APPLICATION FORM RELEASE

Application to import for release or to release from containment new organisms under the Hazardous Substances and New Organisms Act 1996

Send by post to: Environmental Protection Authority, Private Bag 63002, Wellington 6140 OR email to: [email protected]

Application number

APP201362

Applicant

Auckland Council Private Bag 92-012

Auckland

Key contact

Dr Nick Waipara

www.epa.govt.nz 2

Application to import for release or to release from containment new organisms

Important

This application form is to seek approval to import for release or release from containment new organisms (including genetically modified organisms). The application form is also to be used when applying to import for release or release from containment new organisms that are or are contained within a human or veterinary medicine. Applications may undergo rapid assessment at the Authority‘s discretion if they fulfil specific criteria. This application will be publicly notified unless the Authority undertakes a rapid assessment of the application. This application form will be made publicly available so any confidential information must be collated in a separate labelled appendix. The fee for this application can be found on our website at www.epa.govt.nz. If you need help to complete this form, please look at our website (www.epa.govt.nz) or email us at [email protected]. This form was approved on 1 May 2012.

May 2012 EPA0160

3

Application to import for release or to release from containment new organisms

1. Brief application description Provide a short description (approximately 30 words) of what you are applying to do.

To import and release the yellow leaf spot Kordyana sp. as a biological control agent for the weed (Tradescantia fluminensis).

2. Summary of application Provide a plain English, non-technical description of what you are applying to do and why you want to do it.

A collective comprising 13 regional councils and the Department of Conservation (DOC) supports the development of biological control as the strategy most likely to achieve widespread, cost-effective and environmentally acceptable control of tradescantia (Tradescantia fluminensis). This application seeks approval to introduce a white smut fungus – the Brazilian yellow leaf spot fungus (Kordyana sp.) – as a biocontrol agent of tradescantia. The EPA has already approved the introduction of the chrysomelid beetles Neolema ogloblini (approval code NOR000043), Lema basicostata and Neolema abbreviata (approval code ERMA200683) as biocontrol agents for tradescantia. These feed on the leaves, stems and shoot tips of the weed respectively (http://www.epa.govt.nz/new-organisms/popular-no-topics/Pages/biocontrol-for-tradescantia.aspx). Leaf infection by Kordyana sp. will complement the effects of these insects and this fungus would also be introduced from Brazil, the native range of the weed.

Auckland Council makes this application on behalf of the National Biocontrol Collective. Landcare Research is the science provider for this research, and contracted Professor Robert Barreto (Departamento de Fitopatologia, Universidade Federal de Viçosa, Brazil) to determine if the yellow leaf spot fungus would be suitable to release in New Zealand. Landcare Research has contracted Jane Barton to provide expert input on pathology, and Richard Hill & Associates to prepare the application and manage the application process on behalf of Auckland Council.

Tradescantia (sometimes called wandering Willie or wandering Jew) forms thick mats at many forest margins, in forest clearings, and on stream margins in northern New Zealand. Mats overshadow and kill low-growing , including native tree seedlings that are essential for forest regeneration. Without intervention, heavy tradescantia infestations guarantee the eventual destruction of small forest remnants and shrinkage of larger stands over time as forest margins retreat. Tradescantia is a hated weed in suburban backyards and civic parks, and commonly causes severe allergic reactions in dogs that walk in it. Tradescantia has no significant beneficial attributes.

The biological control programme aims to gain control over this weed by establishing a range of natural enemies that damage it in a variety of ways. Together, the damage caused by control agents is expected to benefit the environment in two ways: by reducing the growth rate and bulk of tradescantia where it currently exists allowing growth of native seedlings in forest ecosystems, and by stopping the development of damaging mats at new sites of invasion.

May 2012 EPA0160

4

Application to import for release or to release from containment new organisms

The following potentially adverse effects of introducing Kordyana sp. have been identified:

The risk of direct damage to native plants and valued exotic ornamental plants Indirect effects on flora and fauna as a result of disruption of trophic relationships The removal of tradescantia as a habitat for native fauna. None of these risks is considered to be significant. Host range tests indicate that no New Zealand native plants will be at risk from Kordyana sp. Tests indicated that valued ornamental plants related to tradescantia will not be significantly affected. The agent is host specific to the weed, and because significant infections will occur only where the weed is abundant, no significant disturbance of ecological relationships can occur outside of tradescantia infestations. The presence of tradescantia itself massively modifies natural interactions between species, and any reduction in the weed will help reverse those impacts. Tradescantia has been observed to suppress the populations of some native invertebrates, but to enhance others. Any net biodiversity benefit is unlikely to ever outweigh the adverse effect of the weed on other species, particularly native plants. The main beneficial and adverse effects identified in previous consultations will be addressed in the application form and background information will be posted on the Landcare Research website (Landcare Research 2012).

3. Describe the background and aims of the application This section is intended to put the new organism(s) in perspective of how they will be used. You may use more technical language but please make sure that any technical words used are included in a glossary.

The adverse effects of Tradescantia fluminensis on forest ecosystems are better documented than those of any other weed in New Zealand and the weed is an acknowledged threat to biodiversity values nationwide. The effects of tradescantia can be managed at particular sites by hand removal of the plants or by applying herbicides, but successful control is resource-intensive and often damaging to associated vegetation. The weed continues to spread and accumulate in areas that are beyond the reach of conventional weed management strategies. Kordyana sp. is expected to contribute to biological control of tradescantia, reducing those adverse effects. Biological control is the only hope of sustained and widespread management of those adverse effects because agents persist from year to year, and can disperse to colonise even hidden tradescantia populations. Background reviews, reports and published papers about the potential of biological control, the nature of the tradescantia threat, and the biology of Kordyana sp. can be found on the Landcare Research website (Landcare Research 2012).

Tradescantia is now seen by many as one of the weeds that most threaten the integrity of important forest remnants in the North Island (Landcare Research 2012). The dense mats of interleaved tradescantia stems that form on the forest floor in New Zealand are often 50 cm deep. A single stem may be 1.5 m long with many branches. Kelly and Skipworth (1984) estimated that a square metre of ground with a standing crop of 1400 g could comprise 300 tradescantia plants with a total stem length of 900 m. Where tradescantia is present on the forest floor tall native seedlings tend to be absent, implying the death of small seedlings under this weed (Kelly & Skipworth 1984). Heavy shading by tradescantia kills all seedlings, no matter how many are present. This means that robust forest ecosystems with abundant seedlings are just as much at risk from tradescantia as struggling urban reserves (Standish et al. 2001; Landcare Research 2012). Edge effects extend at least 40–50 m into forest May 2012 EPA0160

5

Application to import for release or to release from containment new organisms

remnants, so areas of less than 9 ha are dominated by such edge effects, including tradescantia invasion (Standish et al. 2004). Tradescantia mats can persist on the forest floor even once light gaps close. The adverse effects are evident at several trophic levels. It grows in a radically different way from any native vegetation, fundamentally altering ecosystem processes such as litter decomposition, nutrient cycling (Standish et al. 2004), and the process by which lowland podocarp–broadleaved forests, swamps and stream margins regenerate (Landcare Research 2012). There is also an indirect relationship between the presence of tradescantia and the nature of the invertebrate communities in lowland forests (Yeates & Williams 2001; Landcare Research 2012).

Physical removal by ‗rolling up‘ and removing the mat and then painstakingly examining the exposed ground for fragments that might regenerate is regarded as the most effective ‗herbicide-free‘ technique available for small patches of tradescantia. This tactic minimises damage to non-target vegetation. Weed management can remove the shading threat, but current control measures themselves have significant side effects. Potential damage to desirable vegetation by control measures (Esler 1988) needs to be weighed against the estimated impact of continued weed invasion in the event of no weed control (Standish et al. 2002).

Standish (2001) estimated that regeneration of even the most shade tolerant native species was prevented where biomass of tradescantia on the ground was greater than 200 g dry matter/m2. She found that this level was exceeded in 83% of measurements in New Zealand, and levels can range from 116 to 1200 g/m2. In Brazil, comparable dry biomass samples ranged from 46 to 296 g/m2, with only 12% exceeding 200 g/m2. Just how much of this biomass reduction is due to the action of potential biocontrol agents has not been studied in Brazil. If it is due to natural enemies then a classical biological control programme has good prospects for success in New Zealand.

4. Information about the new organism(s)

Provide a taxonomic description of the new organism(s) (if the organism is a genetically modified organism, provide a taxonomic description of the host organism(s) and details of the genetic modification). Describe the biology and main features of the organism including if it has inseparable organisms. Describe if the organism has affinities (e.g. close taxonomic relationships) with other organisms in New Zealand. Could the organism form an undesirable self-sustaining population? If not, why not? What is the ease with which the organism could be eradicated if it established an undesirable self-sustaining population?

The organism that is the subject of this application is also the subject of: a. an innovative medicine application as defined in section 23A of the Medicines Act 1981. No b. an innovative agricultural compound application as defined in Part 6 of the Agricultural Compounds and Veterinary Medicines Act 1997. No

4.1. Provide a taxonomic description of the new organism

May 2012 EPA0160

6

Application to import for release or to release from containment new organisms

Nine fungal species with potential as biocontrol agents were collected during a 5-year survey of diseased Tradescantia fluminensis plants in Brazil. One of these was a white smut fungus that caused distinctive yellow spots on infected leaves (Figure 1). Molecular analysis has shown that this organism has the following :

Kingdom = Fungi Phylum = Sub-phylum= Class = Order = Family = Brachybasidiaceae = Kordyana

This phylogenetic analysis showed that the organism is not K. tradescantiae (another Kordyana species known to infect Tradescantia hosts) and is distinct from any other Kordyana species (Barreto et al. 2010). It will be formally described using the specific epithet brasiliensis, the name used in unpublished reports (Prof. R. Barreto, pers. comm.). Until the paper is published the organism remains unnamed and will be referred to as Kordyana sp. in this application.

4.2. Describe if the organism has affinities (e.g. close taxonomic relationships) with other organisms in New Zealand There are no close relatives of Kordyana sp. present in New Zealand. There are no fungi in the same family (Brachybasidiaceae) known in New Zealand. The closest relatives of Kordyana sp. that do occur in New Zealand are seven species (2 exotic, 5 indigenous) of (Family Exobasidiaceae), which are in the same order as Kordyana (Exobasidiales) (http://nzfungi2.landcareresearch.co.nz). None are found on Tradescantia spp. or any close relatives. Also in the same order are two exotic species of (family ) that have been reported on palm trees in New Zealand.

4.3. Describe the biology and main features of the organism including if it has inseparable organisms Basidiospores of Kordyana sp. that land on the surface of Tradescantia fluminensis leaves germinate and form germ-tubes. These penetrate the plant‘s epidermis. Fungal mycelium then grows into the tissues of the host plant causing disease. Yellow (chlorotic) lesions develop and expand. Infected plant tissues become yellow initially, and later, starting centrally, become brown (Figure 1). As lesions expand and coalesce with each other the leaves become blighted and die but generally remain attached to the stem. High levels of humidity favour the formation of groups of basidia (called caespituli) that emerge through the stomata of infected tissues and sporulate. Basidiospores form and are forcibly released. These are small and lightweight and easily spread through air turbulence. The cycle begins again when these basidiospores land on a new T. fluminensis plant.

A drawing and photograph of spores of Kordyana sp., and the disease symptoms it causes on T. fluminensis, are provided (Figures 1 & 2, and figs 4 & 5 in Barreto et al., 2010). Figure 2 shows the reproductive structures of the

May 2012 EPA0160

7

Application to import for release or to release from containment new organisms

rust emerging through the stomata of an infected tradescantia plant. The smut produces single-celled, hyaline basidiospores that are oblong to elliptic, 3.0 x 5.0 — 8.0 x 15.0 µm, and smooth-walled. The basidiospores form on basidia, which emerge through the stomata of the host plant (Figure 2).

Figure 1. Typical yellow leaf spots caused by Kordyana sp. on a tradescantia leaf.

Figure 2. Reproductive structures of Kordyana sp. as seen under a microscope.

White smuts such as this Kordyana species do not have alternative hosts.

No inseparable organisms have yet been recorded associated with this species (Prof. R. Barreto, pers. comm.). As the fungus does not produce fully functional spores on artificial media, and it is not possible to collect and ship spores produced naturally because of their tiny size and fragility, Kordyana sp. will be imported as actively growing colonies on living tissues of T. fluminensis plants. Kordyana sp. will be isolated from any other associated fungi in a pathogen-proof containment facility to establish a pure colony of the smut on New Zealand T. fluminensis plants.

May 2012 EPA0160

8

Application to import for release or to release from containment new organisms

4.4. Could the organism form an undesirable self-sustaining population? If not, why not? No populations of Kordyana sp. established in New Zealand are expected to be undesirable. The pathogen is specific to Tradescantia fluminensis, an acknowledged plant pest under the National Pest Plant Accord (Biosecurity New Zealand, accessed 2012). It is not expected to have any significant adverse ecosystem effects (see Section 6.2). Although initial releases will be made in the Auckland Region it is expected that the pathogen will be able to colonise tradescantia populations wherever these occur in New Zealand. The object of introducing this fungus is to establish desirable, self-sustaining populations contributing to the suppression of tradescantia populations and/or the maintenance of control.

4.5. What is the ease with which the organism could be eradicated if it established an undesirable self- sustaining population?

The object of introducing Kordyana sp. to New Zealand is to establish desirable, self-sustaining populations of this smut as a biocontrol agent for Tradescantia fluminensis wherever it occurs. If Kordyana sp. establishes in New Zealand it will spread very rapidly and eradication will not be possible.

5. Detail of Māori engagement (if any)

Discuss any engagement or consultation with Māori undertaken and summarise the outcomes.

The EPA Māori National Network was consulted in 2007 and again in 2011 over plans to introduce a suite of control agents, and the likely costs and benefits of the biological control strategy for tradescantia. The EPA has advised that no additional national consultation on this proposal is necessary. Instead, the relevant issues already raised in consultation over applications to introduce the three other control agents for tradescantia are addressed here. The original submissions can be found on the Landcare Research website (Landcare Research 2012), along with information about the use of fungi as biocontrol agents. The applicant also intends to discuss the proposal in face-to-face meetings with Iwi Māori in the Auckland Region.

Previous consultation identified no significant positive effects other than those accruing to the general population (Section 6). Responses highlighted the potential of biological control to alleviate the increasing effects of tradescantia on native ecosystems, and to reduce any side-effects of herbicides used to manage tradescantia, both now and in the future. One response noted the environmental costs associated with doing nothing. Another sought to collaborate in the project (Landcare Research 2012).

Previous consultation revealed the following potential adverse effects that are of particular concern to Māori (Landcare Research 2012):

Effects on taonga species and the risk of Kordyana sp. becoming a pest Effects on native flora and fauna Plans for monitoring the positive and negative effects of biological control

May 2012 EPA0160

9

Application to import for release or to release from containment new organisms

Impact on mauri and whakapapa. The predominant concern of respondents was the safety of taonga and other native species. The history of biological control clearly records that the plant species most at risk from any control agent with a narrow host range are those that are closely related to the known host (e.g. Barton 2004). The evidence supporting this conclusion is presented on the Landcare Research website (Landcare Research 2012). The risk of disease appearing on any native species in Aotearoa New Zealand is considered to be insignificant. Kordyana sp. caused disease symptoms (Figure 1) on the leaves of Tradescantia fluminensis but not on other plants tested in Brazil. Even species of the same genus were immune (see Section 6.3). There are no plants in the native New Zealand flora that are even remotely related to tradescantia (Section 6.3). Adverse effects on native species other than plants could only occur through direct competition with native fungi or insects on tradescantia, or through a third species such as a mycoparasite. These effects are considered highly unlikely for the reasons discussed in Section 6.3. If agents reach sufficient numbers to reduce tradescantia mats to low levels, agents will not be able to colonise the remaining plants, and will die back to low numbers. If tradescantia comes back to damaging levels, then so will the agents. Tradescantia will never be eradicated, and this dynamic equilibrium is a standard/common feedback in biological control projects.

History shows that biological control of weeds can succeed in Aotearoa New Zealand but may not become evident for many years (Fowler et al. 2000). The level of tradescantia control achieved nationally will depend on the population levels that beetles will achieve once released in Aotearoa New Zealand, the frequency and virulence of Kordyana sp. on leaves, and how all of the agents interact. The outcome can only be known once all agents are well established. Although it is known that the agents will be introduced to Aotearoa New Zealand without the natural enemies that limit their performance in Brazil, we cannot be certain what mortality factors here (such as climate) might limit populations of insects or Kordyana sp. until those agents are released. Kordyana sp. uses windblown spores to disperse, and is likely to disperse faster than the invertebrate control agents. Comments on plans to assess the value of this project can be found in Section 8.2.

One respondent stated that if resources important to Māori are affected, so are the people, and Māori must be certain of the potential impacts on their resources. This reinforced past submissions that stressed the role of Māori as kaitiaki of taonga species, of tapu and whakapapa, and particularly of mauri (the principle interconnecting all things in the natural world, including human activity). Landcare Research chooses control agents that will have the smallest possible ecological footprint (see Section 6.3). However, introducing a new species to Aotearoa New Zealand is a permanent change to te taiao. In the end it comes down to whether it is acceptable to introduce a new, highly focused species to an ecosystem to try to overcome the largely uncontrolled effects of the weed on the mauri of that ecosystem. This is a trade-off that is acceptable to many but uncomfortable to some.

6. Identification and assessment of beneficial (positive) and adverse effects of the new organism(s)

Adverse effects include risks and costs. Beneficial or positive effects are benefits. Identification involves describing the potential effects that you are aware of (what might happen and how it might happen).

May 2012 EPA0160

10

Application to import for release or to release from containment new organisms

Assessment involves considering the magnitude of the effect and the likelihood or probability of the effect being realised.

Consider the adverse or positive effects in the context of this application on the environment (e.g. could the organism cause any significant displacement of any native species within its natural habitat, cause any significant deterioration of natural habitats or cause significant adverse effect to New Zealand‘s inherent genetic diversity, or is the organism likely to cause disease, be parasitic, or become a vector for animal or plant disease?), human health and safety, the relationship of Māori to the environment, the principles of the Treaty of Waitangi, society and the community, the market economy and New Zealand‘s international obligations.

6.1. Scenario We have made the following assumptions in assessing the risks, costs and benefits of introducing Kordyana sp.:

1. Without intervention, the area affected by tradescantia nationwide will continue to increase, as will its adverse ecosystem effects. 2. Biological control could reduce both the severity and frequency of tradescantia impacts on ecosystems, and reduce the rate of invasion. 3. The degree to which biological control will be successful nationwide remains uncertain. 4. The pathogen will establish throughout the range of the weed. 5. Any positive or negative effects of Kordyana sp. will occur across its range. 6. All four biocontrol agents are expected to co-occur. While the larvae of the chrysomelid beetles feed in different parts of the plant, Kordyana sp. may compete with beetles for tradescantia leaves. 7. The occurrence and performance of agents will vary both within and between infestations. The intensity of competition between agents is expected to be low because in Brazil insect defoliation occurs on the same leaves as Kordyana sp. infection to the extent expected by chance: there is no evidence of one damage type excluding the other. 8. Overall, the effects of the agents nationally are expected to be cumulative.

Beneficial or adverse effects of this proposal could arise from:

The introduction to New Zealand of the yellow spot fungus, Kordyana sp. The reduction in the abundance and vigour of tradescantia by biocontrol agents. The applicant and Landcare Research identified all reasonable risks and benefits by literature review, by public consultation, and by formal brainstorming during consultation over EPA applications submitted in 2007 and 2011 to introduce the control agents Neolema ogloblini, N. abbreviata and Lema basicostata. All effects identified are listed on the Landcare Research website (Landcare Research 2012). Those considered to be potentially significant are addressed in Sections 6.2 to 6.6.

May 2012 EPA0160

11

Application to import for release or to release from containment new organisms

6.2. Beneficial effects on the environment In itself, the introduction of Kordyana sp. is not expected to have any beneficial environmental effect. If the introduction of Kordyana sp. enhances the biological control programme and reduces tradescantia infestations, the environmental benefits would be:

To halt or reduce the degradation of forest remnants caused by tradescantia, resulting in: (a) Increased diversity of native plants and improved regeneration ability in recovering forests (b) Improved habitats for native fauna (c) Restoration of ecosystem processes affected by tradescantia.

Reduced loss of habitat to tradescantia and improved diversity of native plants

Biological control will reduce the adverse effects of tradescantia on biodiversity values by restricting the rate at which tradescantia mats build, and by slowing the establishment of new infestations.

The ecological health of forest remnants and stream margins in the North Island would be significantly improved if biological control reduced the biomass and percentage cover of tradescantia sufficiently to allow regeneration of native plant seedlings to resume, and to allow other vegetation to coexist with the weed. The biomass of tradescantia often reaches 800 g/m2 in the zone of medium shade surrounding forest remnants. Reduction of this biomass by 75% to 200 g/m2 or less (estimated to be equivalent to 70–90% cover) would allow regeneration of tolerant native species (Standish 2001). R. Grimmett (pers. comm.; Landcare Research, 2012) suggested that a 50–70% reduction in tradescantia cover would be required to re-establish native seedling regeneration. Persistent suppression of native seedling regeneration by tradescantia skews the age structure of populations, and must inevitably lead to decline of diversity in even large, long-lived plant species (Esler 1988). Reduction in the cover of tradescantia below the threshold required to restore a measure of seedling survival (between 50 and 90% cover) would restore the pattern of plant succession to resemble that of un-infested forest. Only biological control is capable of providing this type of habitat protection, as agents seek out infestations wherever they occur, and agent populations persist from year to year. Biological control may achieve this alone, or it may facilitate other control methods by minimising biomass accumulation. A range of beneficial outcomes across the range of the weed is expected because the four control agents will have different developmental requirements, and the net impact on tradescantia will vary regionally and from year to year with climate variation.

Improved habitats for native fauna

The strongest predictor of species richness of soil invertebrates and microfauna within a forest is the species richness of the vegetation. Tradescantia infestations change forest structure and vegetation species richness, and therefore influence the species richness of the insect fauna (Toft et al. 2001; Standish 2004; Landcare Research 2012). Resumption of regeneration of native seedlings within infested areas as a result of successful biological control would increase vegetation diversity, and help restore the abundance and range of invertebrates and micro- organisms associated with those plants to a more natural state (Yeates & Williams 2001; Winks et al. 2003; Standish 2004).

May 2012 EPA0160

12

Application to import for release or to release from containment new organisms

Tradescantia clearly has major adverse effects on natural habitats, but simultaneous beneficial effects have also been recorded. Standish et al. (2002) found that the endangered endemic snail Powelliphanta traversi traversi did better in five of seven forest remnants that had tradescantia because the weed was an effective refuge. However, the presence of the weed wasn‘t critical to that success because the snail was present in 11 other forest remnants where tradescantia was absent. Similarly, the growth of tradescantia in the two remaining forest remnants on Stephens Island has prevented or greatly reduced the establishment of seedling tree and understory plants, and greatly impaired the access of tuatara and fairy prions to their burrows (Brown & Rees 1995). At the same time, they found that tradescantia-infested areas on the island carried higher densities of striped gecko, snails (Rhytida stephensensis) and native earthworms than tradescantia-free areas, possibly because the weed mat provided partial protection from predation by tuatara. Tradescantia is detrimental to the integrity of natural forest ecosystems overall but provides ecosystem services in some habitats. Standish et al. (2002) suggested that gradual control of tradescantia, with replacement by native cover, would simultaneously benefit snails while achieving other biodiversity maintenance goals. Rather than eliminating tradescantia infestations, biological control is expected to achieve this style of control by reducing the biomass accumulation of existing stands and limiting invasion of new sites (see Section 6.1).

Restoration of ecosystem processes

Tradescantia mats grow in a radically different way from any native vegetation, and fundamentally alter the way infested habitats work. Mats change how plants succeed each other as vegetation regenerates in lowland podocarp–broadleaved forests, swamps and stream margins (Landcare Research 2012), and probably adversely affect the integrity of other native ecosystems. The process of litter decomposition and nutrient recycling is profoundly different around tradescantia than it is on open forest floor. Standish et al. (2004) found that the productivity of several sites examined was high, but that litter breakdown was particularly rapid where tradescantia occurred. Available nitrogen was higher under weed mats than in non-tradescantia plots. They concluded that these differences were probably due to differences in vegetation structure between tradescantia-infested plots and tradescantia-free plots, and associated differences in microclimate. Some of these effects could be seen as positive for the forest environment, others negative, but in either case the presence of tradescantia changes the systems from a natural trajectory. Biological control will partially reverse those changes by reducing the biomass of existing mats, providing a mosaic of new habitat opportunities for native seedlings and limiting the establishment of new infestations.

6.3. Adverse effects on the environment Kordyana sp. would be detrimental to the environment if it:

Significantly reduced the size or fitness of native plant populations Competed significantly with native fungi on leaves of tradescantia or native plants Hybridised with native species Interacted with mycoparasites in a way that adversely affected native fungi, or Caused replacement of tradescantia with a worse weed.

May 2012 EPA0160

13

Application to import for release or to release from containment new organisms

Risks of Kordyana sp. to non-target plants

Wapshere (1974) states that the plants most likely to be attacked by a proposed biocontrol agent are those most closely related to the target weed and growing in the area where that agent will be used. He developed the centrifugal phylogenetic testing system for selecting test plants, which will accurately predict the range of hosts that the control agent will use in its new environment. This system has been universally adopted by biological control practitioners worldwide and continues to be effective at preventing unexpected attack of non-target plants in the field, especially for pathogens (Barton (née Fröhlich) 2004, Barton 2012). Fowler (2007) selected the test-plant species to be included in host range testing of potential biocontrol agents for Tradescantia fluminensis (both invertebrates and fungi). The list is presented in table 1, and his methodology can be found on the Landcare Research website (Landcare Research 2012).

The most important consideration when compiling a list of plants to be included in host range testing of Kordyana sp. was: ―what are the closest relatives of T. fluminensis that grow in New Zealand?‖ Tradescantia fluminensis belongs to the family (order , Table 1). There are no plants native to New Zealand belonging to this family (or even to this order). Exotic plants in the Commelinaceae that occur in New Zealand include a range of ornamental house and garden plants, some of which have become established in the wild: Commelina diffusa (present in cultivation), Gibasis pellucida (occasionally found outdoors), G. schiediana (naturalised), Tradescantia albiflora (naturalised), T. cerinthoides (naturalised), T. virginiana (naturalised) and T. zebrina (in cultivation) (Data from (http://nzflora.landcareresearch.co.nz/ accessed May 2012 and/or G. Houliston and I. Schönberger, Landcare Research, pers. comm.; common names can be found in Table 1). There is also an entity referred to as T. fluminensis ―big‖ here, but New Zealand botanists have not yet been able to name this (Peter Heenan, Landcare Research, pers. comm.). There are two other families in the Commelinales that occur as exotics in New Zealand, the and the Pontederiaceae. Within the Haemodoraceae there are two monospecific genera that occur in New Zealand: the ornamental Anigozanthos flavidus (kangaroo paw), a native of Australia, and thyrsiflora (red root), which has naturalised here. In the Pontederiaceae there are three species: Pontederia cordata, Eichhornia azurea (anchored water hyacinth, status in New Zealand not given on database) and E. crassipes (water hyacinth, naturalised) (http://nzflora.landcareresearch.co.nz/, accessed May 2012). While several of these species are well known as ornamental plants, they are not significantly valued in New Zealand. Most of them are considered undesirable weeds, or as potential weeds (particularly the Tradescantia and Eichhornia species).

Less closely related to the Commelinales are the orders Arecales, Poales and the Zingiberales (Table 1). The only native plant in the Arecales is the endemic palm Rhopalostylis sapida (nīkau) and this is arguably the closest relative to tradescantia in New Zealand. This plant was included in host range tests of the insect biocontrol agents released in New Zealand and was completely unharmed (Landcare Research 2012). Unfortunately, it proved impossible to source this plant in Brazil for testing of Kordyana sp. The plant could not be tested in New Zealand because we do not have a pathogen-proof quarantine facility. There are many plants native to New Zealand in the large order Poales, including members of the families Typhaceae, Juncaceae, , Restionaceae and . Again, it was not possible to source these in Brazil, or test them in New Zealand, but in this case it was possible to test other plants that belong to these families as surrogates (e.g. Cyperus rotundus in the Cyperaceae

May 2012 EPA0160

14

Application to import for release or to release from containment new organisms

and a Juncus species in the Juncaceae; Table 1). There are no plants native to New Zealand in the Zingiberales, but there are a few plants in this family grown here as ornamentals: e.g. Canna sp. (canna lilies) in the Cannaceae; Musa sp. (bananas) in the Musaceae; Strelitzia sp. in the Strelitziaceae; and, Zingiber sp. and Hedychium spp. in the Zingiberaceae.

The Kordyana species relevant to this application has only been collected from T. fluminensis in the field in Brazil (Prof. R. Barreto, pers. comm.) but its host range across Brazil has not been fully studied.

Taking all of the information above into account, a final list of 21 plant species (including the target weed) in six families was selected for host range testing of Kordyana sp. (Table 1).

The methods used in host range testing in Brazil, and the results of tests, are given in detail in three unpublished reports (Barreto et al. 2010; Barreto & Macedo 2011; Barreto 2012; Landcare Research 2012). Since these results are crucial to assessing the risks of Kordyana sp. attacking non-target plants, a brief summary is also provided here.

Kordyana sp. has an intimate relationship with its host and grows much better in living plant tissues than on artificial media. Researchers at Viçosa were able to grow it on artificial media, but the fungal structures produced in this way were unable to infect plants. Field observations showed the smut had no trouble moving from plant to plant on its own, so host range tests were designed to take advantage of this natural spread.

The 20 non-target plants listed in Table 1 were placed in a shade house with T. fluminensis plants that had become naturally infected with Kordyana sp. in the field. Tradescantia fluminensis plants propagated from material originating in New Zealand were also included as positive controls (i.e. the test would only be valid if the target plants became infected under the conditions provided). Tradescantia fluminensis plants developed symptoms of Kordyana sp. infection one month after they were placed near infected plants in the shade house, but none of the other test plants developed symptoms during 12 months of observation (Table 1). In addition to the 20 test plants required to define the host range of Kordyana sp. (Fowler 2007; Table 1), Barreto (2012) monitored a further 42 plant species of 24 families that were growing in the shade house. No Kordyana sp. disease symptoms were recorded on these plants either.

Meanwhile, a method was developed that allowed the spores of Kordyana sp. to be applied more directly to test plants. Leaves of infected T. fluminensis plants were collected and attached (with Vaseline) to the underside of a sheet of glass with the leaf surface pointing downwards so fungal spores could fall onto test plants below. This sheet of glass was then suspended over the test plants in a dew chamber where ideal conditions for infection were provided for 48 hours. Then plants were transferred to a greenhouse for observation. This direct method was applied to 14 species in the Commelinaceae family that had already been tested in the shade-house experiment.

The results were exactly the same as for the shade-house test: the Kordyana sp. only caused symptoms on T. fluminensis of New Zealand origin. These results demonstrate that under conditions either ideal for the fungus or similar to those it is likely to experience in the field, the yellow leaf spot is highly host specific, and unlikely to infect any species other than T. fluminensis. Given the inability of Kordyana sp. to infect closely related plants, and the taxonomic distance between T. fluminensis and New Zealand native plants, it is extremely unlikely that Kordyana sp. could damage any non-target native or ornamental plants in New Zealand. May 2012 EPA0160

15

Application to import for release or to release from containment new organisms

Both the outdoor and indoor host range tests used field-collected T. fluminensis plants that had been naturally infected by Kordyana sp. as a source of inoculum. Consequently, the tests were of a random subset of the Kordyana sp. population, not just a single strain of the smut. Since all of the material tested gave the same result (it was unable to infect anything other than T. fluminensis) it seems reasonable to conclude that the whole species Kordyana sp. is specific to the host species T. fluminensis. This application therefore seeks to import the whole species, not any particular strain. Note that host specificity at the strain level is not a universal characteristic of all pathogens. It is common in rust fungi and some genera of fungi that can reproduce asexually (e.g. Colletotrichum species) but the authors have no information that would suggest strain-specificity is a typical feature of fungi such as Kordyana sp. that belong to the Brachybasidiaceae family.

Competition with native fungi

Kordyana sp. will not out-compete native fungi on native foliage because:

It is specific to Tradescantia fluminensis (see Section 6.3), and It won‘t be found on native plants. Winks et al. (2003) surveyed the fungal flora of tradescantia foliage at 14 sites through New Zealand. Twenty-five fungi were isolated. Most were cosmopolitan saprophytic or opportunistic pathogens that are likely to be common on the foliage of a range of plants. A Colletotrichum species was present in just two of 14 sites, but a Phomopsis species was present at ten. The rate of colonisation of leaves was low (62%) suggesting that competition for leaf area between fungal species was not high (Winks et al. 2003). Both Colletotrichum and Phomopsis are genera that include both saprophytic and pathogenic species. If these are native species then they must have existed on plants other than T. fluminensis before that plant was introduced to New Zealand and therefore this exotic plant cannot be their only host. Consequently, competition on the foliage of T. fluminensis would be unlikely to cause a significant decline in the abundance of these indigenous fungi, and thus the risk of significant competitive displacement is unlikely.

Hybridisation with native species of fungi

There are no Kordyana species that are either native or naturalised in the New Zealand region (see Section 4.2 above). Nor are there any other fungi that belong in the same family as Kordyana (Section 4.2). Since hybridisation only occurs between close relatives, there is no possibility that Kordyana sp. would hybridise with a resident fungus.

Interactions with mycoparasites

Some fungi, known as mycoparasites, can attack other fungi. These often have a wide host range and some are known to occur in New Zealand. Hypothetically, the introduction of an additional species of smut could increase mycoparasite loads, which in turn could suppress other fungi, including native smuts. However, no mycoparasites have been found associated with this particular Kordyana sp. in Brazil, and the white smuts that occur in New Zealand, unlike many rusts, are not known to be particularly susceptible to attack from such fungi (Eric McKenzie, Landcare Research, pers. comm. 2012). For example, the white smut Entyloma ageratinae, which was introduced as a biocontrol agent against the weed mistflower (Ageratina riparia), has never been observed being attacked by mycoparasites in the field or in the laboratory (Jane Barton, pers. obs.). Note that if permission to May 2012 EPA0160

16

Application to import for release or to release from containment new organisms

release Kordyana sp. into New Zealand is granted then any material imported from Brazil will be checked for mycoparasites before leaving the country of origin, and again before being released from containment.

Risk that tradescantia is replaced with a more damaging weed

Tradescantia is regarded as a serious weed of the forest floor (Kelly & Skipworth 1984). Biological control is very likely to lead to the replacement of tradescantia by native or exotic species. The nature of the replacement vegetation will vary across habitats. In infested areas with moderate to high incident light, it is very likely that some replacements will be undesirable species. Studies indicate that, where it occurs, there are unlikely to be any weeds that are as damaging to biodiversity values as tradescantia (Landcare Research 2012). Overall, the vegetation assemblage following successful biological control is likely to support greater biodiversity and promote more natural ecosystem processes than a tradescantia mat.

6.4. Beneficial and adverse effects on human health Spores of fungi in the Exobasidiales are not known to cause allergic responses in humans or animals and are no more likely to do so than any other resident fungus, or pollens (Eric McKenzie, Landcare Research, pers. comm.; Landcare Research 2012).

6.5. Beneficial effects on society, communities, and the market economy The adverse environmental effects of tradescantia (Section 6.2) are so severe that management of this weed is a major focus for the Department of Conservation (DOC), territorial local authorities, NGOs and private landowners seeking to protect and enhance forest remnants and reserves nationwide. In itself, the introduction of Kordyana sp. would have no benefits to society or communities or the market economy. However, Kordyana sp. is expected to contribute significantly to biological control of tradescantia, and hence to:

Protect national conservation values Protect reserves near urban areas and other forest remnants Better apply scarce community volunteer effort for reserve maintenance Assist tradescantia management in home gardens Protect dogs from allergic responses.

Protection of national conservation values

Tradescantia is a serious problem in the national conservation estate. Tradescantia is not the subject of any weed- led projects and so it is not possible to isolate expenditure on tradescantia from other management costs on Conservation land. However, responses from DOC staff made it clear that there are large tracts of tradescantia that cannot be treated at present because of limited resources (Landcare Research 2012). The QEII National Trust indicated in 2007 that on 183 of 317 covenants registered to protect conservation values on private land, tradescantia infestations were considered to be of high or medium severity. Eradication or progressive

May 2012 EPA0160

17

Application to import for release or to release from containment new organisms

management was indicated on 240 covenants (Landcare Research 2012). The monetary costs of these infestations are not known because control devolves to landowners rather than the Trust.

Protection of urban reserves and other forest remnants

Conservation of regional biotas worldwide depends heavily on the retention and management of fragmented areas of the original vegetation. In New Zealand little of the pre-European landscape remains intact, but occur as forest remnants in heavily modified or peri-urban areas (Whaley et al. 1997; Smale & Gardner 1999). Those reserves close to towns have more weeds than those further away (Timmins & Williams 1991), often as a result of the dumping of garden rubbish (Sullivan et al. 2005). Tradescantia is often a major component of such garden waste and invasion into margins and along waterways is continuing (see comments in Landcare Research (2012)).

In 2007 Auckland City Council administered 200 bush and natural areas, and Te Ngahere Native Forest Management managed 29 of these. In a total area of 260 ha, Te Ngahere was actively managing 45 ha of tradescantia (17% of the area reserved). Te Ngahere staff treated tradescantia in 211 public reserves in the Auckland Region from 2005 to 2007 (D. Moverley, Landcare Research, pers. comm., 2012). If these reserves are to continue to conserve natural values, they will require regular visits to prevent the establishment of weeds (Timmins & Williams 1991). Future population pressure on such amenities will make this a difficult task without the sustained suppression that biological control provides.

Community volunteer effort for reserve maintenance

Maintaining the regional floristic and conservation values of community reserves in New Zealand (e.g. Anon. 1995) requires considerable human resources. It proved difficult to quantify the volunteer-effort directed at tradescantia in 2007, but the importance of tradescantia as a weed of national significance is reflected in the comments of those land managers who deal with this threat (Landcare Research 2012). Environmental grants from many regional councils and NGOs support numerous community restoration projects nationwide, many of which focus heavily on tradescantia. In 2007, respondents presented a number of case histories (Landcare Research 2012). Sixteen of these provided sufficient information to conservatively estimate the total cost of tradescantia management in the previous year to be over $100,000. This is a minimum estimate of the national cost because responses represent only a small proportion of the organisations and individuals involved in tradescantia management in New Zealand. Some examples are:

http://www.weedbusters.co.nz/awards/Nominee.asp?id=28

http://www.trelissickpark.org.nz/weeds.htm

The presence of other weeds within the reserves means that successful biological control would not achieve economic gains in a strictly market sense, but would allow reallocation and more effective use of the limited resources available for weed management on reserve land. Successful biological control would extend the amount of land under management, and unlike current methods, biological control would be self-sustaining and responsive to weed resurgence.

Better management of tradescantia in home gardens

May 2012 EPA0160

18

Application to import for release or to release from containment new organisms

Home gardeners are the sector of society most likely to benefit from the introduction of another biocontrol agent for tradescantia. This is one of the most persistent and difficult weeds in North Island gardens, and removal of living tradescantia mats, or even removal of dead material following herbicidal control, requires hard physical labour. Successful biological control would reduce the rate of biomass accumulated by tradescantia, and so reduce the time required to manage gardens. Tradescantia causes allergic responses in some gardeners, and biological control may reduce the incidence of this.

Protection of dogs from allergic responses

The leaves and stems of plants in the family Commelinaceae snap readily, exuding a mucilaginous sap that can attach to the skin of dogs and other pets to give prolonged exposure to its components. This can lead to allergic responses in dogs including acute dermatitis, pain and distress, and anaphylactic shock (Lee & Mason 2006). These symptoms are not uncommon in dogs in New Zealand, where the cause is contact with tradescantia (see Section 9 for allergic response references). Reduction in the frequency or size of tradescantia patches by moderate biological control would reduce the incidence of these distressing allergic responses (Landcare Research 2012).

6.6. Adverse effects on society, communities, and the market economy No significant adverse effects on society and communities or on the market economy have been identified. The results of host range tests summarised in Section 6.3 suggest that Kordyana sp. will only infect T. fluminensis. Closely related species used as ornamentals were tested and were not susceptible to attack, as were a wide range of economically important plant species (Barreto et al. 2010). It is extremely unlikely that any plants of ornamental or economic value in New Zealand will be attacked by this smut.

7. Could your organism(s) undergo rapid assessment?

If your application involves a new organism that is or is contained within a veterinary or human medicine, could your organism undergo rapid assessment (s38I of the HSNO Act)?

Describe the controls you propose to mitigate potential risks (if any). Discuss what controls may be imposed under the ACVM Act (for veterinary medicines) or the Medicines Act (for human medicines).

Discuss if it is highly improbable (after taking into account controls if any): the doses and routes of administration of the medicine would have significant adverse effects on the health of the public or any valued species; and the organism could form an undesirable self-sustaining population and have significant adverse effects on the health and safety of the public, any valued species, natural habitats or the environment.

Do not include effects of the medicine or new organism on the person or animal being treated with the medicine.

Not Applicable

If your application involves a new organism (excluding genetically modified organisms), could your organism undergo rapid assessment (s35 of the HSNO Act)?

Discuss if your organism is an unwanted organism as defined in the Biosecurity Act 1993.

May 2012 EPA0160

19

Application to import for release or to release from containment new organisms

Discuss if it is highly improbable that the organism after release: could form self-sustaining populations anywhere in New Zealand (taking into account the ease of eradication) could displace or reduce a valued species could cause deterioration of natural habitats, will be disease-causing or be a parasite, or be a vector or reservoir for human, animal, or plant disease will have adverse effects on human health and safety or the environment.

Not Applicable

8. Other information

Add here any further information you wish to include in this application including if there are any ethical considerations that you are aware of in relation to your application.

There are no ethical considerations to be addressed.

8.1. Section 36 minimum standards Displacement of native species

Kordyana sp. is a pathogen of plants, but is apparently specific to Tradescantia fluminensis. It cannot live on other hosts (see Sections 4 and 6.3) and so will not compete with fungi on plants other than T. fluminensis. Winks et al. (2003) did not find any fungi of significance growing on tradescantia in New Zealand, and so Kordyana sp. will not compete significantly on tradescantia leaves either. No mycoparasites (parasitic fungi whose hosts are another fungus) have been recorded on Kordyana sp. but two were found on another novel Kordyana species infecting Commelina sp. plants in Brazil (Prof. R. Barreto pers. comm.). Mycoparasites are present in New Zealand. Kordyana sp. could be susceptible to resident native or introduced mycoparasites in New Zealand, but this is unlikely to significantly inhibit the smut or build up in sufficient levels on the Kordyana sp. to present a danger to any fungi that are native to New Zealand. It is not known whether the net environmental effects of such hyperparasitism would be adverse or beneficial.

Deterioration of native habitats

Tests indicate that the agent will not attack native plants if released in New Zealand. Where tradescantia is currently abundant, the natural habitat and its processes have been degraded. Successful biological control would help to restore those habitats and their associated natural processes (such as regeneration) to a more natural state. Where the weed is less abundant, suppression of tradescantia build-up and dispersal will halt the decline in the integrity of natural habitats. No significant deterioration in habitat quality is envisaged from the introduction of this agent.

Adverse effects on human health and safety

No significant effects on human health and safety are expected from the introduction of this species to New Zealand.

Adverse effect to New Zealand’s inherent genetic diversity

May 2012 EPA0160

20

Application to import for release or to release from containment new organisms

Kordyana sp. has no close relatives in New Zealand (Section 4) so hybridisation is not possible. We consider that there is no reasonable risk that Kordyana sp. will evolve a wider host range in future. The evidence for these views is presented in Barton (née Fröhlich, 2004). No inseparable organisms have been recorded on Kordyana sp. in its native range.

Causing disease, being parasitic, or becoming a vector for disease

Kordyana sp. is a disease-forming organism but experimental evidence indicates that its only likely host is the weed Tradescantia fluminensis.

8.2. Post-release monitoring and measurement of impact Auckland Council and Landcare Research intend to monitor release sites for the establishment of Kordyana sp. They intend to measure the impact of Kordyana sp. and how it interacts with the invertebrate control agents already released. Standish et al. (2001) have already provided baseline estimates of tradescantia biomass from which to assess the effects of control agents, and assessment of the role of invertebrate control agents has begun. However, further planning is not justified until it is known whether Kordyana sp. will establish in New Zealand, and whether it will form damaging infections.

9. Appendices(s) and referenced material (if any) and glossary (if required)

The references cited in the application are provided in this section along with a peer-review report of the application‘s treatment of unpublished data.

9.1. References Allergic responses (accessed June 2012)

http://www.petvet.co.nz/news.cfm?content_id=42 ,

http://60.234.160.229:81/LinkClick.aspx?fileticket=ETgNtaieu9A%3d&tabid=92&mid=526 ,

http://www.albertanimalhospital.com.au/AllergyPlants.pdf

Anon. (1995). The Battle for Kitchener Park. New Zealand Geographic 26: 9-12.

Barreto RW (2012) Indirect host range test in shade house. Report. Viçosa, Brazil: Universidade Federal de Viçosa (Unpublished report in Landcare Research 2012).

Barreto RW, Macedo DM (2011) Biological control of Tradescantia fluminensis with pathogens. Report August 2011. Viçosa, Brazil: Department de Fitopatologia, Universidade Federal de Viçosa, Viçosa, Brazil (Unpublished report in Landcare Research 2012).

Barreto RW, Macedo DM, Pereira OL, Waipara NW, Dodd S (2010) Biological control of Tradescantia fluminensis with pathogens. Interim report August 2010. Departmento de Fitopatologia, Universidad Federal de Viçosa, Viçosa, Brazil, Landcare Research, Auckland, New Zealand (Unpublished report in Landcare Research 2012).

May 2012 EPA0160

21

Application to import for release or to release from containment new organisms

Barton (née Fröhlich) J (2004) How good are we at predicting the field host-range of fungal pathogens used for classical biological control of weeds? Biological Control 31: 99–122.

Barton J (2012) Predictability of pathogen host range in classical biological control of weeds: an update. BioControl 57: 289–305.

Biosecurity New Zealand (accessed 2012) http://www.biosecurity.govt.nz/nppa.

Brown D, Rees D (1995) Control of Tradescantia on Stephens Island. Ecological Management 3: 6–9.

Esler AE (1988) Naturalisation of plants in urban Auckland, New Zealand 5: Success of the alien species. New Zealand Journal of Botany 26: 565–584.

Fowler S (2007) Proposed test plant list for the New Zealand biological control programme against Tradescantia fluminensis. Report. Lincoln: Landcare Research (Unpublished report in Landcare Research 2012).

Fowler SV, Syrett P, Hill RL. (2000) Success and safety in the biological control of environmental weeds in New Zealand. Austral Ecology 25: 553–562.

Kelly D, Skipworth, JP (1984) Tradescantia fluminensis in a Manawatu (New Zealand) forest: I. Growth and effects on regeneration. New Zealand Journal of Botany 22: 393–397.

Landcare Research (2012) http://www.landcareresearch.co.nz/science/plants-animals- fungi/plants/weeds/biocontrol/approvals

Lee SE, Mason KV (2006) Immediate hypersensitivity to leaf extracts of fragrans (inch plant) in a dog. Veterinary Dermatology 17: 70–80.

Smale MC, Gardner RO (1999) Survival of a Mount Eden bush, and urban forest remnant in Auckland, New Zealand. Pacific Conservation Biology 5: 83–93.

Standish R (2001) Prospects for biological control of Tradescantia fluminensis Vell. (Commelinaceae). Doc Science Internal Series 9. Wellington, Department of Conservation.

Standish RJ (2002) Experimenting with methods to control Tradescantia fluminensis, an invasive weed of native forests remnants in New Zealand. New Zealand Journal of Ecology 26: 161–170.

Standish RJ (2004) Impact of invasive clonal herb on epigaeic invertebrates in forest remnants in New Zealand. Biological Conservation 116: 50–58.

Standish RJ, Robertson AW, Williams PA (2001) The impact of an invasive weed Tradescantia fluminensis on native forest regeneration. Journal of Applied Ecology 38: 1253–1263.

Standish RJ, Bennett SJ, Stringer IAN (2002) Habitat use of Tradescantia fluminensis by Powelliphanta traversi. Science for Conservation 195. Wellington, Department of Conservation. Pp. 1–26.

Standish RJ, Williams PA, Robertson AW, Scott NA, Hedderley DI (2004). Invasion of a perennial herb increases decomposition rates and alters nutrient availability in warm temperate lowland forest remnants. Biological Invasions 6: 71–81.

Sullivan JJ, Timmins SM, Williams P (2005) Movement of exotic plants into coastal native forests from gardens in northern New Zealand. New Zealand Journal of Ecology 29:1–10.

Timmins S, Williams PA (1991) Weed numbers in New Zealand forest and scrub reserves. New Zealand Journal of Ecology 15: 153–162.

May 2012 EPA0160

22

Application to import for release or to release from containment new organisms

Toft RJ, Harris RJ, Williams PA (2001) Impacts of the weed Tradescantia fluminensis on insect communities in fragments forests in New Zealand. Biological Conservation 102: 31–46.

Wapshere A (1974) A strategy for evaluating the safety of organisms for biological weed control. Annals of Applied Biology 77: 201-211.

Whaley PT, Clarkson BD, Smale MC (1997) Claudelands Bush: the dynamics of an urban kahikatea (Dacrycarpus dacrydioides) forest remnant. Tane 36: 131–155.

Winks CJ, Waipara NW, Gianotti AF, Fowler SV (2003) Invertebrates and fungi associated with Tradescantia fluminensis (Commelinaceae) in New Zealand. Landcare Research Contract Report LC0203/153.

Yeates GW, Williams PA (2001) Influence of three invasive weeds and site factors on soil microfauna in New Zealand. Pedobiologia 45: 367–383.

9.2. Peer review of unpublished information Dr. Stan Bellgard, Landcare Research, Auckland I have reviewed the information presented in two reports written by Professor Robert Barreto and co-authors, covering the biology, aetiology and pathology of Kordyana sp. I have also reviewed the EPA Application to be submitted for the importation of Kordyana sp. for the biocontrol of Tradescantia fluminensis (tradescantia). As requested by Dr Jane Barton and Dr Richard Hill, I have addressed the following specific questions:

1. Is Professor Barreto‘s research appropriate for the estimation of the host range of Kordyana sp.? 2. Do the two tables in the Application accurately summarise the host range testing data provided in the reports? 3. Does the text of the Application accurately portray the results summarised in the reports and tables? 4. Is the account of Kordyana sp. biology contained in section 4 of the Application an accurate interpretation of the technical data provided by Prof. Barreto?

Host range testing

The first of Prof. Barreto‘s reports (Barreto et al. 2010) documents the in-country surveys (conducted over five years) of tradescantia and the pathogens associated with diseases recognised in the field survey. This included the discovery of a third species of Kordyana (the current agent being assessed). They then examined the feasibility of culturing the various pathogenic agents, and commenced the indirect host-range-screening testing to assess the specificity of the agents. The results of these tests suggested that the new Kordyana sp. (brasiliensis) was pathogenic to the New Zealand biotype of tradescantia.

The second report of Barreto et al. (2011) further examined host-specificity and also attempted to develop an inoculation methodology, which utilised inoculum from infected tradescantia leaves. This was indeed successful, with symptoms being observed on T. flumenensis but not any other species after 18 days. There were no observed non-target impacts after 65 days on any other species.

This research also suggested that the importation of the biocontrol agent may need to be facilitated using infected plant material. Barreto et al. consider that the infected plant material could be hand-couriered as part of an aeroplane consignment directly from Brazil to New Zealand.

From the information provided it is considered that estimation of host range of Kordyana sp. is adequate.

May 2012 EPA0160

23

Application to import for release or to release from containment new organisms

Table in Application

The table in the Application accurately conveys the data provided by Prof. Barreto.

Text of the Application

The text of the Application accurately conveys the results contained in the technical reports provided by Prof. Barreto.

Account of biology in Section 4 of the Application

Section 4 provides an accurate and well-written summary of the biology of Kordyana sp.

Post-release monitoring should be considered in partnership with the regional biosecurity staff, especially in relation to the combined (synergistic) effects of insect and pathogen biocontrol agents.

10. Signature of applicant or person authorised to sign on behalf of applicant

I request the Authority to waive any legislative information requirements (i.e. concerning the information that shall be supplied in my application) that my application does not meet (tick if applicable).

I have completed this application to the best of my ability and, as far as I am aware, the information I have provided in this application form is correct.

Signature Date

May 2012 EPA0160

Table 1. Summary of host specificity tests conducted in Viçosa, Brazil. The full results can be found in unpublished reports (Barreto et al. 2010; Barreto & Macedo 2011; Barreto 2012) on the Landcare Research website (Landcare Research 2012). There were three replicates in the shade-house test, and two replicates in each of two spore-drop tests. Status information is from http://nzflora.landcareresearch.co.nz/ (accessed May-June 2012) unless otherwise stated in the text. Authorities for plant names can be found in Barreto (2012, in Landcare Research 2012). A dash indicates no test.

Order Family Relationship Plant species Common name Status in NZ Disease Disease to target weed symptoms in symptoms in shade-house test spore drop test

Commelinales Commelinaceae Target weed Tradescantia fluminensis Wandering Jew Target weed Yes Yes

Same genus Tradescantia pallida Purple heart Absent No No Tradescantia spathacea Moses in a basket Absent No No Tradescantia zebrina Inch plant In cultivation No No Tradescantia zononia Absent No No

Same sub-tribe Callisia repens Baby‘s tears In cultivation No No Callisia warszewicziana Absent No No Tripogandra diuretica Absent No No Gibasis schiediana Bridal veil Naturalised No No

Same tribe Dichorisandra thyrsiflora Blue ginger Present in cultivation No No Siderasis fuscata Absent No No Tinantia sp. Absent No -

Same family Commelina benghalensis Status not given No No Commelina diffusa Present in cultivation No No Commelina erecta Absent No No

Pontederiaceae Same order Eichhornia crassipes Water hyacinth Naturalised No -

Poales Cyperaceae Same class Cyperus rotundus Purple nutsedge Naturalised No -

Juncaceae Juncus spp Rushes Some spp. native No -

Zingiberales Strelitziaceae Strelitzia reginae Bird of paradise Present in cultivation No -

Zingiberaceae Hedychium coronarium White ginger Present in cultivation No - Zingiber officinale Edible ginger Present in cultivation No -