ER-AF-NO1-4 3/99

FORM 1

Application for approval to

IMPORT FOR RELEASE OR RELEASE FROM CONTAINMENT ANY NEW ORGANISM

under Section 34 of the Hazardous Substances and New Organisms Act 1996

Office use only

Fees $ Date received / / Verified date / / Job manager

ER-AF-NO1-4 3/99 Application for approval to import for release or FORM 1 release from containment any new organism under Section 34 of the Hazardous Substances Page 1 and New Organisms Act 1996

IMPORTANT Before you fill in this application form please talk to ERMA New Zealand. We can help you scope and prepare your application. The scale of information we need should match the potential significance of the application. For example, applications which may pose a significant risk to the environment or to human health need to be supported with more substantial information than applications which clearly pose a more minor risk. We need all relevant information early on in the application process. Quality information up front will speed up the process. Any extra material that does not fit in the application form must be clearly labelled and cross-referenced in the application form. Commercially sensitive information should be collated in a separate document. This form is in three parts. If you think your application may qualify for rapid assessment please check with us first and then complete only Parts A and B. Non rapid assessment applicants should complete Parts A and C only. All applicants must sign the form at the end of Part A and enclose the correct application fee. Please check ERMA New Zealand’s current pricing policy, we are unable to process applications that do not contain the correct fee. All references to regulations in this form, unless otherwise noted, refer to the Hazardous Substances and New Organisms (New Organisms Forms and Information Requirements) Regulations 1998. Copies of all our application forms will soon also be available on our website: www.ermanz.govt.nz, and also in electronic form (MS Word format). If you have any suggestions for improvements to this form, please contact our operations staff at the address below. You can get more information at any time by telephoning, writing to, or calling in at our Wellington office. One of our staff members will be able to help you.

List of application forms for new organisms: These are all our application forms related to new organisms. Please check you have the right one. Form 1 Application for approval under section 34 of the Act to import for release, or release from containment, any new organism – including rapid assessment (this form). Form 2 application for approval under section (40)(1)(a) of the Act to import into containment any new organism. Form 3 application for approval under section 40(1)(b) of the Act to develop in containment any genetically modified organism – including rapid assessment. Form 4 application for approval under section 40(1)(c) to field test (including large scale fermentation) in containment new organism. Form 5 application for approval under section 47 to use a new organism in an emergency. Form 6 application for approval under section 62 for grounds for reassessment of a new organism in containment.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 2

Applicant details

1. Name and address in New Zealand of the applicant: This should be the organisation or person formally responsible for this application. Name: Hieracium Control Trust Address: c/- Federated Farmers of New Zealand (Inc.) P.O. Box 665 Timaru

Phone: (03) 443 7155 Fax: (03) 443 7150

2. The applicant’s address for service in New Zealand (if different from above): Address: as above 3. Name of the contact person for the application (if different from applicant): This person should have sufficient knowledge to respond to queries and have the authority to make decisions on behalf of the applicant that relate to processing the application. Name: Mr John Aspinall Position: Chairman, Hieracium Control Trust Phone: (03) 443 7155 Fax: (03) 443 7150 Email: [email protected]

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 3

4. Summary Provide a summary of the information contained in this application relating to the identification of the organism. The information should include summaries of: - the identity of the organism; - if it is a genetically modified organism, the source of the donor nucleic acid material and the purpose of the modification; - what the organism will be used for and why it has been selected. Provide a summary of the information contained in this application relating to the assessment of the effects of the organism. The information should include summaries of: - the risks, costs and benefits of releasing the organism and the assessment of these. This summary will be used to provide information for those people and agencies (eg, Minister for the Environment, Department of Conservation, Regional Councils, etc.), who shall be notified of the application, and for potential submittors who request information. This information will also be used to prepare the public notice of the application. For these reasons, applicants should ensure that this summary information does not contain any commercially sensitive material. [ Yes/No? ] further information

Approval is sought by the Hieracium Control Trust to release from containment three new organisms (Insecta: Diptera): the hieracium gall fly, Macrolabis pilosellae (Binnie), (Cecidomyiidae), the root-feeding hover fly, Cheilosia urbana Meigen (formerly C. praecox (Zetterstedt)) and the crown-feeding hover fly, C. psilophthalma (Becker) (Syrphidae). These three insect species have been identified by Landcare Research and CABI Bioscience (CAB International,UK) as part of a suite of six potential control agents likely to have the greatest impact as biological control agents for hawkweeds, Hieracium species (Asteraceae), in New Zealand. All three flies feed only on hawkweed, and cause no harm to humans or other organisms.

Hawkweeds (Hieracium spp.) are serious weeds of hill and high country pastoral areas, as well as conservation land through large areas of New Zealand. As a result of invasion by Hieracium species a recent estimate placed one million stock units potentially at risk from reduced pasture production, representing a loss of up to $76M per annum. This figure does not include the environmental costs of losing large areas of native grasslands to these invasive weeds. The potential economic benefits of a successful biological control programme are such that costs should be recovered about 14 years following the release of agents. Environmental benefits of successful biological control include a return to healthier native grasslands, with increased biodiversity, and reduced erosion.

Permission to release two other insect species, hieracium plume moth, Oxyptilus pilosellae Zeller (Lepidoptera: Pterophoridae), and hieracium gall wasp, Aulacidea subterminalis Niblett (Hymenoptera: Cynipidae), was granted by MAF Regulatory Authority on 18 February 1998 and 19 November 1998 respectively. The accidentally introduced rust fungus, Puccinia hieracii var. piloselloidarum, was under study as one of the six potential control agents when it was found to be established already in New Zealand.

The most significant risk identified from the proposed release of these three new insects is the possible damage to non-target plants. However, host range tests have shown that this risk is very low: the insects fed and developed only on plant species in the genus Hieracium. Tests were carried out both in Switzerland and in an insect containment facility at Lincoln, New Zealand.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 4

Several complementary control agents are generally required for a successful weed biological control programme. In this case there are two reasons for this. One is that there are four separate weeds, and not all control agents will attack all of them. The other is that it is usually necessary to use agents that feed on different parts of the plant so that it is weakened from several directions. The rust affects the leaves of H. pilosella, but is not effective on all genetic variations of plants within this species. The gall wasp, Aulacidea subterminalis, also attacks only H. pilosella, causing damage to stolons. The plume moth, Oxyptilus pilosellae, attacks both H. pilosella, and H. caespitosum, feeding in the crown of plants. The three insect species that are the subject of the current application feed on a broader range of plants. The gall fly, Macrolabis pilosellae, damages leaves of H. praealtum as well as of H. pilosella, and H. caespitosum. The two Cheilosia species developed on all four New Zealand weedy Hieracium species, including the non-stoloniferous H. lepidulum. Cheilosia urbana feeds on roots of the plants, and C. psilophthalma damages the above-ground parts of the plant.

Organism details

5. The identification of the organism: This should include all information necessary to identify the organism and should include: - the taxonomic classification and name of the organism; - the essential characteristics that identify the organism and its behaviour in the environment; - sufficient information to enable the Authority to uniquely identify the organism in the register as required by section 20(2)(b) of the Act. (This section may also include the name by which the organism is generally known.) The information in this section would include, for example, information on the habitat range and climatic sensitivity of the organism. References to the scientific literature supporting this information should be given here if appropriate. In the separate box below the applicant should provide the name of the organism suitable for inclusion in the Authority’s public register. Information that is commercially sensitive should be clearly identified. If supplied separately, a cross-reference to it should be included. Taxonomic Name: Macrolabis pilosellae (Binnie 1878) (Insecta: Diptera: Cecidomyiidae), hieracium gall fly; Cheilosia urbana Meigen (formerly C. praecox (Zetterstedt 1843)) (Insecta: Diptera: Syrphidae), hieracium root-feeding hover fly; Cheilosia psilophthalma (Becker 1894) (Insecta: Diptera: Syrphidae), hieracium crown-feeding hover fly Characteristics: Macrolabis pilosellae (Binnie 1878). Adults are minute (less than 2 mm long), fragile flies, with tiny, poorly developed heads. The body of the fly is pale fawn in colour, the legs are darker, and the head appears dark, almost black. Antennae are relatively short, and multi-segmented. The head, consisting mostly of a pair of distinctive elongate, black, compound eyes that curl around the bases of the antennae, is tucked under the thorax. The dorsal surface is covered with golden brown hairs, and the delicate, hairy wings have reduced venation. Adult flies emerge from galls between June and early September in the Northern Hemisphere. Flies are short-lived, males, 2–13 days, females, 2–9 days. Female flies lay tiny, elongate, white eggs externally in leaf axils. Developing larvae cause gall formation in leaves at stolon tips, and in developing rosettes.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 5

Flies were collected from the Black Forest in Germany, where they undergo three generations per year. In New Zealand, climatic conditions in most hawkweed-infested areas are such that we predict a similar number of generations here. In the laboratory in Switzerland flies lived up to 9–13 days at 15°C and up to 5–7 days at 25°C. Female flies laid 26–50 eggs on average (Grosskopf & Hassler, 1998). Host range tests were completed on 69 test plant species comprising 9 species from the genus Hieracium, 24 species in the family Asteraceae, including the closest relatives to Hieracium among native New Zealand plants, and 36 species from an additional 27 families of plants. Oviposition and larval development tests were carried out in Switzerland, or under insect containment conditions in New Zealand for those plants that could not be grown in Switzerland. Galls developed only on plants in the subgenus Pilosella of the genus Hieracium. Characteristics: Cheilosia urbana Meigen (formerly C. praecox (Zetterstedt, 1843)) These are medium- sized (7–8 mm long) stout-bodied flies with large heads. In males the compound eyes meet on the top of the head. Females have eyes separated by about one-quarter of the width of the head. The body is dark coloured, and covered with pale-coloured hairs, which give it a furry appearance. Tarsal claws are distinctly bicoloured, dark at the base with brownish-yellow tips. Legs are generally brown in colour, with a variable amount of yellow. Larvae feed externally on the roots of the host plant, in the soil. Distinct setae are visible on the surface of larvae, which are generally dark in colour. The pupae are brown in colour, the respiratory horns at the anterior end of the pupae are the same colour as the rest of the pupa, and are cylindrical in shape. Flies were collected from the Black Forest in Germany and from the Swiss Jura. They are univoltine, completing only one generation per year. Field-collected female flies laid 70–84 eggs on average. (Grosskopf & Hassler, 1998; Grosskopf & Murphy, 1999). This may be an underestimate, as flies may have already laid some eggs before being collected. Behaviour in New Zealand is expected to be similar, as critical climatic factors are not dissimilar from the region of origin. Host range tests were completed on 71 test plant species comprising 9 species from the genus Hieracium, 31 species in the family Asteraceae, including the closest relatives to Hieracium among native New Zealand plants, and 31 species from an additional 25 families of plants. Oviposition and larval development tests were carried out in Switzerland, or under insect containment conditions in New Zealand for those plants that could not be grown in Switzerland. In addition an open field test was carried out in Switzerland. No eggs were laid on plants other than Hieracium species, and no larvae survived on any plants other than Hieracium species. Characteristics: Cheilosia psilophthalma (Becker 1894). Morphologically C. psilophthalma is difficult to distinguish from C. urbana. Most of the distinguishing characters are only visible under a stereo microscope. The tarsal claws do not have a yellowish tip– both the base and tip of the claws are dark, with the base only slightly lighter. Larvae feed on the above-ground parts of the host plant, at the rosette centre, at the base of stolons, in leaves, leaf axils, and in stolon tips. The larvae are generally lighter in colour than C. urbana larvae, and have several rows of very short bifurcated structures along the body, but do not have obvious setae like C. urbana. Respiratory horns on the pupae of C. psilophthalma are shorter than those of C. urbana, are tapered, and mid-brown in colour, somewhat darker than the rest of the pupa. Flies were collected from the Black Forest in Germany and from the Swiss Jura. They are univoltine, completing only one generation per year. Field-collected female flies laid 52–70 eggs on average (Grosskopf & Hassler, 1998; Grosskopf & Murphy, 1999). This may be an underestimate, as flies may

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 6

have already laid some eggs before being collected. Behaviour in New Zealand is expected to be similar, as critical climatic factors are not dissimilar from the region of origin. Host range tests were completed on 66 test plant species comprising 9 species from the genus Hieracium, 29 species in the family Asteraceae, including the closest relatives to Hieracium among native New Zealand plants, and 28 species from an additional 25 families of plants. Oviposition and larval development tests were carried out in Switzerland, or under insect containment conditions in New Zealand for those plants that could not be grown in Switzerland No eggs were laid on plants other than Hieracium species, and no larvae survived on any plants other than Hieracium species.

[ Yes ] further information [ No ] commercially sensitive information

Name of the organism that may be used for the Authority’s public register: Macrolabis pilosellae (Binnie 1878) (Insecta: Diptera: Cecidomyiidae), hieracium gall fly Cheilosia urbana Meigen (Insecta: Diptera: Syrphidae), hieracium root-feeding hover fly Cheilosia psilophthalma (Becker 1894) (Insecta: Diptera: Syrphidae), hieracium crown-feeding hover fly

6. If the organism is a genetically modified organism, information on the details of the genetic modifications: This information shall include full details of the genetic constructs and modifications and the source and characteristics of the foreign nucleic acid. This information should clearly identify the source of the donor genetic material and the characteristics. The desired characteristic (eg, herbicide resistance) and any other significant characteristics that may be expressed by the donor genetic material in the organism should be described. Information on the stability and homogeneity of the construct should be given, if known. If this information is not known then this should be explicitly stated. References to the scientific literature supporting this information should be given here if appropriate. Information that is commercially sensitive should be clearly identified. If supplied separately a cross-reference to it should be included. [ No ] further information [ No ] commercially sensitive information

Not applicable

7. The reason why an application is necessary for the organism: Refer to the definitions set out in Section 2 of the Act, to the prohibited organisms in the Second Schedule of the Act, and for genetically modified organisms, to the exemptions in the HSNO (Organisms Not Genetically Modified) Regulations 1998.

Macrolabis pilosellae, Cheilosia urbana, and C. psilophthalma are ‘new organisms’ as defined under section 2 of the HSNO Act 1996 by way of each of them being: (a) an organism belonging to a species that was not present in

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 7

New Zealand immediately before 29 July 1998, and therefore they require approval to be imported for release, or to be released from containment; M. pilosellae, C. urbana, and C. psilophthalma are not prohibited organisms under the Second Schedule of the HSNO Act 1996.

8. Information on the potential use of the organism: The information shall include all the potential uses of the organism and how well the organism performs these uses. The information should also provide sufficient details on the purpose of the application to enable the Authority to provide the information required in the register (under section 20(2)(c) of the Act). We suggest the information in this section be as expansive as possible. While the applicant may have only one potential use in mind, an approval may enable other uses as well. To enable the Authority to have access to all relevant information all the potential uses of the organism should be provided. The information on how well the organism performs these uses is necessary to enable the Authority to determine the performance characteristics of the organism. Information that is commercially sensitive should be clearly identified. If it is supplied separately a cross- reference to it should be included. [ Yes ] further information [ No ] commercially sensitive information

This application falls under Section 34 of the HSNO Act 1996: Application for approval to import or release. Permission is sought to release the three new organisms from containment (Section 34(1)(b)). Copies of MAF import permits Nos. 2000009543 and 200009607 are attached (Appendix V).

The sole purpose and use of the three new organisms, Macrolabis pilosellae, Cheilosia urbana, and C. psilophthalma is as biological control agents for weedy Hieracium species, to complement the activity of two agents already released for the same purpose: Oxyptilus pilosellae and Aulacidea subterminalis. The insects have no attributes that would make them useful for any other known purpose, nor do they have any negative qualities that would produce unwanted side effects.

Although it is not possible to predict with certainty the outcome of any specific agent introduction for biological control, an economic evaluation of biological control of Hieracium undertaken at the outset of this programme (Grundy, 1989) concluded that a biological control project would recover its costs about 14 years after the release of control agents. The five agents selected were those judged to have the greatest potential for damaging Hieracium, based on their impact on plants in Europe, while posing minimal threat to other plants. Surveys were carried out in 1992–94 in Europe, and a number of insects were found that attacked species of Hieracium (Jordan, 1993; Grosskopf, 1994; Sárospataki, 1999). Seed-feeding species such as Tephritis spp. were accorded low priority since Hieracium spp. are perennial weeds that are already widespread in New Zealand. Instead, insects that limited the growth and vegetative spread of the weeds were accorded high priority, and the final choice was made of five insects that would together act on a range of plant parts, and attack the range of problem Hieracium species present in New Zealand.

Flies from both families Cecidomyiidae (gall flies) and Syrphidae (hover flies) have been used previously as weed biological control agents. Cecidomyiids have been successful in controlling their target weeds: for example the gall fly Zeuxidiplosis giardi has effectively limited St John's wort in Hawaii, South Africa, and parts of New Zealand and Cystiphora schmidti is an effective control agent for skeleton weed in Australia (Julien & Griffiths, 1998). Only one syrphid (also a Cheilosia species, C. corydon, that was released in the early 1990s in the USA for control of Carduus species of thistles) has been used as a biological control agent, and it is too early to say yet whether it has

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 8

been effective. There are many examples where single control agents, or, more commonly, a suite of complementary agents, has effectively limited a problem weed (McFadyen, 1998). Establishment rates for introduced weed biological control agents in New Zealand are currently estimated at between 76 and 94% (Syrett et al. 2000). It is very difficult to predict whether any new agent in particular will establish or not, but given adequate effort, and the lack of obvious factors likely to prevent establishment, probability of establishment is around 90%. Experience indicates that univoltine species of biological control agents reach damaging populations at sites in New Zealand 5 years or so after their release there. The two hieracium- feeding Cheilosia species are therefore likely to take 5 years from time of release at a site to reach damaging populations. Our experience with multivoltine species is more limited, but gorse spider mite attained outbreak populations in the third season following release at most sites. Macrolabis pilosellae is multivoltine, so we predict that it may reach damaging populations within 3 years of release.

The rate of dispersal is dependent on fecundity, number of generations per year, and average distance travelled by individuals of the agent species. It will also be influenced by climatic conditions following release. Many univoltine species have taken at least 20 years to establish populations throughout New Zealand. Active re-distribution may be used to accelerate widespread establishment. For multivoline control agents such as gorse spider mite and old man’s beard leaf miner establishment occurred throughout weed-infested areas of New Zealand within 5 8 years and 3 5 years respectively (A.H. Gourlay, pers. comm.). Gall flies such as M. pilosellae are weak fliers, but are readily transported by wind. Hover flies (the two Cheilosia species) are strong fliers, but are perhaps less likely to be taken long-distance on the wind. From the known distributions of the three insects in Europe, we expect that they will establish throughout Hieracium-infested areas of New Zealand. The rate at which they will reach maximum distribution is difficult to predict, but we estimate that with active involvement of interested farmers and other land managers this will be achieved within 5 to 20 years. Comparing the climatic and altitudinal ranges occupied by Hieracium species in New Zealand with the conditions experienced by the insects in their native Europe, we do not expect them to be limited by adverse conditions. Exceptions might occur, for example, in particularly dry seasons that prevented growth of the host plants, and temporarily slowed their establishment and population growth.

Reference Syrett, P.; Briese, D.T.; Hoffmann, J.H. (2000). Success in biological control of terrestrial weeds by arthropods. Pp. 189 230 in Gurr, G.; Wratten, S. (eds.), Biological control: measures of success. Kluwer, Dordrecht, The Netherlands.

Provide in this box a statement describing the purpose for making the application. This statement may be included in the Authority’s public register (please use a maximum of 255 characters): To release from containment the insects Macrolabis pilosellae (Binnie 1878), Cheilosia urbana (Zetterstedt 1843), and Cheilosia psilophthalma (Becker 1894) for the purpose of biological control of hawkweeds, Hieracium spp.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 9

9. Information on any likely inseparable organisms: Information should be provided on any organism which is unable to be separated from any new organism at the time of making the application. Examples may include foot and mouth and scrapie causing organisms in animals and viruses in plants. [ Yes/No? ] further information

Macrolabis pilosellae, Cheilosia urbana, and C. psilophthalma are not known to host any inseparable organisms, or any organisms pathogenic to insects (CABI abstracts, 1984–1999).

The three Hieracium-feeding insects will be imported from Switzerland into secure containment in New Zealand in the MAF Regulatory Authority (MAFRA) approved facility at Lincoln. Cheilosia urbana flies will be imported as eggs or pupae, C. psilophthalma as pupae, and Macrolabis pilosellae in pupal cocoons. All three fly species will be reared through to adult under containment, so as to eliminate any parasitoids from the culture. Flies will then be tested for the presence of pathogens and, provided results are negative, application will be made to MAF Quarantine for permission to release from quarantine. Procedures will be as required by the MAFRA Standard 154.02.08: Standard for Invertebrate Quarantine Facilities and Landcare Research's operating procedures: Gourlay, H. & Hill, R. (1994) Procedures for the importation and quarantine of live insects and other arthropods into the Landcare Research Invertebrate Quarantine Facility. Pursuant to Section 40 of the Biosecurity Act 1993, Mr Hugh Gourlay is the registered operator of this facility. The facility also complies with the standards set out in the Third Schedule, Part II of the HSNO Act 1996. Operating under these regulations, the facility offers a very high level of security with very low levels of risk.

Risks associated with the importation into containment of Macrolabis pilosellae were assessed in ERMA application number 98001. The initial importations of Cheilosia urbana and C. psilophthalma were made under MAFRA, and risks identified were similar to those with M. pilosellae.

10. Information on the affinities of the organism with other organisms in New Zealand: Information should be provided on taxonomically related organisms in New Zealand with which the new organism is likely to interact. The information should also include any positive or negative effects of this interaction, including other relationships (eg, predator/prey/parasites, etc.).

Cecidomyiidae: Macrolabis pilosellae Cecidomyiids comprise a large cosmopolitan family of small to minute flies. Larvae of most of this family of flies live in galls of living plants or are scavengers in decomposing organic matter. There are at least 36 species in the family Cecidomyiidae, from 14 genera, recorded in the literature as present in New Zealand (Barnes, 1937; Scott, 1984). However, there are more than those published, and it is likely that the eventual number of species will be substantially greater. Dr Martin has compiled a database with a total of 48 species, only seven of which are described, and from one small-leaved Coprosma sp. five different gall-forming cecidomyiids have been recorded (N.A. Martin, Crop & Food Research, pers. comm.). Most of the genera present in New Zealand are of worldwide distribution, but the genus Macrolabis is not represented in New Zealand. At least six of the species present in New Zealand have been accidentally introduced, and one intentionally (Zeuxidiplosis giardi Kieffer), as a weed biological control agent. Apple leaf-curling midge (Dasyneura mali Kieffer) and hessian fly Mayetiola destructor (Say) are examples of accidentally introduced cecidomyiids that have achieved pest status here (Scott, 1984). Parasitoids from six families within the Hymenoptera have been recorded from species of Cecidomyiidae in New

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 10

Zealand (Valentine & Walker, 1991). Species of cecidomyiids that have been used as weed biological control agents elsewhere have sometimes been subjected to high levels of attack from parasitoids. Although the introduced Z. giardi is attacked by a torymid (Torymoides sp.) parasitoid in New Zealand, plants are heavily galled, and stunted, by the control agent (Syrett, 1989). Evidence does not suggest that the impact of M. pilosellae will be significantly lessened by parasitism or predation.

Syrphidae: Cheilosia urbana, C. psilophthalma The genus Cheilosia contains about 130 Palaearctic species, but neither the genus, nor the tribe Cheilosiini to which it belongs, are represented in New Zealand. Larvae of the Syrphidae, or hover flies, are often predaceous. Those of the subfamily Eristalinae, to which Cheilosia belongs, may also be saprophagous or phytophagous. But there are no close relatives in New Zealand. The only syrphids of economic importance in New Zealand are the bulb flies, Merodon equestris (F.) and Eumerus strigatus (Fallén), which are both introduced species, and the native aphid-feeding Melangyna novaezelanidae (Macquart) and Melanostoma fasciatum (Macquart). Parasitoids of the latter have not been observed (N.A. Martin, pers. comm.) although Valentine & Walker (1991) note that Diplazon laetatorius F. (Ichneumonidae) has been recorded from native species of Syrphidae. Adult syrphids can be useful pollinators.

Potential impacts on other organisms in the environment As tests have demonstrated a high degree of host specificity for all three insects (see below, and in Appendices I III) the likelihood that any non-target species will be negatively affected is very low. Native plant species, and useful exotic plant species, are likely to benefit positively from the suppression of hawkweeds. Very few insects are found on hawkweeds in New Zealand, so replacing the weeds with other plants is likely to increase faunal as well as floral diversity. The genus Hieracium comprises 4 subgenera: Pilosella is strictly European, Hieracium has a Holarctic distribution, Stenotheca is of ‘New World’ origin, and species of the sub-genus Mandonia occur at high elevation in South America. Two sub-genera (Pilosella and Hieracium) are represented in New Zealand by weedy species introduced from Europe.

The gall fly Macrolabis pilosellae has been recorded in the field only from plants within the sub-genus Pilosella. The field host ranges of the hover flies Cheilosia urbana and C. psilophthalma in Europe are not well known, but they are readily collected from H. pilosella there. Host range tests indicated that the two hover flies attacked Hieracium plants in both sub-genera Pilosella and Hieracium, although showing a preference for plants within the subgenus Pilosella. In summary from results of host range tests, M. pilosellae is likely to be a useful control agent for H. pilosella, H. praealtum and H. caespitosum, but is unlikely to attack H. lepidulum. The root-feeding hover fly C. urbana is likely to be useful for control of all four problem Hieracium species in New Zealand, including H. lepidulum. The crown-feeding hover fly C. psilophthalma showed preference for species within the sub-genus Pilosella, but some development occurred on H. lepidulum as well indicating that this species may also be attacked in the field.

Two insects and a pathogen have already been released for control of H. pilosella. The pathogen Puccinia hieracii var. piloselloidarum attacks only some strains of H. pilosella, and so far does not seem to be very damaging in New Zealand. The gall wasp, Aulacidea subterminalis, has been recovered from H. pilosella in the field, on plants where it was released. However, it is too early to say whether it will be effective, or whether it will attack any Hieracium species other than H. pilosella. Host range tests indicated that this was unlikely, and that the gall wasp would only be useful against one weedy species, H. pilosella. The crown-feeding moth, Oxyptilus pilosellae, has been released in the field, but so far, no recoveries have been made. The moth is difficult to rear, so only small numbers

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 11

have been released so far. We hope to improve rearing methods, and make further releases during 2001/02. This moth is likely to have a broader host range than either of the other agents already released, and may be a useful control agent for all four weedy Hieracium species. However, it did show a preference for H. pilosella and H. caespitosum in host range tests, so may be less effective on the other two species.

High populations of phytophagous insects have been known to cause outbreaks of parasitoid or predatory species. It has been hypothesised that such abnormally high populations may have detrimental impacts on the normal prey species of the parasitoids or predators. It is not known whether native cecidomyiids or syrphids will be positively or negatively affected by a possible increase in their parasitoids or predators, but there is no evidence to suggest that the impacts are likely to be significant.

Note: If this application is not for a rapid assessment, go to Part C.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 12

PART C

Assessment of Effects The information to be provided in these sections should cover the assessment of effects (both adverse and positive) of the organism. Where appropriate these sections may be combined in section 14 below. Effects should be clearly assessed where relevant, including details as to how the risks will be controlled by the proposed containment system. Where these adverse effects are identified, in the first instance by the applicant, as being minor then these do not require in-depth assessment.

11. Information on all the possible adverse effects of the organism on the environment: This should include information on the effects of the organism on ecosystems, public health, and Maori culture and taonga. It should also include information relevant to the matters in sections 4, 5, 6, 7, 8, and 37 of the Act and any regulations made under section 41 of the Act. The assessment should identify and assess risks, costs and benefits. The information should give particular regard to: Environmental and ecosystem effects (section 6(a) and (b) of the Act) - assessment of the known and possible adverse effects throughout the life cycle of the organism on the sustainability of native and valued introduced flora and fauna and on the intrinsic value of ecosystems. [Include an assessment of the ability of the organism to establish an undesirable self-sustaining population and the ease with which the organism could be eradicated if it was established.] Public health effects (section 6(c) of the Act) - assessment of the known and possible adverse effects throughout the life cycle of the organism on public health. [Assessment should take account of aspects of public health and safety including, where appropriate, effects from occupational exposure and effects from environmental exposure to the organism.] Relationship of Maori with taonga (section 6(d) of the Act) - assessment of the known and possible adverse effects throughout the life cycle of the organism on the relationship of Maori and their culture and traditions with their ancestral lands, water, sites, wahi tapu, valued flora and fauna, and other taonga. [Include details of consultation (if any) carried out.] The ability of the organism to escape from containment.

See Section 14

12. In the identification and assessment of risks, costs and benefits and other impacts which may occur as a result of the escape of the organism include those matters set out below. The information should comprise of the risks identified and include: - the nature of the adverse effects of the organism. - the probability of occurrence and the magnitude of each adverse effect. - the risk assessed as a combination of the magnitude of the adverse effect and the probability of its occurrence. - the options and proposals for managing the risks identified.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 13

- the uncertainty bounds on the information contained in the assessment, expressed quantitatively where possible but otherwise through narrative statements. The identification and assessment of costs and benefits required in each application must include. - the nature of the costs and benefits associated with the proposed new organism and whether they are monetary or non-monetary; - the magnitude or expected value of the costs and benefits and the uncertainty bounds on the expected value. Relevant costs and benefits will be those which pertain to the New Zealand economy, society and environment and which would not arise if the application was not approved (ie the opportunity cost to New Zealand). They shall include the long term as well as short term, and consequential as well as direct costs and benefits. The information on risks, costs and benefits shall include the distributional effects over time, space and groups in the community. It shall also include the uncertainty intervals associated with these estimates.

Not applicable

13. Information on the positive effects of the organism:

See Section 14

14. Assessment of effects If the assessment of effects is combined into this section, applicants should clearly indicate how the information requirements in sections 11, 12 and 13 of this form are addressed. [ Yes/No? ] further information [ Yes/No? ] commercially sensitive information

Section 36 of the HSNO Act 1996 Minimum standards will be met by this application.

(a) The three new organisms are not likely to ‘cause any significant displacement of any native species within its natural habitat’. These three new insects are dependent on introduced Hieracium for their survival, so although they are likely to build to high populations in the presence of substantial infestations of Hieracium, in areas where there is little Hieracium they will not be prominent, and will not have any impact in areas where their host plants are absent. In areas where Hieracium is abundant, native herbivores are uncommon, as surveys for insects on the weed have shown (Syrett & Smith, 1998). Therefore the impact on native fauna (and flora, since weedy Hieracium displaces native plant species) will be enhanced by successful establishment of effective biological control agents that suppress Hieracium. High populations of Hieracium-feeding insects could conceivably result in an increase in generalist predator populations, which might then disperse to feed on populations of native species. However, this is a hypothetical scenario, which in any case would be only a transient effect.

(b) These introductions are unlikely to ‘cause any significant deterioration of native habitats’. In high quality native habitats, Hieracium is not present, and therefore these new insects are not likely to occur there. In lower quality habitats, where Hieracium species are present, the insects will improve the habitat quality by reducing Hieracium cover, and allowing the re-establishment of native plant species with their attendant

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 14

fauna. There may be a temporary increase in bare ground and litter, which might have the detrimental effect of allowing increased erosion. However, our trials to simulate biological suppression of Hieracium are showing that other vegetation does increase to replace the weed (see section on risk of increased erosion and decline in water quality under ‘environmental and ecosystem effects’).

(c) None of the organisms will ‘cause any significant adverse effect on human health or safety’ as they have no undesirable characteristics, and their feeding, and any consequent damage, is confined to their host plants. In addition, areas infested by Hieracium are not adjacent to major areas of human habitation or occupation.

(d) None of the three insects is likely to cause any significant adverse effects to New Zealand’s inherent genetic diversity. Although the successful establishment of these insects will increase the number of exotic insect species in New Zealand by three, they should not result in any loss of native diversity. The successful suppression of Hieracium species will enhance native diversity by increasing the area of habitat available to native species, and removing the threat to several vulnerable plant species (see section on threats to indigenous biodiversity, under ‘environmental and ecosytem effects’.

(e) The three new organisms are unable to ‘cause disease, be parasitic, or become a vector of human, animal or plant disease’. None of them, or their close relatives, has ever been recorded as a disease organism, a vector of disease, or to be parasitic. All three species form distinct species, and do not interbreed with other organisms, so they are highly unlikely to change their current behaviours.

The probability and magnitude of the potential adverse effects are summarised in Table1, and of the benefits, in Table 2.

Economic, social and cultural effects

Economic benefits The economic benefits from successful biological control of Hieracium are high. Grundy (1989) conducted an economic evaluation in which an estimate of the value of production losses caused by Hieracium was up to $4.4M annually in the moderate rainfall, high country areas alone. A more recent estimate placed up to million stock units at risk (McMillan, 1991), representing a loss of $76M per annum (last 5-year to 1999 average returns, Meat and Wool Economic Service of New Zealand, 2000). Hieracium species were dominant to common over 42% of the South Island in 1990 (Hunter, 1991). These weeds are still spreading (Colin Meurk (Mackenzie Country), Alan Rose (Marlborough and Harper/Avoca, Rose et al., 1998), Brian Molloy (Canterbury), Peter Espie (Otago), and many people believe that if nothing is done to curb them, even greater losses will be sustained in the future. Non- market values, such as the threat to valued native plants, the loss of scenic values, and the loss of soil stability and water-holding capacity were not taken into account in these estimates. A study by Greer & Sheppard (1990) of the benefit of research into biological control of old man's beard showed that New Zealanders placed a very high value on preserving native forest. If tussock landscapes were valued at only one-tenth that of forest, the perceived value of research for conservation of natural vegetation would exceed the economic valuation obtained by Grundy (1989). There is reputed to be a small industry that harvests Hieracium seed for export that is worth about $10k annually. This industry would be severely affected by successful biological control. However, those farmers who currently harvest seed are likely to gain larger benefits through successful biological control than their current returns from harvesting seed.

Alternative methods of managing Hieracium In certain situations, chemicals are available for management of Hieracium. Recommendations were obtained by John Aspinall from Mark Christie (Du Pont) and John Henderson (Dow).

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 15

Herbicide treatment: 2,4-D Ethylheyl ester (Pasturekleen, 520g/l a.i.) + clopyalid (Versatill, 300g/l a.i.) + dimethicone copolyol (Boost, an organo-silicone surfactant) The cost of this treatment is $73.75/ha + $25/ha for ground application or $100/ha for helicopter application. In trials the treatment gave 91% control of Hieracium after 400 days. A disadvantage of this treatment, even though it is effective, is that Versatill also removes clovers for at least a year. To maintain this level of control over the longer term, it is essential to apply fertiliser. Without Versatill 90% control was achieved at 150 days, but only 65% by 400 days, a rapid return of Hieracium.

Fertiliser treatment: A product called ‘Biostart’ is a seaweed-based liquid fertiliser with added finely ground elemental sulphur (S) plus Bo, Mn, and Na that has been publicised as a useful treatment for Hieracium. The finely ground sulphur results in a rapid initial response of strong clover growth, but it is recommended that the treatment is repeated every second year to maintain effectiveness. The estimated cost of treatment is $120/ha, plus application costs. A similar response is likely to be obtained by applying superphosphate, although it would not be as rapid.. For equivalent expenditure, superphosphate could be applied at 450kg/ha.

None of these treatments is cost-effective on most Hieracium-infested land. They are only useful for land that is potentially highly productive.

Predicting the safety of biological control agents from host-range tests A list of test plant species for each of the three insect species was compiled using criteria established by Zwölfer & Harris (1971) and Wapshere (1974, 1975), and revised by Forno & Heard (1997). Total numbers of species tested were 69 for Macrolabis pilosellae, 71 for Cheilosia urbana, and 66 for C. psilophthalma. Particular emphasis was given to taxonomically closely related species, i.e. representatives of the family Asteraceae, in particular the subfamily Cichorioideae, tribe Lactuceae, and genus Hieracium to which the host plants belong. In no-choice tests with M. pilosellae, gall development occurred only on plants in the genus Hieracium, subgenus Pilosella (Appendix I). Adult flies were produced from only three species: H. pilosella, H. caespitosum, and H. praealtum. In single choice oviposition and larval development tests larvae and adult flies were produced from H. pilosella, H. caespitosum and H. praealtum, but not from H. aurantiacum or H. lepidulum. It is concluded that M. pilosellae has potential as a control agent for the three problem species H. pilosella, H. praeatum, and H. caespitosum. In no- choice larval development tests with Cheilosia urbana larvae were found on all Hieracium species tested except H. murorum, and no plants outside the genus Hieracium were attacked (Appendix II). In single-choice oviposition tests eggs were laid on all Hieracium species, but not on either of the other two plant species tested. Results indicate that C. urbana may be a useful control agent for all four problem species. In no-choice larval development tests with C. psilophthalma larvae were found on all Hieracium species tested, and on all Hieracium species a proportion of larvae developed to adult (Appendix III). Plants outside this genus were not attacked: no larvae, pupae, or feeding marks were found on any plants other than Hieracium. Conclusions were that C. psilophthalma is restricted to Hieracium species, but shows a preference for stoloniferous species. However, small numbers of eggs were laid on H. lepidulum, and larvae were able to develop to adult, indicating that C. psilophthalma might be a possible control agent for this non-stoloniferous problem hawkweed.

In general, if results of host tests are incorrectly interpreted, so that when insects are released in the field they attack non-target plants, economic damage to valued plant species may occur. However, the likelihood of this happening is very low. Over 1150 releases of more than 365 species of insects and pathogens for control of 133 weeds in 75 countries have been made (Julien & Griffiths, 1998), and where Wapshere's (1975) protocols have been applied, there has been negligible impact on non-target plants (McFadyen, 1998).

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 16

Predicting the success of biological control It is difficult to predict the likely success of a specific biological control programme, and hence the probability of achieving a high economic benefit. Some previous projects have been spectacularly successful, and others have had a negligible impact. The proportion of projects that have achieved successful control have been calculated as 83% for South Africa, 50% for Hawaii (McFadyen, 1998), and, using similar criteria, 83% for New Zealand (Fowler et al., in press). The New Zealand figure included projects classified as ‘full’ or ‘partially successful’. For one project only (St John’s wort), no other control methods are necessary, but five projects were judged as partially successful with other controls still needed. One project failed, but the majority (7 projects) were at too early a stage for an assessment to be made. It is rare for a single control agent to be completely successful alone, although often one species may have the greatest impact. In the case of a suite of related weeds, such as the Carduus group of thistles, a single agent has often been successful against more than one weed species (Julien & Griffiths, 1998). Analyses have been conducted to determine whether some kinds of agents make better control agents than others (Crawley, 1989, Julien, 1989), but generalisations have not been very useful: for example, chrysomelid beetles are among both the best and worst biological control agents.

In the case of Hieracium, two insects have already been released as biological control agents, and an accidentally introduced rust fungus is established. It has been predicted that a plume moth and a gall wasp will contribute to the control of some, but not all, of the problem Hieracium species (Syrett et al., 2000). Host range tests (see Appendices I–III) indicate that the three insect species that are the subject of the current application will complement the activity of the two already released by attacking a different range of Hieracium species (Table 3). It is difficult to predict efficacy based on similar organisms used previously for weed biological control because the only other syrphid to have been used previously (Cheilosia corydon for Carduus thistles) is not yet established, so there is no information available on its impact. Other cecidomyiid flies are successful biological control agents, however (see p.8)

In Australia three control agents, a gall fly, a gall mite and a rust fungus, have effectively controlled another weed, rush skeleton weed, Chondrilla juncea, in the same family (Asteraceae) as Hieracium. Four insect control agents (two species of ragwort flea beetle, cinnabar moth and a leaf-rolling caterpillar) have contributed to effective control of another asteraceous weed ragwort (Senecio jacobaea) in a number of countries (Julien & Griffiths, 1998).

Complementarity of control agents The suite of insects proposed as potential control agents for the four weedy species of Hieracium fill different feeding niches within their host plants as well as attacking a different range of Hieracium species. The five insects selected for introduction to New Zealand as biological control agents were chosen from about 40 species that are reportedly host specific to Hieracium species (Jordan, 1993). Of the two insects already released in New Zealand, caterpillars of the plume moth feed on all the above-ground parts of the plants, weakening them, and occasionally causing the death of rosettes. Immature stages of the gall wasp feed and develop inside galls that are formed at the end of stolons. These galls inhibit further stolon elongation, and hence vegetative reproduction of the plants. The rust fungus develops on all of the above-ground parts of the plant and, if the infestation is severe, can kill rosette plants. However, the rust attacks only some forms of H. pilosella, and none of the other weedy Hieracium species. The three new insects also attack their host plants in different ways. Larvae of the gall-fly Macrolabis pilosellae develop inside galls, which they induce on leaves of the host plants. Unlike any of the other insect control agents, M. pilosellae is mulitvoltine (passes through several generations in a single year), so is likely to exert pressure on plants when other insect species (all univoltine) are inactive. Larvae of the syrphid Cheilosia urbana feed externally on roots, while C. psilophthalma larvae feed in the crown of the plant. The cumulative impact of the five insects is likely to be greater than any one alone.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 17

It has been suggested that biological control agents for a target weed should be introduced one at a time, and monitored for impact, so that if adequate control has been achieved, further introductions may be unnecessary. In practice this is not a viable strategy because most insect biological control agents take a long time to achieve widespread establishment and impact. During this time (usually in excess of 10 years) the weed is not being controlled, and exerts increasing levels of economic and environmental damage. For early biological control projects, all agents were introduced that were believed to be sufficiently host specific that risks of impact on non- target plants were minimal. For Lantana camara 36 different species were released (Julien & Griffiths, 1998). Today a more parsimonious approach is taken (McEvoy 1996), and only those agents predicted to have the greatest impact are selected for introduction. So for control of purple loosestrife in the USA, six out of 14 species were selected as the most promising, and four of these were eventually released (Julien & Griffiths, 1998). It is too early to say how successful the project has been although insects are causing damage to plants in the field.

Parasitism and predation In Europe, several Cheilosia urbana larvae have been found parasitised by an undescribed Phygadeuon species (Ichneumonidae), but no parasitoids have been reared from C. psilophthalma. Parasitised larvae of Macrolabis pilosellae have been dissected from field-collected galls, and the rate of parasitism varied from zero to 80%. No information is available about predation on the three species of Hieracium-feeding insects. It is possible that native New Zealand parasitoids may transfer to these insects as new hosts, but little is known about likely candidates. As the larval stages feed largely inside the host plant, and adult flies are highly mobile, predation is unlikely to be an important factor.

Predicted negative impacts Adult syrphid flies feed on pollen, and two scenarios have been suggested that could be detrimental. Bees depend on pollen for food and, if large numbers of syrphid larvae were to build up on Hieracium so that large numbers of adults were produced, it is conceivable that competition for pollen could occur. However, Hieracium-feeding syrphids are present in the adult stage during summer when pollen is plentiful, and it is also highly unlikely that Hieracium-feeding syrphid populations would become dominant pollen feeders. Therefore, bee-keepers are unlikely to be affected. If adults of Hieracium-feeding syrphids did become dominant pollen feeders, another possible scenario is that weed populations might increase as a result of increased levels of successful pollination. However, there is no evidence that reproduction of weed species is limited by pollinators, nor that Hieracium- feeding syrphids would significantly add to the pool of available pollinators. There have been instances where swarms of flies (Cecidomyiidae) have attained sufficiently high populations to achieve nuisance value. Swarms of Hieracium-feeding flies have not been observed in Europe, and if swarming did occur in heavily Hieracium-infested areas, it would be a transient effect, as flies are short-lived, and unlikely to be a nuisance as Hieracium occurs in areas that have only low populations of people.

Consequences of successful control of Hieracium species A fear that is sometimes expressed by critics of biological control is that when a weed has been successfully controlled it may be replaced by another weed that is even more damaging, and in some cases control of one weed has resulted in its replacement by another (McEvoy & Coombs, 1999). Manipulation studies in vegetation have repeatedly shown, however, that elimination of one component species results in an increase in species already present to fill the gaps (Likens, 1985). Nevertheless, Groves (1991) has emphasised the importance of promoting competing native plants in conjunction with biological control programmes in natural ecosystems. In the case of H. pilosella, which is infesting hill and high country pasture in New Zealand, replacement of Hieracium species by a worse weed could pose an economic threat to pastoral farmers. However, some farms are so badly affected that their land has no longer any value for pasture production. In this case it is difficult to imagine how any other weed could make the situation worse. Preliminary results from a trial in which biological control of Hieracium is being simulated suggest that a mixture of native and exotic plants is replacing the weed (Syrett et al., 2000).

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 18

Tussock landscapes have cultural value for the vistas they provide for travellers, tourists, and residents of the hill and high country. Hieracium species are continuing to spread through these areas, with the associated reduction in native tussock (Festuca and Chionochloa species) and smaller inter-tussock herb species. Biological control has the potential to reverse this transition.

Environmental and ecosystem effects

Risk of increased erosion and decline in water quality As tussock-covered hillsides become predominantly Hieracium covered, rate of water run-off increases, with subsequent increases in erosion, and decline in water quality. It has been argued that with the successful establishment of biological control agents, Hieracium species will decline, and there will be an increase in bare ground that will result in worse erosion and water quality problems. Experiments being undertaken to simulate the action of biological control agents are showing that although there have been initial increases in bare ground and litter in treatment plots, vegetation cover increased in treated plots compared to the controls (Syrett et al., 2000). Suppression of weedy species by biological control agents is a gradual process compared to suppression using herbicides, when risks of erosion are substantially greater. The risk of negative impacts of successional change initiated by biological control of Hieracium species is probably low, and indications from the simulation experiments in progress are that changes in these environments are very slow. There is no evidence from these experiments that Hieracium species are being replaced by other weed species. The restoration of native grassland species should be associated with an increase in native biodiversity of both the flora and fauna of affected habitats, and restoration of the ecosystem to a more natural state. Where H. lepidulum is invading forest understorey, it is expected that it will be replaced by native species, as there are few other exotic species established in these habitats.

Risk of impacts on non-target species In the event that the host specificity tests have been incorrectly interpreted, or that they fail, and the newly introduced insects feed on desirable non-target species, there could be a risk of suppression, or even extinction, of rare and endangered species. The number of cases of non-target plants being attacked by biological control agents is small (McFadyen, 1998), but the risk of damage is significant when potential control agents with a relatively broad host range are used (McEvoy & Coombs, 1999). For example, the caterpillar Cactoblastis cactorum for control of cactus weeds has also attacked rare native cactus species in North America.

Representatives of cultivated species used in our host tests included plants within the subfamily Cichorioideae to which Hieracium spp. belong, as well as native New Zealand plants from the other subfamily of the Asteraceae, the Asteroideae. Test plants included the crop plants chicory and lettuce, and endemic species Embergeria grandifolia, Sonchus kirkii, and Kirkianella novae-zelandiae. Results of host tests (insect oviposition, and development, see Appendices I–III) indicate that all three insect species being considered for biological control of Hieracium species in New Zealand are host specific, in that they are confined to the genus Hieracium, and show preference for the subgenus Pilosella. Not all Hieracium species were suitable hosts, and no plants outside the genus Hieracium were attacked. There are no native Hieracium species in New Zealand, nor any cultivated exotic species. Therefore the risk of unwanted non-target impacts occurring is very low.

Threat to indigenous biodiversity Indigenous biodiversity may be threatened if an introduced biological control agent competes directly with native species. In the case of the three Hieracium-feeding insects, this is unlikely to be significant because the existing fauna of Hieracium in New Zealand is meagre (Syrett & Smith, 1998), and native species found on the plants are more common on alternative host plants. By reducing the dominance of Hieracium in affected areas, the diversity

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 19

of both flora and fauna should be enhanced. Keesing (1993) showed that, in Tongariro National Park, most native communities invaded by heather had lower diversity of invertebrates than those that had not been invaded, and none showed an increase.

Some ecologists fear that the continual release of biological control agents will have an adverse effect on indigenous biodiversity through a dilution effect. This fear is unfounded because the main source of additions to the fauna comes from unscreened, accidental introductions. Currently insect biological control agents comprise less than 1% of the exotic insect fauna, which is about 13% of the total insect fauna (Emberson, in press). However, the naturalised flora is now greater in number of species than the indigenous flora, and continues to grow. The main causes of homogenisation of the New Zealand flora and fauna are the purposeful introduction of plants for horticulture (which has resulted in 22 000 species now in cultivation, with the potential to escape, that have had minimal screening for weediness), and the accidental introduction of insects and other invertebrates.

Risk to ecosystem function Another way in which an introduced plant-feeding insect might disrupt existing food webs is if its populations increased to the extent that it became an important alternative food source for parasitoids or predators, that might in turn outbreak, and have an abnormally high impact on their normal native hosts. However, the introduced insects could just as easily enhance the populations of rare parasitoids and predators.

Accidental release of associated organisms

The risk of accidental release of parasitoids or disease organisms associated with these three Hieracium-feeding insects is low to negligible (see Section 9 for details). However, the consequence of an accidental release would most likely impact on the efficacy of the insects as biological control agents. If limited by disease or parasitism, they are likely to be less effective. Other consequences might result if the released organisms were not host specific, and were able to develop on alternative hosts. Since the host range of associated organisms is not investigated, it is not possible to predict the outcome of these scenarios, which have very low probability of occurrence.

Proposed release and monitoring programme

A research objective has been submitted to the Hieracium Control Trust to devise a protocol to monitor the establishment and impact of biological control agents for Hieracium. Percentage cover of Hieracium and other components of vegetation will be recorded at each of six sites. These data will provide baseline information on vegetation composition before insects are released. The incidence of the rust fungus will be scored at each of the six sites. Insects will be released at each of the sites as they become available, but for logistical reasons it is likely that they will be added over a period of 1–3 years. A schedule of optimal dates for sampling insect establishment will be constructed. A sampling strategy to measure the rate of spread of insects from the original release point will be devised. Numbers and pattern of release will be determined from experience with similar species that have been used elsewhere, but recent research shows that relatively small numbers of insects are required for establishment (Memmott et al., 1996).

A rearing method has already been developed for M. pilosellae and, as this fly is multivoltine, it is relatively straightforward to adjust the life cycle of Northern Hemisphere populations to Southern Hemisphere conditions. A colony of M. pilosellae from CABI Bioscience in Switzerland is currently held in insect containment at Lincoln, so that field releases may be made in February 2001. Five releases of approximately 300 flies each would be made in 2001, and a further 20+ releases are anticipated the following year (summer 2001/02). The syrphids are much more difficult to rear: they have not yet been reared successfully under artificial conditions. Therefore, we plan to import late instar larvae or pupae in July 2001, and release emerging adult flies after the colony has been tested for

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 20

the presence of pathogens, and parasitoids have been eliminated. Both C. urbana and C. psilophthalma are univoltine, so rephasing to New Zealand seasons is more difficult. The length of the pupal stage will be artificially extended by varying amounts, and glasshouse-grown Hieracium plants will be advanced so that they are at an acceptable stage for adult flies. A small number of trial field releases may be made during the summer of 2001/02, but the bulk of field releases will be made in summer 2002/03. These should comprise at least 10 pairs of adult flies at each of 20 release sites.

In 2000/01 costs of shipments from Switzerland and the rearing, release and monitoring programme are estimated to be approximately $65k, in 2001/02 $90k, and in 2002/03 $70k. Resources necessary for effective monitoring are expected to increase in future years as biological control agents become established and have measurable impact on host plant populations. The total future costs for the entire Hieracium biological control programme are estimated at approximately $500k.

For most weed biological control agents it is several years before establishment is confirmed through finding increasing numbers of the agents at or near the original release site. However, galls of the hieracium gall wasp, Aulacidea subterminalis have already been recovered in the field at two out of six sites checked only one year after release. On average, it takes about 5 years for control agents to attain sufficiently high populations at a site for it to be practicable to harvest insects for transfer to new sites. Dispersal rates can vary from a few metres to tens of kilometres per year. It is likely to take at least 10 years for these three flies to reach all populations of Hieracium throughout New Zealand, and from 5–10 years for populations in the area around a release site to have achieved maximum impact. From the information currently available on the three flies, it appears likely that they will establish on Hieracium species throughout their range in New Zealand.

Public health effects

None of the three flies proposed for introduction sting, bite, have an offensive odour, nor any other undesirable characteristics. They are similar in form and behaviour to related indigenous species already present in New Zealand. There are no human health risks associated with flies from the family Cecidomyiidae, nor from plant- feeding Syrphidae. Some predaceous Syrphidae can cause intestinal myaisis if accidentally ingested, when larvae are feeding on aphids on salad vegetables, for example (Lane & Crosskey, 1993). All three flies proposed for introduction have been bred in captivity and handled by research workers for several years, and no ill effects have been observed. All three insect species form distinct species that do not interbreed with any other species (Chandler, 1998).

Cost benefit analysis

Hunter (1991) reported Hieracium as dominant, conspicuous, or common over an area that included 42% of the South Island in 1990. The weeds are also well established in the Central North Island, at least as far north as Taupo, and east into Hawkes Bay. Observations from farmers and conservationists record increased distribution and density since this survey. McMillan (1991) estimated 1M stock units at risk from invasion of Hieracium. To date 300 000 to 500 000 stock units have been removed, largely breeding ewes and wethers. Farmers have reduced stock numbers, or intensified development in more productive areas rather than carrying stock of reduced productivity. In many areas carrying capacity has decreased from 2 stock units/ha to well below 1 stock unit/ha. In areas with the most severe Hieracium infestation carrying capacity has been reduced to 0.1 stock units/ha.

Herbicides are rarely a practical management option for Hieracium control. Also, there are few, if any, effective herbicides. They are expensive, non-selective, and not appropriate for use in conservation areas where valued non-target species are at risk. In a few areas, cultivating, oversowing, and topdressing with fertiliser may be

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 21

options, where potential stocking rates exceed 5 6 stock units /ha. The vast majority of Hieracium-infested areas (see figure 1, Appendix IV) are either low-grade pasture or conservation areas, where the costs of any existing management options make them prohibitive.

The average overseas earnings over the last 5 years from meat and wool were $76.18 per stock unit (Meat and Wool Economic Service of New Zealand, pers. comm., 2000). So between $23M and $38M is already being lost annually in pastoral export earnings alone, and a further $38M is at risk from further invasion of pastoral lands by Hieracium.

Potential savings from successful biological control are therefore up to $23 76M per annum. On average New Zealand biological control programmes have been partially effective, so predicted savings will be substantially less than this. In calculating an expected annual benefit of $205 000 15 years after release of control agents, Grundy (1989) conservatively assumed the costs of Hieracium to be $1.1 4.4M per annum. Adjusting for the current higher potential savings that are being predicted, the expected annual benefit becomes $3.7M.

There are also substantial non-monetary benefits of successful biological control of Hieracium. Water-holding capacity of soil under tussock is superior to that under Hieracium. Increasing shade and shelter results in lower soil temperatures, with lower evaporation losses. Soil organic matter is higher under tussock, also improving water retention ability (B. Wills, pers. comm., 2000). Other benefits result from reducing the threat to indigenous plants (B. Molloy, pers. comm, 2000), and increasing indigenous biodiversity of both flora and fauna. A summary of expected benefits and costs is presented in Table 4.

We estimate that approximately 80% of the 2.5M hectares of pastoral lease land is affected by Hieracium invasion, up to1.5M hectares of which is grazed. An additional 1M hectares approximately of hill and dryland country, that is also largely grazed, is moderately to seriously affected by Hieracium invasion. A further 1M hectares of productive land is probably at risk of invasion (John Aspinall, pers. comm). In addition to the farmed land that is affected, large areas of recreational and conservation land are infested and threatened by further invasion. Alpine grassland areas at higher altitudes, that until recently had been free of Hieracium, are increasingly being invaded.

A decline in abundance of Hieracium in grassland would increase the options for use of this land. In appropriate areas, stocking rates would increase as availability of palatable pasture species increased. A sustainable stocking rate might be between 1.5 and 2 stock units/ha on average. The ultimate stocking rate will of course also be influenced by changes in other factors such as management of rabbit and hare numbers, and climate. Hill and high country areas are increasingly being used for a variety of tourism activities, both by private individuals and groups, and through commercial ventures. Tourists prefer to undertake activities in areas not invaded by Hieracium, and as they become more highly aware of environmental values, they will increasingly seek out those areas where indigenous species predominate. Unless Hieracium is reduced, the degraded nature of many areas already infested, or susceptible to invasion, will make them inceasingly unattractive to tourists.

Forestry has been identified as possibly the only option for some areas already severely degraded. However, this activity is only viable on land at lower altitudes. Even there, trees grow slowly, which combined with the high cost of freight, makes forestry a marginal operation economically.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 22

Cost-benefit ratio

Assuming that there will be no benefit from biological control until year 10, and that between years 11 and 21 benefits increase to $3.7M at present day value and is constant thereafter, using a 10% discount rate, Present value of monetary benefits = $6.7M Present value of monetary costs = $774k B : C ratio is 8.6 : 1

Relationship of Maori with taonga

To assess the impact of this application on Maori taonga the following material was compiled using the ERMA Revised Protocol 1, Taking Account of Maori Perspectives (September 1999). Assistance with the consultation process was obtained from Bevan Tipene Matua (Senior Policy Advisor (Maori), ERMA New Zealand).

Initially we identified that the iwi most affected by the proposal to introduce new biological control agents for Hieracium were those whose lands were infested by the weed, namely Ngai Tahu and Tuwharetoa. Dialogues with these two iwi were initiated between November 1999 and January 2000 (Appendix VI).

Ngai Tahu

A written response from Linda Constable, on behalf of Te Runanga o Ngai Tahu .was received in early August 2000. In response to this letter, John Aspinall visited with Linda Constable and Bob Penter on 29 August and discussed the following issues with them:

The threats posed by Hieracium to Ngai Tahu assets and taonga Concerns regarding the insects switching to other plant Evaluation of present biological control agents Concerns relating to dilution of indigenous insects Other options for control of Hieracium including chemical herbicides, fertiliser and ‘biostart’ liquid fertiliser The wide range of groups funding the Hieracium Control Trust

On 31 August 2000 Pauline Syrett called Bob Penter to discuss any outstanding issues. Bob Penter reported that he had no further concerns following his discussions with John Aspinall

Tuwharetoa

John Aspinall first approached Tikitu Nathan of Tuwharetoa to canvas the iwi’s response to our application in January 2000. We sent a copy of a letter providing details of the proposal (Appendix VI) and offered to meet with representatives of iwi. Tikitu Nathan indicated that this would not be necessary. He responded to a telephone call from Bevan Tipene-Matua in October to say that they had some concerns about the application because of possible impacts on other species. They are concerned for the protection of the mauri of these species. Tikitu Nathan requested that we forward a copy of Ngai Tahu’s response to assist them in formulating a response of their own. With Linda Constable’s agreement, a copy of this letter was forwarded to Tikitu Nathan on 6 October 2000.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 23

Te Tau Ihu O Te Waka A Maui (North of the South Island iwi)

Ngati Koata

Jim Elkington of Ngati Koata responded on 4 August 2000 to a letter from us to say that Ngati Koata had no problems with the application.

Other iwi

Representatives of 42 other iwi throughout New Zealand (including Te Tau Ihu O Te Waka A Maui that had not formally responded) were sent a letter on 3 October providing information on the current application, offering to provide additional information if required, and requesting that they send us their comments, and raise any concerns they have with us (Appendix VI).

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 24

Issues of potential significance to Maori

Potential issues of significance in relation to Treaty outcomes

Key Treaty Outcomes Significant adverse effect? Comment

Yes No

The relevance of unresolved Treaty claims to the Waitangi Tribunal The continued ability of Maori to exert their developmental right as implied by the Treaty, where these are recognised by the Waitangi Tribunal.

Potential issues of significance in relation to environmental outcomes

Key environmental outcomes Significant adverse effect? Comment

Yes No The continued and improved availability, Insects proposed for introduction quantity and quality of traditional food feed only on weeds of no food resources (mahinga kai) value to Maori The continued availability, quantity and quality of traditional Maori natural resources The retention of New Zealand’s diverse This will be enhanced by range of indigenous flora and fauna successful biological control of hawkweeds The protection of indigenous flora and fauna valued by Maori The purity of water (inland and coastal) Water quality will be enhanced and the need to retain and extend its following restoration of taller productive and life-sustaining capacity vegetation The purity of land and the need to retain This will be improved through and extend its productive and life- suppression of weedy sustaining capacity hawkweeds and restoration of grasses The purity of air and the need to retain and extend its productive and life- sustaining capacity The purity of human health and well-being The insects will not cause disease, bite or sting, or pose any health risk The restoration and retention of natural Natural habitats will be enhanced habitats by suppression of weedy hawkweeds

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 25

Potential issues of significance in relation to cultural outcomes

Key cultural outcomes Comment Significant adverse effect Yes No The recognition of Maori cultural, spiritual, See below ethical, or socio-economic values The protection of the mauri of peoples The preservation and maintenance of traditional Maori knowledge by Maori The maintenance, expression and control by Maori of their traditional practices e.g. kaitiakitanga, tapu, rahui The protection of the mauri (spiritual integrity or life-force) of valued flora and fauna The protection of the mauri of the land The protection of the mauri of waterways (inland and offshore) The protection of the mauri of air and other taonga

In the application to ERMA New Zealand by the New Zealand Citrus Growers Inc. and the Avocado Industry Council for release of Thripobius semiluteus, (NOR99002) the applicants concluded that the cultural and spiritual aspects of introduced biological control agents were still being determined by Tangata Whenua, and that these general issues needed to be discussed in a wider forum. As with their application, this one does not raise any additional issues that are not generic to all biological control applications.

Potential issues of significance with relation to health and well-being outcomes

Key health and well-being outcomes Significant adverse effect? Comment Yes No The protection of taha wairua: spirituality balance with nature, protection of mauri The protection of taha whanaunga: responsibility to the collective, the capacity to belong, to care and to share The protection of the taha hinengaro: mental health and well-being, the capacity to communicate, to think and to feel The protection of the taha tinana: physical growth and development

The applicants have identified the following ways in which successful control of Hieracium could benefit Maori: • restoring the integrity of native grasslands;

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 26

• minimising the risk of extinction of rare native plants; • increasing indigenous biodiversity in Hieracium-infested areas; • protecting water quality in streams and rivers; • reducing the degradation of less severely affected areas.

Ngai Tahu is the iwi with the greatest potential interest in this application, because their lands include areas with the most severe infestations of the weed. In response to the draft application (see Appendix V), Ngai Tahu say that they ‘consider bio-control preferable to most insecticides’, and they support the view that chemicals can be damaging, and should therefore be used as little as possible. However, they also express general concerns for the safety of biological contol, and urge ERMA New Zealand to take a conservative approach in considering applications for further introductions. They point out that insufficient time has elapsed for the effectiveness of earlier introductions to be assessed. The applicants agree with this, but believe that delaying release of further beneficial organisms will allow Hieracium to expand its range and impact still further, especially for those species other than H. pilosella, which are unlikely to be affected by the releases made so far.

The ability of the organism to escape from containment

The intention is that the three insects be released from containment and spread throughout the Hieracium-infested areas of New Zealand to establish self-sustaining, and multiplying, natural populations as quickly as possible.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 27

Table 1. Systematic identification and assessment of reasonable and foreseeable risks associated with this application.

Potential adverse risk Source of risk Element of risk Method used to Full assessment identify risk Life-supporting capacity Sudden removal of Erosion following Common sense Effects likely to be of air/water and Hieracium spp. from increased bare assessment, temporary, and minor soil/ecosystems ecosystems ground research, feedback on previous applications for Hieracium-feeding and other insect biological control agents Increased rate of " Negative effects likely water run-off, decline to be temporary and in water quality minor Successional change " Risk of negative impact very low Capacity of people to Insects attack non- Economic damage to " Probability of attack, provide economic, target plants non-target and magnitude of social, and cultural well- agricultural, forestry, effect is very low, being and foreseeable or horticultural crops, (see Appendices I–III) needs of future and garden plants generations Syrphids feed on Loss to bee-keepers " The risk that this pollen from competition with occurs is high, but the bees for pollen magnitude of the effect is likely to be very low Increase in weed " Magnitude of effect is populations through rated as very low, increased pollination impact of effect negligible compared with that of other pollinators Gall flies attain high Nuisance value to " Negligible risk, flies populations the public short-lived Increases in vespid " Risk and magnitude wasp populations of risk low, flies would from increase in prey comprise very small pool proportion of potential prey pool Hieracium spp. are Economic loss from " Risk is moderate, removed from seeds unavailable for losses are low grasslands harvesting relative to the economic costs to farmers of Hieracium- infested pasture

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 28

Potential adverse risk Source of risk Element of risk Method used to Full assessment identify risk Replacement of Economic costs to " Risk is assessed as Hieracium spp. with farmers negligible, as current more damaging value of land severely weeds infested with Hieracium is zero. Sustainability of native Insects attack Suppression, or " The risk of the effect and valued introduced desirable plants extinction, of valued, is negligible to low, flora and fauna rare, or endangered but the magnitude of plant populations the effect might be high (see Appendices I–III) Accidental release of Predators, " Risk is very low, if associated parasitoids, or insect quarantine organisms pathogens released procedures adhered to (Sections 9 and 14) Species becomes Competes with " Risk low, negative established as a new native, or desired effect likely to be less element of the fauna introduced element than that of the weed of the fauna species on threatened fauna Increased prey pool " May result in for parasitoids, environmental costs predators, and or benefits, risk diseases of insects uncertain, magnitude adversely affects of effect likely to be food web low Hieracium spp. Increased bare " Risk low, research removed from native ground inhibits indicates that native grasslands growth of native plants grow better plants when Hieracium is suppressed Hieracium spp. are " Risk is low, no replaced by worse evidence for this weeds scenario in any experimental trial to date Intrinsic values of New species Dilution of native " The risk is high, but ecosystems become common biodiversity by the effect is negligible elements of the introduced species to low. Most exotic fauna invertebrates are accidental introductions Public health Insects are harmful Members of the " Risks negligible, the to humans public receive bites, species proposed for stings, swallow flies introduction pose no and are then known health risks poisoned or choke (see Section 14).

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 29

Potential adverse risk Source of risk Element of risk Method used to Full assessment identify risk Insects carry disease Levels of disease " Risks negligible, the harmful to humans increase species proposed for introduction pose no health risks (see Section 14) Relationship of Maori to consultation with Significant generic their taonga, wahi tapu, Maori, responses to issues of possible water, etc. previous applications impact of biological control on cultural values

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 30

Table 2. Likelihood and consequences of beneficial effects resulting from this application. Likelihood is based on the probability of achieving control.

Potential benefit to Source of benefit Element Method used to Assessment of producing benefit identify benefit size and probability of benefit Life-supporting capacity Replacement of Improved water Common-sense Benefit is likely to of air/water and Hieracium spp. by quality through assessment, research, be high, probability soil/ecosystems more desirable slowed run-off feedback on previous of achieving it, plants applications to introduce medium Hieracium-feeding and other insect biological control agents Succession to " Benefit is likely to more diverse flora, be high, probability increased cover of achieving it, medium Reduced invasion " Benefit is likely to of Hieracium into be high, probability vulnerable of achieving it, ecosystems medium Capacity of people to Increase in Increased pasture " Benefit estimated provide economic, palatable pasture production as $4.4–78M, social and cultural well- species probability medium being and foreseeable needs of future generations Restoration of Increased " Benefits high, tussock appreciation of probability medium landscapes tussock landscapes Sustainability of native Substantial Restoration of " Benefit high, and valued introduced reduction in the native grasslands probability medium flora and fauna populations, vigour, and spread of Hieracium species Replacement of H. " Benefit high, lepidulum by native probability medium understorey species Increased " Benefit high, biodiversity of flora probability medium and fauna in affected habitats Intrinsic values of Substantial Restoration of " Benefit high, ecosystems reduction in the ecosystem to more probability medium populations, natural state vigour, and spread of Hieracium species

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 31

Potential benefit to Source of benefit Element Method used to Assessment of producing benefit identify benefit size and probability of benefit Relationship of Maori to Substantial Protection of flora “ Benefit high, their taonga, wahi tapu, reduction in and fauna, probability medium water, etc. populations, improved water vigour, and spread quality of Hieracium species

Table 3 Predicted level of attack on five weedy species of Hieracium by five insect biological control agents. equivalent level of attack to that on H. pilosella, minor level of attack compared to that on H. pilosella.

Hawkweed species Oxyptilus Aulacidea Cheilosia Cheilosia Macrolabis pilosellae subterminalis urbana psilophthalma pilosellae H. pilosella

H. praealtum -

H. caespitosum -

H. aurantiacum -

H. lepidulum - -

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 32

Table 4. Summary table of costs and benefits Beneficial Name of benefit/cost Best estimate Uncertainty Spatial distribution Temporal effects range distribution Direct Increased stock $3.7M $0 9.7M Hill and high country Annual monetary numbers pastoral areas benefits Direct non- Improved soil moisture Hill and high country Long-term monetary retention pastoral and conservation benefits areas Increased indigenous Hill and high country Long-term biodiversity (flora and pastoral and conservation fauna) areas Sub-total $3.7M $0 9.7M

Indirect Increased scenic and Hill and high country Long-term benefits cultural value of land pastoral and conservation areas Data from post-release Nation-wide Medium-term monitoring programme will increase understanding of biocontrol effects, and support future decision making

Direct Rearing, distribution and $500k $450 550k Hill and high country Over 3 years* monetary establishment costs for pastoral and conservation costs all 3 organisms areas Measuring impact of $350k $300 400k Hill and high country Over 10 years* organisms post-release pastoral and conservation areas Loss/reduction in $7k $0 15k Hill and high country Long-term Hieracium seed industry pastoral and conservation areas Indirect None identified costs * We estimate that no further costs will be incurred after 3 and 10 years respectively on these items

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 33

International and related matters

15. Information on all occasions where the organism has been considered by the government of any prescribed State or country or by any prescribed organisation and the results of such consideration: Where no countries or organisations are prescribed by regulations made under section 140(1)9k) of the Act, this section can be omitted. If the applicant is aware that the organism has previously been considered by, for example, any OECD or APEC country, information on the nature of that consideration, including the result, should be provided if known. [ Yes/No? ] further information

None of these three insects are known to have been released anywhere outside their natural range in Europe for any purpose, including biological control. However, tests have been conducted by CABI Bioscience in Delémont, Switzerland, with test material from the USA on behalf of the state of Idaho that has an interest in biological control of H.caespitosum.

16. Information on New Zealand’s international obligations that may be relevant to the application: Where the applicant is aware that New Zealand’s international obligations may be relevant to the application, indicate the nature of the obligation and the effect this may have on the application. If the applicant is aware of obligations such as the WTO Agreements, the Convention on International Trade in Endangered Species (CITES), Trans Tasman Mutual Recognition Agreement and the like that may be relevant to the application, then information on these obligations should be provided, if known. [ Yes/No? ] further information

None known

Previous considerations

17. If the application relates to an organism that has been previously considered by the Advisory Committee on Novel Genetic Techniques or the Minister for the Environment on the recommendation of the Interim Assessment Group, details of the consideration and its results: [ Yes/No? ] further information

Risks associated with the importation of Macrolabis pilosellae have been assessed in ERMA application no. 98001. Cheilosia urbana, and C. psilophthalma were imported into containment under MAF, but risks were similar to those for M. pilosellae.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 34

Other relevant legislation

18. Information on other legislation relevant to the organism and its use throughout its life cycle. If the organism is also subject to other legislation (eg. an Import Health Standard under the Biosecurity Act 1993, or resource consent under the Resource Management Act 1991), details should be provided. [ Yes/No? ] further information

Insects imported into containment are bound by the Biosecurity Act 1993, currently administered by MAF Quality Management, under which disease and parasitoid screening is undertaken (see Sections 9 and 14). MAF permit no. 2000009607 for importing Cheilosia urbana and C. psilophthalma is valid until 29 June 2001, and permit no. 2000009543 for importing Macrolabis pilosellae is valid until 21 June 2001.

Hieracium species are not currently included in any regional councils' Regional Pest Management Strategies, so there is no legislation requiring these weeds to be controlled. One of the reasons for this is that there is no workable control measure available that could be enforced.

Glossary

19. A glossary of scientific and technical terms used in the application. This may be appended to the application on a separate form if desired. [ Yes/No? ] further information

Biological control: the use of one living organism to control another: in this case by establishing sustaining populations of a new organism. Dorsal surface: upper surface Gall: a fleshy deformity induced on a plant by an insect, or other organism, to provide shelter and nutrients. Host-specific: using only a single host, or a range of closely related host plants Larva(e): juvenile stage of an insect(s) Leaf axil: the upper angle between a leaf and its adjoining stem Multivoltine: undergoing several generations per year Parasitoid: an insect that lays its eggs in, or on, its host. Developing young kill the host. Phytophagous: plant-feeding Pupa(e): stage of insect development between larva and adult Rosette: plant in which leaves radiate around a centre point Setae: bristle, or 'hair'

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 35

Stolon: horizontal stem running along the ground, capable of rooting and forming a new plant at nodes Univoltine: passing through only one generation per year

Other relevant information

20. Provide here any other information required by the Act or regulations not included under any other section of this form. [ yes ] further information [ no ] commercially sensitive information

References

Barnes, H.F. (1937) Checklist of the Cedidomyidae of New Zealand. Transactions of the New Zealand Institute 67, 115–121.

Chandler, P. (1998) Checklists of insects of the British Isles (new series) Part 1: Diptera. Handbooks for the identification of British insects Volume 12, Royal Entomological Society, 234 p.

Crawley, M.J. (1989) The successes and failures of weed biocontrol using insects. Biocontrol News and Information 10, 213 223.

Emberson, R.M. (in press) Endemic biodiversity, natural enemies, and the future of biological control. In: X International Symposium on Biological Control of Weeds, Bozeman, Montana, USA, 4–14 July 1999.

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 No. 49, Canberra, Australia, pp. 71–75.

Fowler, S.V.; Syrett, P. ;Jarvis, P. (in press) Will the New Environmental Risk Management Authority, together with some expected and unexpected non-target effects, cause biological control of broom to fail in New Zealand? In: X International Symposium on Biological Control of Weeds, Bozeman, Montana, USA, 4–14 July 1999.

Greer, G.; Sheppard, R.L. (1990) An economic evaluation of the benefits of research into biological control of Clematis vitalba. Agribusiness and Economics Research Unit, Lincoln College, Research Report No. 203, 49 p.

Grosskopf, G. (1994) Investigations on potential biocontrol agents of mouse-ear hawkweed, Hieracium pilosella. Unpublished report. International Institute of Biological Control, Delémont, Switzerland, 10 p.

Grosskopf, G.: Hassler, A. 1(998) Investigations on potential biocontrol agents of mouse-ear hawkweed, Hieracium pilosella L. Annual Report, CABI Bioscience, Delémont, Switzerland, 40p.

Grosskopf, G.; Murphy, S. (1999) Investigations on potential biocontrol agents of mouse-ear hawkweed, Hieracium pilosella L. Annual Report, CABI Bioscience, Delémont, Switzerland, 43p.

Groves, R.H. (1991) Status of environmental weed control in Australia. Plant Protection Quarterly 6, 95–98.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 36

Grundy, P. (1989) An economic evaluation of biological control of Hieracium. Agribusiness and Economics Research Unit, Lincoln College, Research Report No. 202, 41 p.

Hunter, G.G. (1991) The distribution of hawkweeds (Hieracium spp.) in the South Island, indicating problem status. Journal of the New Zealand Mountain Lands Institute review 48, 21 31. Jordan, T. (1993) Investigations on potential biological control agents of mouse-ear hawkweed, Hieracium pilosella L., in Europe. Unpublished report. International Institute of Biological Control, Delémont, Switzerland, 12 p.

Julien, M.H. (1989) Biological control of weeds worldwide: trends, rates of success and the future. Biocontrol News and Information 10, 299 306.

Julien M.H.; Griffiths, M. (1998) Biological control of weeds a world catalogue of agents and their target weeds. 4th edit., CABI Publishing, Wallingford, UK, 223 p.

Keesing, V. (1993) The impact of Calluna vulgaris on native invertebrate communities in Tongariro National Park. In: Proceedings of the Second Heather Control Workshop, Turangi, 19–21 August 1993, Williams, K. & Keys, H. (Eds.) Department of Conservation, Tongariro/Taupo Conservancy Technical Series No. 1 pp. 19–21.

Lamb, K.P. (1960) A check list of New Zealand plant galls (Zoocecidia). Transactions of the Royal Society of New Zealand 88, 121–139.

Lane, R.P.; Crosskey, R.W. (1993) Medical insects and arachnids. Chapman & Hall, London, 723 p.

Likens, G.E. (1985) An experimental approach for the study of ecosystems. Journal of ecology 73, 381–396.

McEvoy, P.B. (1996) Host specificity and biological pest control. BioScience 46, 401 405.

McEvoy, P.B.; Coombs, E.M. (1999) Why things bite back: unintended consequences of biological weed control. In: P.A. Follett & J.J. Duan (eds.) Non-target effects of biological control. Kluwer Academic Publishers, Boston, Massachusetts, USA.

McFadyen, R.C. (1998) Biological control of weeds. Annual Review of Entomology 43, 369–393.

McMillan, G. (1991) Hieracium – an ecological perspective. Journal of the New Zealand Mountain Lands Institute review 48, 8 20.

Memmott, J.; Fowler, S.V.; Harman, H.M.; Hayes, L.M. (1996) How best to release a biological control agent? In: Moran V.C. & Hoffmann, J.H. (eds.) Proceedings of the IX International Symposium on Biological Control of Weeds, 19–26 January 1996, Stellenbosch, South Africa, pp. 291–296.

Rose, A.B.; Basher, L.R.; Wiser, S.K.; Platt, K.H.; Lynn, I.H. (1998). Factors predisposing short-tussock grasslands to Hieracium invasion in Marlborough, New Zealand.

Sárospataki, M. (1999) Phytophagous insects associated with Hieracium pilosella (Asteraceae) in Hungary, Central Europe. Environmental Entomology 28, 1–8.

Scott, R.R. (1984) New Zealand pest and beneficial insects. Lincoln University College of Agriculture, 373 p.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 37

Syrett, P. (1989) Hypericum perforatum L., St John's wort (Clusiaceae). In: Cameron, P.J.; Hill, R.L.; Bain, J.; Thomas, W.P. (eds.) A review of biological control of invertebrate pests and weeds in New Zealand 1874–1987. CABI Technical Communication No. 10, pp. 353–356.

Syrett, P.; Smith, L.A. (1998) The insect fauna of four weedy Hieracium (Asteraceae) species in New Zealand. New Zealand Journal of Zoology 25, 73–83.

Syrett, P.; Smith, L.A.; Grosskopf, G.; Meurk, C. (2000) Predicted contributions of a plume moth and a gall wasp to biological control of hawkweeds in New Zealand. In: Proceedings of the 7th Australasian Conference on Grassland Invertebrate Ecology, 27 Sept –1 Oct 1999, Perth, Western Australia, .

Valentine, E.W.; Walker, A.K. (1991) Annotated catalogue of New Zealand Hymenoptera. DSIR Plant Protection Report No. 4., New Zealand Department of Scientific and Industrial Research, 84 p.

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.

Zwolfer, H.; Harris, P. (1971) Host specificity determination of insects for biological control of weeds. Annual Review of Entomology 16, 159–178.

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz

ER-AF-NO1-4 3/99 Application for approval to import for release or release from FORM 1 containment any new organism under Section 34 of the Hazardous Substances and New Organisms Act 1996 Page 38

Summary of Application Contents (Please check the application is complete and identify attachments)

[ Yes/No? ] Fees enclosed [ Yes/No? ] Assessment of effects included [ Yes/No? ] Confidential information supplied [ Yes/No? ] Signed and dated [ Yes ] Appendices attached and cross-referenced (list below) Appendix I Host range of the hieracium gall fly, Macrolabis pilosellae Appendix II Host range of the hieracium root-feedinghover fly, Cheilosia urbana Appendix III Host range of the hieracium crown-feeding hover fly, Cheilosia psilophthalma Appendix IV Figure 1: Map of areas of New Zealand affected by Hieracium Appendix V Copies of permits issued by MAF Appendix VI Summary of Maori consultation, and copies of correspondence Appendix VII Personal communications cited in application

Signature of applicant or person authorised on behalf of applicant Date:

20 Customhouse Quay, Cnr Waring Taylor & Customhouse Quay PO Box 131, Wellington Phone: 04-473 8426 Fax: 04-473 8433 Email: [email protected] Website: www.ermanz.govt.nz