides monilifera subsp. monilifera (L.) Norlindh (Family: Asteraceae)
Barbara I.P. Barratt, AgResearch Invermay, PO Box 50034, Mosgiel
1) Please provide a brief review of adequacy of the testing methodology used including issues relating to number of plants tested, replicates, choice/non-choice methods and interpretation of results. We would appreciate your comments on the ecotype/strain variability, possibility parasitoids associated with the moth, is there anything that could have been done to strengthen the host specificity testing, ie is there potential for non-target effects, and the potential for the biological control agent to establish in New Zealand.
Testing methodology The approach used by Australian researchers to assess the likely host specificity of boneseed leafroller (BSLR) pre-release in Australia included:
field surveys to determine the ‘natural’ host range of the herbivore in part of its natural range; also included Australian natives; laboratory no-choice experiments in Australia with test plants selected using the centrifugal phylogentic testing protocol; field oviposition choice experiments using as far as possible plants which tested positive in the laboratory tests; field larval feeding (no-choice) experiments using 21 spp. of Asteraceae and 14 plant spp. from other families; release of over 1000 BSLR pupae in the field where Banksia was present, an Australian native which supported full larval development in the laboratory.
The field surveys were thorough including 18 sites where 127 spp. of plants were examined. There were no records of BSLR on any plants other than Chrysanthemoides. Plants physically in contact with Chrysanthemoides were avoided, but it would have been useful to include these in a separate category. In natural conditions the target plant will be in contact with non-target plants and the frequency of (or lack of evidence for) larvae moving to another species would have provided useful data.
The laboratory no-choice tests involved 10 neonate larvae being placed on (mostly) 6 replicate plants of 96 species in 33 families. By most standards the number of test plants would be considered far more than adequate, although it was not explained whether several tiers of tests were conducted in view of the quite wide range of plants found to support full larval development. BSLR larvae were able to complete development on 36 species in 11 plant families indicating that the species has a potentially wide physiological host range. Replication of larval numbers and plants was adequate in most cases.
Page 63 of 84
Four field host range tests basically tested ovipostion preferences of BSLR females in the presence of Chrysanthemoides, with different BSLR release methods. Large numbers of insects were used with good rates of adult emergence from pupae. The experimental design, replication etc. appeared to be robust. The experiments showed that there were instances of oviposition on six non-target plant species even in the presence of Chrysanthemoides, but these were all within about a metre of the release point.
A futher field experiment was carried at two sites out to determine whether larvae hatching from egg batches placed on non-target plant species could feed and develop. Replication at these sites was adequate. In these tests, some larval development occurred on 14 non-target species, but after 40 days live larvae remained (from 2-48 individuals) on five non-target species, all in the Asteraceae.
Release of BSLR in the field at the Banksia site showed that the species did not establish. Banksia was one of the plants used in the laboratory no-choice tests where full larval development occurred (resulting in 20 pupae from 110 neonate larvae), and in the field oviposition tests, there was some oviposition on Banksia (Fig. 14b) although this result does not appear in Table 5.
The Australian report concludes that “Tortrix sp. is regarded as monophagous”. Given the wide range of plants species on which larvae can develop successully through to pupation, and the demonstrated incidence of oviposition and larval development on non-target plant species in field experiments, it is hard to concur with this conclusion. The field test with Banksia, and the results of the field survey of ‘natural’ host range support the conclusion that in the field in Australia, non-target species are not likely to sustain severe defoliation by BSLR. Further oviposition and larval feeding tests on offspring from individuals that developed successfully on non-target species would have been valuable to determine whether there was any change in the proportion of the subsequent generation that select alternative plant species for oviposition.
The approach used by NZ researchers to assess the likely host specificity of BSLR pre-release in NZ included:
a field oviposition choice test was carried out at one site using Chrysanthemoides and 10 non-target species (all Asteraceaea), three of which were native to NZ, in a design similar to the Australian oviposition field tests.
This test used 6 replicates of most plants, and 3293 BSLR pupae resulted in 1445 adults which produced 333 egg batches all on the target plant species. This test was adequately replicated and included suitable numbers of insects, but limited by the number of test plants used, especially NZ natives. The rationale for selection of the test plants was not given except that they were “important to New Zealand”. There was no evidence that the centrifugal phylogenetic testing protocol was used. The experimental design varied from that used in most of the Australian field tests in that the central release plot contained only Chysanthemoides and the test plant plots were more distant (5m) from the central release plot. In the Australian field experiment, the central plot also contained test species other than the target plant, and it was on these plants that some oviposition occurred, not on plants in the surrounding plots. Only in
Page 64 of 84 the “supplementary field oviposition test” carried out by the Australians did the central plot contain only the target plant. In this case no oviposition on non-target plants occurred. Therefore we have no data from tests with NZ plants equivalent to those where non-target oviposition occurred in the Australian tests.
Ecotype/strain variability The applicants have indicated (section 3.5) that should the release in NZ be approved, they will establish a rearing facility for BSLR in South Africa using specimens collected from western Cape Province, which is the area from which the insects used in host range tests were collected. These will be shipped to NZ from this facility. If this intention is adhered to, then there should be no problem with ecotype variability which might result in the release of ecoytypes which have different host range characteristics.
Parasitoids The applicants have indicated that they will follow containment procedures required by a MAF Import health Standard issued under the Biosecurity Act 1993. This will require insects to be reared in quarantine for a period which was not specified by the applicant. This should ensure that no parasitoids are released in to NZ along with BSLR.
Potential to establish in NZ Boneseed occurs around much of the NZ coastline, so presumably a biological control agent that will establish throughout NZ is required. The BSLR occurs naturally in South Africa, has established in Australia including Tasmania, so it could be assumed that it will establish throughout NZ. The applicants have not specifically mentioned the potential for BSLR to establish in NZ. It is possible that the number of generations (and hence speed at which populations build up) is temperature dependent and therefore might vary between northerly and southerly populations.
The impact of BSLR in Australia has apparently been limited by predation. After release, it is likely that parasitoids present in NZ will attack BSLR, as indicated by the applicant (section 3.4). A number of parasitoids were found attacking tortricids feeding on boneseed in NZ and there is a risk that these parasitoids will also attack BSLR. This could be investigated in containment, but the extent to which the biological control programme in NZ will be compromised as a result of natural enemy activity cannot be predicted with any accuracy pre-release.
2) Please respond to the following issues raised by submitters regarding the testing regime. Please note your information will form the basis of the project team's response to the issues raised in sections 9.2.2 and 9.2.3 of the E&R report.
9.2.2: There is a risk that host range tests carried out overseas will not be indicative of results that might be obtained in NZ because of any number of differences in biological, ecological and physical conditions. However, we are clearly unable to carry out tests in the field in NZ pre-release, so it is frequently argued that tests in the field in the country of origin of the proposed biological control agent are likely to give more realistic data on potential host range than tests in the laboratory in NZ.
Page 65 of 84 9.2.3: No-choice tests are more conservative than choice tests because they determine the range of species upon which a herbivore is able to develop physiologically (Hill, 1999). These tests are usually carried out initially on a range of test plants chosen using the centrifugal phylogenetic testing protocol to establish a profile of potential host species. Choice tests are then carried out to add more information on host preferences, especially using test plant species that were attacked in the no-choice tests. Oviposition tests are often considered to be more indicative of ecological host range than larval feeding tests. In this case no-choice tests were carried out in the laboratory by the Australian but not the NZ researchers. The former showed that full development of BSLR occurred on 36 spp. of plants in 11 families. The choice-testing carried out indicated quite clearly that boneseed leafroller has a strong preference for boneseed when that species is present. However, choice tests also showed that oviposition would occur occasionally on other species or even non-plant material in the presence of boneseed. Similarly choice tests showed that larvae were able to develop fully on alternative species. These results in combination indicate that there must remain a degree of uncertainty in predicting post-release host specificity of BSLR in NZ.
a/ Three submitter's (6324, 6327 and 6331) concerns about the extrapolation of data obtained overseas to New Zealand conditions, can you comment on the validity of this practice. In particular 6331 raised concerns about the use of material not sourced in New Zealand and regional/continental influences on foliage quality that could impact on results.
In all risk assessments for biological control agents it is necessary to extrapolate to some extent, whether it is from laboratory tests to field conditions, or field conditions overseas to those in NZ. Conditions under which plants grow can certainly influence foliage quality and hence host selection. However, even within NZ, species can vary locally, and so while it is a factor that should be considered, it might be to some extent outweighed by the benefit of having plants growing outside in semi-natural conditions for evaluation of non-target effects. An argument could be made that field testing overseas with NZ plants is more realistic than testing in containment in NZ. Ideally both should be done as mentioned above. Submitter 6331 recommends that final testing with NZ plant material should be carried out in containment in NZ. It may be preferable, however, to use this an an initial screen for a wide range of NZ plants, and follow this up with field testing overseas using species which have proved to be suitable for larval development in the containment tests. b/ With respect to submissions 6329, 6332, we would appreciate your comment based on the results presented whether or not non-target effects are likely to occur.
Submitter 6329 argues that there is insufficient conclusive evidence that there will be no non-target impacts on indigenous plant or moth species, or exotic food plants, and that BSLR will not mutate post-release to produce a new species that will dominate in the environment. Submitter 6332 similary shares concerns about the level of information available about potential non-target effects.
There can be no hard and fast guarantees about the outcome when a new species is introduced in to an existing ecosystem. There will always be an element of
Page 66 of 84 uncertainty in predicting impacts simply because ecosystems are extremely complex and our knowledge of how they function is consequently relatively rudimentary. However, weed biological control programmes, when founded on well accepted pre- release studies, have an impressive environmental safety record (Fowler et al., 2000). There are two well publicised examples of non-target impacts of weed biological control agents, Rhinocyllus conicus attacking native Cirsium (thistle) species in the USA, and Cactoblastis cactorum attacking rare species of Opuntia in Florida. In neither case did pre-release study fail to predict attack on non-target plants, in fact both species were known to have broad host ranges, but decisions to import and release were made by US regulators despite the evidence. In the case of BSLR the proposal is based heavily upon Australian research. This leaves a degree of uncertainty about the potential impacts of BSLR on NZ native plants, of which only three were included in field-based tests in South Africa.
In the opinion of John Dugdale, who is recognised as an expert on Tortricidae, the likelihood of BSLR inter-breeding with NZ native Tortricide is “extremely improbable” (Appendix 7 of the application). I have no evidence to support a contrary view. Just like the target pest organisms, there is always the possibility that over time biological control agents will evolve post-release as they adapt to new environments e.g. Hufbauer (2002), sometimes to the advantage of the biological control programme.
c/ 6331 and 6333 - can you comment on the need for further testing based on the criteria described by Wapshere and by Withers choice and non-choice methods in a field trial.
The complementarity and benefits of non-choice and choice tests, laboratory containment and open field tests has been outlined above. The choice of NZ native test plants for testing appears not to have been selected on the basis of the generally accepted standards of Wapshere (1974, 1989). Rather the applicants have relied heavily upon the extrapolation of Australian tests which were carried out using standard protocols. However, the Australian research showed that BSLR is not monophagous, that it has a wide physiological host range, female moths occasionally oviposit on a few non-target species, and larvae can complete full development on some non-target species. They also showed from field surveys that this appears not to occur in the field in South Africa. These data raise a level of uncertainty about what is likely to happen in the field in NZ, especially given the limited testing carried out on NZ native plants. I believe further testing to determine the physiological host range of BSLR in containment in NZ would provide the basis for further field testing that could perhaps be carried out in South Africa (or now in Australia). This information would provide for a stronger basis upon which to assess risk to non-target plants in NZ. Notwithstanding concerns for NZ native plants, we are evaluating the risk of introducing a species which has been shown to be capable in the laboratory of developing through to successful pupation on Trifolium repens, estimated to be worth over $3 billion to the NZ economy (Caradus et al., 1996) so caution would seem to be well justified. d/ Can you respond to the comment by 6331 that the Australian testing was "more extensive and was of a higher scientific standard than the Landcare" research.
Page 67 of 84
The Australian testing programme was certainly more comprehensive and appeared to follow standards and protocols now generally accepted for weed biological control. The NZ testing was limited and was based upon the premise that it was possible to extrapolate from the Australian study. e/ Can you comment on the statement made by 6331 that the absence of the host long term selection pressure would result in a non-target favouring population of boneseed leafroller.
The absence of the target host resulting from successful biological control by a herbivore may have several outcomes depending upon the degree of host specificity of the biological control agent. In the case of a strictly monophagous biological control agent, complete removal of the host (which is very unlikely in any biological control programme), population collapse of the herbivore would be expected. Given a herbivore with a very wide host range, then one would expect the biological control agent to utilise other suitable plant species within its acceptable host range. For cases in between these extremes, one would predict a similarly wide range of scenarios. Should an acceptable alternative host plant be available, even if not initially a preferred species, then it is not inconceivable to imagine that a population which can develop successfully on the alternative species will be at a selective advantage.
Page 68 of 84 Appendix 3: Dr Ilse Breitwieser’s Report 1) Please provide a brief review of the phylogeny of boneseed with particular emphasis on its relationship to New Zealand natives or introduced species of economic or amenity value.
The Asteraceae is one of the largest families of eudicots including over 25,000 species in around 1500 genera. They occur worldwide, except Antarctica, and are found in all but the most extreme habitats and reach their greatest numbers in arid and temperate regions such as the Mediterranean region, Mexico, the Cape Province of South Africa and woodland, grassland and bushland vegetation of Africa, South America and Australia. The family includes many economically important plants such a lettuce, artichokes, and sunflowers, as well as, popular garden plants such as dahlias, chrysanthemums, and marigolds along with weeds like dandelions and thistle. There are 287 indigenous species of Asteraceae found in New Zealand; 261 of these are endemic. The Asteraceae surpasses all other families in the number of species listed as threatened or uncommon in New Zealand (see Phylogeny of New Zealand Plants http://nzflora.landcareresearch.co.nz/Phylogeny/WebForms/Home.aspx).
The Asteraceae can be divided into three subfamilies, the small subfamily Barnadesioideae and the much bigger subfamilies Cichorioideae and Asteroideae (e.g., Bremer 1994). More recently Panero and Funk (2002) proposed the recognition of 11 subfamilies. The new subfamilies clarify the phylogenetic relationships of the Cichorioideae. The Asteroideae can be divided into several tribes, from about 10 tribes (e.g., Bremer 1994) to 20 tribes (Panero and Funk 2002). All recent publications treat tribes Senecioneae, Calenduleae, Gnaphalieae, Astereae, Anthemideae, Inuleae / Plucheae as closely related (e.g., Bremer 1994, Bayer and Starr 1998, Wagstaff and Breitwieser 2002). In Panero and Funk (2002) the Calenduleae, the tribe to which Chrysanthemoides monilifera (boneseed) belongs, is sister to the Gnaphalieae, Astereae and Anthemideae. Based on these publications, we need to consider the Gnaphalieae, Astereae, Anthemideae, Senecioneae, Inuleae / Plucheae as closest relatives of the Calenduleae.
Chrysanthemoides monilifera (boneseed) is one of two species of the genus Chrysanthemoides. Chrysanthemoides is one of 8 genera and about 110 species of tribe Calenduleae.
Relationships of Chrysanthemoides to New Zealand’s indigenous plants (based on Wagstaff and Breitwieser 2002): New Zealand does not have any indigenous Calenduleae. The closest related tribes to the Calenduleae are having numerous indigenous genera and species: Gnaphalieae (10 genera, ca. 70 species); Astereae (8 genera, ca. 120 species); Anthemideae (2 genera, ca. 26 species); Senecioneae (6 genera, ca. 58 species). The Inuleae / Plucheae and other Asteroideae are not indigenous to New Zealand (- Centipeda belongs to the Inlueae but it is not certain whether it is indigenous).
List of New Zealand’s indigenous genera in Asteroideae: Anthemideae: Cotula, Leptinella
Page 69 of 84 Astereae: Brachyscome, Celmisia, Damnamenia, Lagenifera, Olearia, Pachystegia, Pleurophyllum, Vittadinia Gnaphalieae: Anaphalioides, Craspedia, Euchiton (incl. here the recently published genus Argyrotegium), Ewartia, Helichrysum, Leucogenes, Ozothamnus, Pseudognaphalium, Rachelia, Raoulia Senecioneae: Abrotanella, Brachyglottis, Dolichoglottis, Haastia, Senecio, Traversia Inuleae: Centipeda (?)
Relationships of Chrysanthemoides to New Zealand’s introduced plants: New Zealand has many naturalised Asteraceae. Most closely related to Chrysanthemoides are Calendula and Osteospermum, the only other genera naturalised in New Zealand that also belong to the Calenduleae.
2) Please comment on the selection of plants for the host range testing. In particular do you feel that any New Zealand natives or introduced species of economic or amenity value that are closely related to boneseed have been omitted.
Chrysanthemoides is a member of the Calenduleae. New Zealand does not have any indigenous Calenduleae. There are no closely related indigenous genera of Chrysanthemoides. Calendula and Osteospermum are the only other genera of Calenduleae naturalised in New Zealand. They were tested by the Australian study.
I have provided comments above on the relationships of Chrysanthemoides and presented a list of Asteroideae genera indigenous to New Zealand. None of these genera is closely related to Chrysanthemoides. All New Zealand indigenous genera of Asteroideae belong to the tribes closest to the Calenduleae. As a plant systematist I can provide information on the systematics (incl. phylogeny) of New Zealand’s species, but it is up to biocontrol experts to judge how many of the about 280 species of New Zealand’s indigenous Asteroideae need testing.
The New Zealand test is based on one species each of all relevant tribes of Asteroideae (two species of Gnaphalieae and Astereae) and two tribes of Cichorioideae. The selection criteria for the species tested were not stated in the application. The selection of the species was obviously not based on whether the species are indigenous or exotic to New Zealand nor on whether the exotic species selected are of particular economic or amenity value. Of the 11 species tested, three are indigenous to New Zealand. None of the exotic species tested are of “importance to New Zealand”. I assume the test was based on representatives of at least one genus of each tribe of Asteroideae indigenous or exotic to New Zealand that were available for testing in South Africa. I assume the applicants’ criterion was that the genus occurs in New Zealand, but not necessarily the species (i.e., Arctotis angustifolia does not occur in New Zealand).
3) Based on the phylogeny review above please comment on the value of including the following plants in host-specificity testing as suggested by submitters (summarised below).
Page 70 of 84
Submitters (see below) were mainly concerned about the lack of testing of Asteraceae species indigenous to New Zealand. They considered the majority of the selected plant species as not being of importance to New Zealand. They recommended to include representatives of New Zealand indigenous species of (1) Asteraceae tribes that are closely related to Calenduleae; (2) Asteracae that have at least some or potential habitat overlap with Chrysanthemoides; (3) other related families that have some or potential habitat overlap with Chrysanthemoides or that are highly palatable species.
The applicants concluded that “The results of this field test add to the evidence already available form field observation in South Africa, and host specifity testing conducted in Australia, that “Tortix” sp. should present a negligible risk to non- target plant species in New Zealand”. The applicants commented on the high degree of host specifity of “Tortix” sp.:“testing would normally concentrate on plants of importance to New Zealand, such as natives, crops and ornamentals, from the same subfamily (Asteroideae) and tribe (Calenduleae)”. The application is based mainly on results of testing the host specifity of “Tortix” sp. for Australia. The applicants supplemented these results by testing representivative genera indigenous or exotic to New Zealand from tribes in the Asteraceae. They selected one species of each tribe closely related to Calenduleae (two species for Gnaphalieae and Astereae) and one species each of two tribes of Cichorioideae.
The application does not contain the criteria for selection of the species used for testing the host specifity of “Tortix” sp. for New Zealand (see point 2). The applicants were obviously not concerned about whether these species are indigenous or exotic to New Zealand nor whether they are “plant species of importance to New Zealand”. However, the submissions recommend numerous species for testing (see below). It needs to be assessed (1) whether some / all of the species suggested for testing in the submissions are available or could be grown at the CSIRO Biological Contro Unit based at the University of Cape Town, South Africa; (2) how many indigenous species of not close relationships and how many species with habitat overlap are standard for biological control testing.
I believe the applicants did not explain sufficiently their criteria for selection of plant species used for testing the host specifity of the boneseed leafroller for New Zealand. Therefore it is not possible to make recommendations on whether the species suggested for testing by the submitters could be or should be tested.
4) Specific comments: Please note your information will form the basis of the project team's response to the issues raised in 9.2.4 of the E&R report a/ 6328 - can you comment on the omission of any species of cultural importance in the existing host-specificity testing? b/ 6330 - can you comment on the need to do further testing on members of the Myrtaceae family and the Leptinella, Olearia, Senecio and Linum c/ 6331 – Please comment on issues raised about the importance of certain species included in the host-specificity testing and the need to do host testing on the
Page 71 of 84 additional species listed below: A further ten plants described as "of importance to New Zealand" were tested for the New Zealand application. It is difficult to see how most of these species could be described as "of importance to New Zealand". Two are endemic species (one common in the wild, one in cultivation), so that's fine. A third species (Pseudognaphalium luteoalbum) is native but since it is virtually cosmopolitan and in most places is regarded as a weed of minor importance, it is hard to see how that could be described as importance to New Zealand. The remaining species tested are introduced. None could be considered of economic importance (there are very few daisies of economic importance except as weeds anyway). Sonchus oleraceus, while not native, is the Maori vegetable puha and for this reason it may be acceptable to consider it is important to New Zealand and worth testing, although it is in a different subfamily from Chrysanthemoides. Cotula turbinata is a South Africa species that was recorded as a casual from ballast heaps around Wellington early last century. It is now occasionally found in cultivation but is not known from the wild any more. Therefore it is hard to see why it has been considered "important". Bellis perennis is a common lawn weed and double flowered forms are sold as ornamentals. Liatris spicata is cultivated, most garden centres have a few plants amongst their perennial selections but it isn't particularly popular. It is recorded from the wild very occasionally as a casual. A couple of species of Arctotis and a hybrid are in New Zealand, both in the wild and cultivated to some extent, although their "importan[ce]" is highly questionable. In the report there is a mention of testing Arctotis angustifolia, which is not listed as being present in New Zealand (MAF Plant Biosecurity Index). This is a fairly uncommon South African endemic species and therefore we would question the relevance of testing this species is for a New Zealand release. The remaining two species are weeds of varying importance. If Tortrix were to attack Erigeron karvinskianus it would be good in that this species is a significant problem in some areas, except that of course that would mean that the agent wasn't very specific (Erigeron is tribe Astereae). Whether or not it attacks Bidens pilosa, a minor weed from the Bay of Plenty northward, seems of little relevance. From the above discussion "a further 10 plants of importance to New Zealand" does not appear to be a correct description of this list. From a conservation perspective only Ozothamnus and the Brachyglottis are relevant at all. The Department therefore considers that even if the Australian results were not ambiguous, this level of testing of nontarget plants in New Zealand is inadequate.
I agree with the submission that most of the plant species selected for testing the host specificty of “Tortix” sp. for New Zealand do not represent plants of “importance of New Zealand”.
Plants for further host testing: The Department believes that particular attention should be focused on native members of the Astereae, Anthemideae and Gnaphalieae, which are the closest tribes to Calenduleae with native members. The following list of plants species are considered appropriate candidates for the further host testing prior to approval of this release. All of the species grow in coastal habitats and potentially could grow with boneseed. Classifications of threatened species are from P. J. de Lange, D. A. Norton, P. B. Heenan, S. P. Courtney, B. P. J. Molloy, C. C. Ogle, B. D. Rance, P. N. Johnson and R. Hitchmough (2004). Threatened and uncommon plants of New Zealand. NZJ Botany 42:45-76. Asteraceae Gnaphalieae
Page 72 of 84 Ozothamnus leptophyllus (not threatened, some habitat overlap with boneseed) Raoulia tenuicaulis (not threatened, some habitat overlap with boneseed) Euchiton keriense (not threatened, some habitat overlap with boneseed)
There is no species called Euchiton keriense. I assume the submission means Anaphlioides keriensis.
Craspedia uniflora var. grandis (not threatened, some habitat overlap with boneseed) Anaphaliodes bellidiodes or Helichrysum filicaule (not threatened, some habitat overlap with boneseed)
Astereae Olearia paniculata (not threatened, some habitat overlap with boneseed) O. pachyphylla (nationally endangered) or O.furfuracea (non-threatened related species)- habitat overlap with boneseed. Celmisia monroi (not threatened species that grows in a range of habitats including coastal, use material from Kekerengu) or C. major (gradual decline) both have habitats, which overlap with boneseed. Lagenifera lanata (sparse) or L. pumila (non-threatened related species)
Anthemideae Leptinella rotundata (gradual decline, highly palatable species)
Senecioneae Senecio sterquilinus (range restricted, grows in areas near where boneseed is) or S. scaberulus (nationally endangered) Brachyglottis kirkii (serious decline) potential overlap with boneseed; also leaves are a similar size, shape and fleshiness B. repanda (not threatened, some habitat overlap with boneseed)
Lactuceae Microseris scapigera (material sourced from NZ) Kirkianella "glauca" (range restricted, grows in areas near where boneseed is) or Sonchus kirkii (gradual decline)
Linaceae Linum monogynum (not threatened, some habitat overlap with boneseed)
Rosaceae Acaena juvenca (not threatened, some habitat overlap with boneseed) Rubus schmideliodes (not threatened, potentially some habitat overlap with boneseed)
Apiaceae Apium prostratum (not threatened, some habitat overlap with boneseed)
Convolvulaceae Calystegia soldanella (not threatened, some habitat overlap with boneseed)
Thymeleaceae Pimelea arenaria (not threatened, some habitat overlap with boneseed)
Page 73 of 84
Brassicaceae Lepidium oleraceum (nationally endangered, habitat overlap, highly palatable species with leaves which are a similar size, shape and fleshiness to boneseed)
Based on most recent research on phylogeny of Asteraceae, tribes Astereae, Anthemideae, Gnaphalieae, Senecioneae and Inuleae / Plucheae need to be regarded as the closest tribes to the Calenduleae (e.g., Bremer 1994, Bayer and Starr 1998, Wagstaff and Breitwieser 2002, Panero and Funk 2002). These tribes are represented in New Zealand by at least 280 indigenous species. None of the species indigenous to New Zealand are closely related to Chrysanthemoides. From these 280 species the Department is listing a selection of those species which have at least some or potential habitat overlap with Chrysanthemoides. The Department also recommends to include species of other families with habitat overlap. The information of the species in the list is correct (except for one species) and all species could be considered as appropriate. The Department’s list of Asteraceae with habitat overlap is not comprehensive (e.g., Gnaphalieae – Craspedia minor, Craspedia uniflora var maritima, Raoulia aff. hookeri, Euchiton audax are not included). I assume the list is not meant to be comprehensive. The list of species from other families represents a good selection.
d/ 6333 - please comment on the need to do host testing on the additional species listed below: Lack of host-specificity testing on New Zealand native plant species. There has been insufficient host specificity testing on New Zealand native plant species. Of the 11 species tested between October and December 2001 (listed on Page 10, Appendix 1), only 5 of these are native to New Zealand. These are Cotula turbinata, Ozothamnus leptophyllus (formerly Cassina leptophylla, tauhinu), Pseudognaphalium luteoalbum, Bidens pilosa, and Brachyglottis greyii.
Cotula turbinata and Bidens pilosa are not native to New Zealand.
Many obvious native Asteraceae candidates for host specificity testing have not yet been tested. Host specificity testing has concentrated on other plant species in the family Asteraceae. There are many other native plant genera within this family which have not been tested, including 1. Celmisia, at least 48 species 2. Haastia, 3 species 3. Helichrysum, 9 species plus varieties 4. Leptinella, 24 species 5. Olearia, 33 species 6. Pachystegia, 3 species 7. Senecio, 18 species Host specificity testing should also have a focus on native plants of coastal environments, as this is the preferred habitat of the weedy shrub Chrysanthemoides.
Page 74 of 84 Some native species to be considered for host specificity testing include: * Sonchus kirkii, puha * Tetragonia tetragonoides, T. trigyna, New Zealand spinach, kokihi * Chenopodium spp., * Lepidium spp, Cooks scurvy grass * Helichrysum lanceolatum, niniao
See comments under point 3
References: Bayer, R. J.; Starr, J. R. 1998: Tribal phylogeny of the Asteraceae based on two non- coding chloroplast sequences, the trnL intron and the trnL/trnF intergenic spacer. Annals of the Missouri Botanical Garden 85: 242-256.
Bremer, K. 1994: Asteraceae - cladistics and classification. Timber Press, Oregon.
Panero, J. L.; Funk, V. A. 2002: Toward a phylogenetic subfamilial classification for the Compositae (Asteraceae). Proceedings of the Biological Society of Washington 115: 909-922.
Wagstaff, S. J.; Breitwieser, I. 2002: Phylogenetic relationships of New Zealand Asteraceae inferred from ITS sequences. Plant Systematics and Evolution 231: 203- 224.
Page 75 of 84 Appendix 4: Decision Pathway
Page 76 of 84 NOTES to Figure 5 - Decision path for applications to import for release or release a new organism from containment (application made under Section 34 of the Act and determined under Section 38 of the Act)
Section 38B of the Act allows the Authority, with the agreement of the applicant, to treat an application for release under section 34 as if it were an application for conditional release. A decision to do this will normally be made prior to the commencement of consideration using this decision path.
Items 1, Information that should be reviewed includes that in the application, the E&R 2 & 3: Report, from experts and in submissions (where relevant). Review should occur in terms of section 38A(2) of the Act and clauses 8, 15, 16, 20, 22 and 23 of the Methodology. Additional information may need to be sought under s58 of the Act.
If the applicant is not able to provide sufficient information for consideration then the application may be declined under section 38 (1)(b)(iii) of the Act.
Item 4: The range of risks, costs and benefits to be identified should be that covered by clauses 9, and 10 of the Methodology. There are two steps within this part of the process:
Step 1: The identification of every possible risk, cost and benefit that can be thought of, to ensure that a starting position is as comprehensive as possible.
Step 2: The elimination of those risks, costs and benefits that can be readily concluded to be negligible or irrelevant, having regard to the characteristics of the organism and the circumstances of the application.
Item 5: All risks (and costs) identified in Item 4 should be assessed, taking into account the matters set out in section 37.
Section 37 of the Act requires that the Authority takes account of the ability of the organism to establish an undesirable self sustaining population and the ease of eradication. This needs to be considered in an integrated way in the assessment process because of the reference to “undesirable”. Undesirable means (in effect) able to create significant risks.
The assessment of risks (and costs) should be carried out in accordance with clauses 22 and 25, 29 to 32 (dealing with uncertainty), 12 to 14 and 33 of the Methodology and relevant clauses of the HSNO Act.
The process of risk assessment is not linear. It is very iterative. In essence all of the steps must be repeated until a satisfactory conclusion is reached. However, in general terms the steps in the assessment process should be as follows, for each risk. Step 1: Assessment of the likelihood of an adverse effect occurring and the magnitude of that effect if it should occur.
Page 77 of 84 Step 2: Consideration of the extent to which the risk will be mitigated by the ability to eradicate the organism if a significant adverse effect eventuated. Step 3: Consideration of how risk averse or cautious the Authority should be in giving weight to the resulting or remaining risk, in terms of clause 33 of the Methodology.
Item 6: Determine whether the organism meets the minimum standards set out in s36.
Item 7: Once the risks (and costs) have been assessed individually, consider all risks (and costs) together.
Item 8: Consider whether the combined risks and costs are negligible. An holistic perspective should be adopted, taking into account the particular characteristics of the organism and the context of the application.
Item 9: This item constitutes a decision made under clause 26 of the Methodology. If risks and costs are negligible and there are no external costs (costs accrue only to the applicant), then the fact that the application has been submitted is deemed to demonstrate existence of benefit, and no further benefits need be considered. However, if significant costs and/or risks exist then all benefits need to be assessed.
Item 10: Although ‘risk averseness’ is considered as a part of the assessment of individual risks, it is good practice to consolidate the view on this if risks are non-negligible. Clause 33 of the Methodology applies, as does section 7 of the Act dealing with caution in the face of scientific and technical uncertainty.
Item 11: Assess benefits in terms of clause 13 of the Methodology.
Item 12: In weighing up adverse and beneficial effects (costs, risks and benefits), clause 34 of the Methodology applies.
Where this item is taken in sequence from items 10 and 11 (i.e. risks are not negligible) it constitutes a decision made under clause 27 of the Methodology, and adverse effects comprise risks and costs.
Where this item is taken in sequence from items 9 and 11 (i.e. risks are negligible, and costs do not accrue only to the applicant) it constitutes a decision made under clause 26 of the Methodology.
Page 78 of 84 Appendix 5: Qualitative scales for risk assessment
Qualitative Risk Assessment
Risks and benefits are assessed by estimating the magnitude of the possible effects and the likelihood of their occurrence. For each effect, the combination of these two components determines the level of that effect, which is a two dimensional concept. Risk assessment may be qualitative or quantitative. Qualitative assessment is informed by quantitative data where this is available.
Qualitative matrices are used to prioritise risks (and benefits), and to identify any risks that are unacceptable. The measure of the level of risk (combination of magnitude and likelihood) is specific to the application therefore measures of level of risk should not be compared between applications. However, the measures (descriptors) for different types of risk (human health, ecological etc) should be established so that they represent relative orders of magnitude.
Magnitude of effect
The magnitude must be a measure of the endpoint (specified by the Act and the Methodology), and is described in terms of the element that might be affected. The magnitude of the effect is not the same as the effect itself. The qualitative descriptors for magnitude of effect are surrogate measures that should be used to gauge the end effect or the ‘what if’ element.
Tables 1 and 2 contain generic descriptors for magnitude of adverse effects (risks and costs) and beneficial effects (benefits). These descriptors are examples only, and their generic nature means that it may be difficult to use them in some particular circumstances. They are included here simply to illustrate how qualitative tables may be used to represent levels of risk.
Page 79 of 84 Table 1 Magnitude of adverse effect Descriptor Examples of descriptions Minimal Mild reversible short term adverse health effects to individuals in highly localised area Highly localised and contained environmental impact, affecting a few (less than ten) individuals members of communities of flora or fauna, no discernible ecosystem impact Low dollar cost of containment/cleanup/repair (<$5,000) No social disruption8 Minor Mild reversible short term adverse health effects to identified and isolated groups9 Localised and contained reversible environmental impact, some local plant or animal communities temporarily damaged, no discernible ecosystem impact or species damage Dollar cost of containment/cleanup/repair in order of $5,000-$50,000 Potential social disruption (community placed on alert) Moderate Minor irreversible health effects to individuals and/or reversible medium term adverse health effects to larger (but surrounding) community (requiring hospitalisation) Measurable long term damage to local plant and animal communities, but no obvious spread beyond defined boundaries, medium term individual ecosystem damage, no species damage Dollar cost of containment/cleanup/repair in order of $50,000- $500,000, Some social disruption (eg people delayed) Major Significant irreversible adverse health effects affecting individuals and requiring hospitalisation and/or reversible adverse health effects reaching beyond the immediate community Long term/irreversible damage to localised ecosystem but no species loss Dollar cost of containment/cleanup/repair in order of $500,000- $5,000,000 Social disruption to surrounding community, including some evacuations Massive Significant irreversible adverse health effects reaching beyond the immediate community and/or deaths Extensive irreversible ecosystem damage, including species loss Dollar cost of containment/cleanup/repair greater than $5,000,000 Major social disruption with entire surrounding area evacuated and impacts on wider community
8 The concept of social disruption includes both physical disruption, and perceptions leading to psychological disruption. For example, some chemicals may have nuisance effects (through odour) that result in communities feeling threatened. 9 Note that the reference to ‘groups’ and ‘communities’ in the context of human health effects includes the notion of groups defined by health status.
Page 80 of 84
The economic effects category has been given a surrogate magnitude. This is for demonstration as a means of illustrating the type of magnitudes that might be encountered. Table 2 Magnitude of beneficial effect Descriptor Examples of descriptions Minimal Mild short term positive health effects to individuals in highly localised area Highly localised and contained environmental impact, affecting a few (less than ten) individuals members of communities of flora or fauna, no discernible ecosystem impact Low dollar benefit (<$5,000) No social effect Minor Mild short term beneficial health effects to identified and isolated groups Localised and contained beneficial environmental impact, no discernible ecosystem impact or species damage Dollar benefit in order of $5,000-$50,000 Minor localised community benefit Moderate Minor health benefits to individuals and/or medium term health impacts on larger (but surrounding) community and health status groups Measurable benefit to localised plant and animal communities expected to pertain to medium term. Dollar benefit in order of $50,000-$500,000, Local community and some individuals beyond immediate community receive social benefit. Major Significant beneficial health effects to localised community and specific groups in wider community Long term benefit to localised ecosystem(s) Dollar benefit in order of $500,000-$5,000,000 Substantial social benefit to surrounding community, and individuals in wider community. Massive Significant long term beneficial health effects to the wider community Long term, wide spread benefits to species and/or ecosystems Dollar benefit greater than $5,000,000 Major social benefit affecting wider community
Likelihood of effect occurring
Likelihood in this context applies to the composite likelihood of the end effect, and not either to the initiating event, or any one of the intermediary events. It includes : . the concept of an initiating event (triggering the hazard), and
Page 81 of 84 . the exposure pathway that links the source (hazard) and the area of impact (public health, environment, economy, or community).
The likelihood term applies specifically to the resulting effect or the final event in the chain, and will be a combination of the likelihood of the initiating event and several intermediary likelihoods10. The frequency or probability solely of the initial incident or hazard event should not be used (as it sometimes is in the safety discipline).
The best way to determine the likelihood is to specify and analyse the complete pathway of the “chain of events” from source to the final environmental impact or effect. Each event in the chain is dependent upon the previous event occurring in the first place.
Likelihood may be expressed as a frequency or a probability. While frequency is often expressed as a number of events within a given time period, it may also be expressed as the number of events per head of (exposed) population. As a probability the likelihood is dimensionless and refers to the number of events of interest divided by the total number of events (range 0-1). Table 3 Likelihood (adverse effect) Descriptor Description 1 Highly improbable Almost certainly not occurring but cannot be totally ruled out 2 Improbable (remote) Only occurring in very exceptional circumstances. 3 Very unlikely Considered only to occur in very unusual circumstances 4 Unlikely (occasional) Could occur, but is not expected to occur under normal operating conditions. 5 Likely A good chance that it may occur under normal operating conditions. 6 Very likely Expected to occur if all conditions met 7 Extremely likely Almost certain
Table 3 provides an example of a set of generic likelihood descriptors for adverse and beneficial effect. Note that when estimating these likelihoods, the impact of default controls should be taken into account.
The table is not symmetrical. This is to allow for classification of very low probability adverse effects.
In practical terms, where the exposure pathway is complex, it may be conceptually difficult to condense all the information into a single likelihood. For any risk where the likelihood is other than ‘highly improbable’ or ‘improbable’, then an analysis of the pathway should include identifying the ‘critical points’; the aspects that are the most vulnerable, and the elements where controls might be used to ‘cut’ the pathway.
10 Qualitative event tree analysis may be a useful way of ensuring that all aspects are included.
Page 82 of 84 Calculating the level of risk
Using these qualitative descriptors for magnitude of effect and likelihood of the event occurring, an additional two-way table representing a level of risk (combined likelihood and measure of effect) can be constructed as shown in Table 4, where six levels of effect are allocated: A, B, C, D, E and F. These terms have been used to emphasise that the matrix is a device for determining which risks (benefits) require further analysis to determine their significance in the decision making process. Avoiding labels such as ‘low’, ‘medium’, and ‘high’ removes the aspect of perception.
The lowest level (A) may be deemed to be equivalent to ‘insignificant’. In this table ‘A’ is given to three combinations; minimal impact and an occurrence of improbable or highly improbable, and minor impact with a highly improbable occurrence. In some cases where there is high uncertainty it may be preferable to split this category into A1 and A2, where only A1 is deemed to equate to insignificant.
For negative effects, the levels are used to show how risks can be reduced by the application of additional controls. Where the table is used for positive effects it may also be possible for controls to be applied to ensure that a particular level of benefit is achieved, but this is not a common approach.
Table 4 Calculating the level of risk (benefit) Magnitude of effect Likelihood Minimal Minor Moderate Major Massive Highly improbable A A B C D Improbable A B C D E Very unlikely B C D E E Unlikely C D E E F Likely D E E F F Very likely E E F F F Extremely likely E F F F F
The table presented here is symmetric around an axis from highly improbable and minimal to massive and extremely likely, however, this will not necessarily be the case in all applications.
Impact of uncertainty in estimates
Uncertainty may be taken into account in two ways. Firstly, when describing a risk a range of descriptors may be used. For example, a risk may be allocated a range of very unlikely- improbable, and minor-major. This would put the range of the risk as B through E.
Alternatively, the level of risk (or benefit) may be adjusted after it has been estimated on the grounds of uncertainty.
Page 83 of 84 Appendix 6: Submissions received by ERMA New Zealand If you are viewing this document via the world-wide-web submissions are available as a PDF file from the same site.
Page 84 of 84