Accepted Manuscript

Review

Benefits from biological control of weeds in New Zealand range from negligible

to massive: A retrospective analysis

David Maxwell Suckling

PII: S1049-9644(13)00047-9

DOI: http://dx.doi.org/10.1016/j.biocontrol.2013.02.009

Reference: YBCON 2907

To appear in: Biological Control

Received Date: 2 November 2012

Accepted Date: 27 February 2013

Please cite this article as: Suckling, D.M., Benefits from biological control of weeds in New Zealand range from

negligible to massive: A retrospective analysis, Biological Control (2013), doi: http://dx.doi.org/10.1016/

j.biocontrol.2013.02.009

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers

we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and

review of the resulting proof before it is published in its final form. Please note that during the production process

errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

Benefits from biological control of weeds in New Zealand range from negligible to

massive: A retrospective analysis

David Maxwell Suckling

The New Zealand Institute of Plant & Food Research Ltd, PO Box 4704, Christchurch 8140,

New Zealand

Email: Max.Suckling@plantandfood.co.nz

Ph+64-3-977344

Fax +64-3-3252074

Abstract

Emerging concern highlighting non-target impacts in classical biological control of

arthropods and weeds has heightened awareness of these risks but raised the risk of obscuring

beneficial effects. This review applied a retrospective assessment of the benefits from weed

biological control in New Zealand, using the framework designed for pre-clearance

assessment of classical biological control. Of those agents released which can be assessed

because of sufficient passage of time (n=33), their impact has been assessed according to the

modern criteria for judging beneficial effects used by New Zealand’s Environmental

Protection Authority (negligible, minimal, minor, moderate, major and massive). Cases with

negligible benefit (n=12) included failures to establish self-sustaining populations, while

cases with minimal benefit (n=11) included some where predation reduced the realised

benefit of established organisms. The remaining cases offered massive (n=2), major (n=1),

moderate (n=5) or minor (n=2) benefit. Suppression of ragwort (Jacobaea vulgaris Gaertn.

(1754), and St Johns wort (Hypericum perforatum L.) were considered to be massive in

magnitude, offering long term ecosystem benefits of controlling invasive weeds. Improved

clarity around risk and benefit could help improve the quality of debate on biological control,

and the five step scale used in New Zealand may prove more widely useful elsewhere.

2

Keywords: Biological control, weeds, non-target, risk, benefit, ecosystem

1 INTRODUCTION

Invasive weeds can be considered to be the most significant of the economic and

environmental pests (Pimentel et al., 2001; Hulme, 2012). Weeds are implicated in species

endangerment (Wilcove et al., 1998) and threaten sustainability in production ecosystems

(Swanton and Murphy, 1996). They typically require removal by labour, herbicide use or

biological control (McFadyen, 1998). All interventions carry their own costs and risks, and

herbicides in particular are being scrutinised for environmental impact in many countries

(Moss, 2010). Weed biological control using arthropods can also be considered to be risky

because of unexpected non-target effects (Simberloff, 2011a). In recent years, there has been

a major upsurge in the concern about non-target risks and impacts from previous biological

control introductions against vertebrates, arthropods, and weeds (Simberloff and Stiling,

1996; Louda et al., 2003; Simberloff, 2011b). Although all non-target ecological effects are

unwanted, effects from weed biological control agents appear to be of much smaller

magnitude than ecological problems involving other forms of biological invasions

(Ehrenfeld, 2010), including ill-conceived deliberate vertebrate introductions (Russell et al.,

2004).

There may be a risk of confounding the ecological consequences of introductions of different

types of organism from different trophic levels, and it may not be informative to generalise

from the consequences of past introductions of vertebrates or even arthropod biological

control agents to weed biological control. By 1998, there were 352 introductions for weed

biological control globally, of which 0.6% (two agents) were known to have had significant

negative impacts on other species (Sheppard et al., 2003). Recognition that release of a thistle

weevil was affecting native thistle biodiversity (Louda et al., 1997; Gassmann and Louda,

2001), while important, was not foreseen as a threat to native biodiversity and represents a

3

change in social attitudes to indigenous species rather than a failure of host range prediction

(Pemberton, 2000). The introduction of the cactus moth to the Caribbean, which led to its

arrival and impact on indigenous cacti in the genus Opuntia in southern North America, is

also frequently quoted in this context (Simberloff and Stiling, 1996; Simberloff, 2011b). Here

the failure seems to have been a lack of prediction of insect movement (Pemberton and Liu,

2007), but the consequences appear dire and inevitable for the Mexican cactus ecosystems,

despite attempts to stop the spread of the moth (Bloem, 2007). There are few other examples

of ecological damage of this magnitude in weed biological control, although others may

emerge in future, given increased scrutiny.

The notions of acceptable risk, and risk of action or inaction merit close consideration

(McCoy and Frank, 2010). In New Zealand, cases with non-target effects have been studied

in an attempt to determine whether host range testing is adequate for predicting ecological

outcomes (Paynter et al., 2004; Withers et al., 2008), and the results do not show significant

adverse impacts on non-target plant populations. Cases without known non-target effects are

being re-examined to determine whether host range testing methodologies would predict a

lack of impacts in the field (Groenteman et al., 2011).

Risk assessment of new organisms is a critical part of the search for sustainable biological

solutions for invasive weeds (Sheppard et al., 2003), and improvements are being made to

methodologies (Simberloff, 2011b), but there is a risk of losing the balance between risk and

reward. This is especially true where the number of weed targets and their degree of impact is

large (Williams and Timmins, 2002). Unexpected non-target feeding does not have to be

biologically significant to have a significant impact on perceptions of biological control, as

can be seen from a recent Australian case. Palmer et al. (2004) described how a highly visible

membracid, introduced for control of lantana, had an apparent seasonal population explosion,

appearing on other plants including one common garden tree. It raised significant concern

4

from the urban public in Brisbane. It was quickly recognised that the insect was in fact only

resting on many of these plants and entomologists found that the insect could not develop on

these alternative plants. However, Palmer et al. (2004) noted that because the urban public

became concerned about the apparent non-target attack to one garden tree, it was inevitable

that those charged with deciding whether agents are safe to release in Australia would be

sensitive to the issue.

Given the increasingly evident adverse environmental impacts of herbicides, and legislative

responses to these (Moss, 2010), it could be prudent for countries to review the issues

limiting their use of biological control of weeds, from perceived risks of non-target impacts.

Although it is possible to estimate the costs of weeds at a given point in time (Pimentel et al.,

2001), estimation of the benefits forgone from failure to make sufficient introductions of

successful biological control agents would challenge any economist. Similarly, estimation of

the marginal benefits of biological control in preventing a situation from worsening can be

difficult. However, the estimation of the benefits of preventing a weed situation from

worsening has been done in several African studies (McConnachie et al., 2004; Van Wilgen

et al., 2004; Van Wilgen and De Lange, 2011). Identification of net benefits from biological

control successes is more readily achieved after a degree of stability which can take decades

(Fowler et al., 2000; Page and Lacey 2004; Sheppard et al., 2003; Hoffmann and Moran,

2007; Chalak-Haghighi et al., 2008).

The approval process for release of new organisms varies between jurisdictions (Sheppard et

al., 2003), but most countries appear to have cumbersome and protracted processes with few

organism considerations per decade. New Zealand is a country with a pioneering role in

arthropod and weed biological control (Cameron et al., 1993), where it has been long

recognised that weeds have a detrimental impact on natural and productive ecosystems

(Williams and Timmins, 2002; Bourdôt et al., 2007). New Zealand’s long history of

5

biological control (Cameron et al., 1989) has been updated and summarised in a database of

all introductions that can be readily accessed from the internet (Ferguson et al., 2007). These

features have combined with progressive risk management legislation to produce the basis for

analysis of the benefits from biological control of weeds from previous introductions.

In New Zealand, the Hazardous Substances and New Organisms Act (1996) provides the

basis for a balanced approach between risks, costs, and benefits on a case by case basis. It

uses risk management principles (EPA, 2011) and a quasi-judiciary reference panel to

consider applications (Sheppard et al., 2003). Overall, the regime is considered to have been

beneficial to weed biological control introductions (Fowler et al., 2010). However, one of the

main problems experienced by the regulatory authority with applicants is that they seldom

adequately describe the benefits. There are many reasons for this, including uncertainty about

establishment and performance of agents, as well as difficulty in providing economic costs

for environmental weeds where benefits in biodiversity may occur. This paper reports a

simple analysis of the benefits from previous releases of weed biological control agents,

using the risk assessment methodology adopted by the New Zealand regulatory agency. The

methodology has been used for risk assessment of new organism applications by the author,

as a member of the board of the Environmental Risk Management Authority of New Zealand

(2003-2011), but has not been used previously for retrospective analysis. Restructuring in

July 2011 accounts for the name change to the Environmental Protection Authority.

2 METHODS

Beneficial effects considered to be above a negligible level were compared across a five-step

scale from minimal to massive, based on the scale in use by New Zealand’s Environmental

Protection Authority prior to any approval to release new organisms (EPA, 2011)(Table 1).

For biological control introductions, it is most pertinent to consider environmental, economic

and social effects, since health effects are less likely to be important (although there could be

6

exceptions). The scale uses verbal descriptors as guidelines for each type of benefit, which

helps to clarify different effects and avoid double accounting in any economic analysis across

environmental, economic and social areas. Dollar values have been assigned on a logarithmic

scale, in order to match the divergence of effects, from minimal to massive. Normally this

scale is used for weighing costs against benefits of a proposed new introduction at the same

time, so units cancel each other out. These per annum values aim to provide a guide only, in

order to indicate the logarithmic scale used for these categories. Community benefits due to

job creation would be included under the economic benefits area, while reduced stress of

farmers might be considered a social benefit, although both could result from the same new

organism introduction. Benefits may occur in one or more areas, but in the case of

environmental weeds it is considered sufficient to have benefits in the environmental benefits

area (such as mist flower, see below). In other cases, impact on pasture weeds (Bourdôt et al.,

2007) can be assigned a value in the economic benefit area.

Table 1 here

A total of 23 weed species have been targeted in New Zealand (Cameron et al., 1989;

Ferguson et al., 2007), since biological control activity began in the 1920s. The cases derived

from the BCANZ database (Ferguson et al., 2007) were individually assessed from available

literature and assigned to a category according to Table 1. About half the cases are thought to

have had negligible effects on the target weeds (Paynter et al., 2010a). While 82 organisms

were investigated, some organisms were not released (n=32), or are pending the outcome of

testing, and 16 cases are considered too recent to assess (mostly under 20 years since release).

One case was self-introduced (Cercospora eupatorii (Capnodiales: Mycosphaerellaceae) and

excluded. Some organisms were imported on multiple occasions, but only the latest record

was used for each species. There is a paucity of solid information on many cases, although

results from 14 species have been summarised previously (Bourdôt et al., 2007), which was

7

taken into account in the assessment. The generic analytical framework and ecological

descriptors used were not specifically designed for weed biological control, and the attempt

to categorise the organism cases raised some challenges to turn evident impact on weed

populations (e.g. reduced percentage cover) into one of the five impact categories. These

cases were addressed as carefully as possible.

3 RESULTS

Filtering left 33 arthropods that could be fairly evaluated (Table 2). A peak in activity has

emerged in the last 30 years (Fig. 1), and a number of organisms are under evaluation before

potential release.

Fig. 1 here

In some cases, the benefits of multiple organism introductions against a weed may be hidden

by one particularly obvious organism, so it is important to recognise that stress factors may

accumulate to affect plants, including the burden of several agents acting in concert. Four

high profile cases studies have been chosen to illustrate the general approach.

Table 2 here

3.1 Nodding thistle crown weevil Trichosirocalus horridus (Coleoptera: Curculionidae)

Jessep (1989) reported that nodding thistle Carduus nutans L. had emerged over previous

decades as a serious agricultural weed affecting thousands of hectares, especially in summer

dry areas where pasture depletion caused by overgrazing was exacerbated by pasture pest

outbreaks. The nodding thistle crown weevil, introduced to New Zealand in 1982, is

considered to have caused a decline in nodding thistle populations in a number of areas

(Groenteman, 2010), although its impact may be augmented by other agents. Thistle

populations appear to have declined about five years after the weevils were released at many

sites. It also attacks other six exotic thistle species to a lesser extent. It is reported to reduce

8

the density of O. acanthium L. locally in Central Otago. Many regions of New Zealand still

list nodding thistle as a target of regional pest management strategies (Giera and Bell, 2009),

indicating either the residual importance of the problem despite biological control, or a lack

of updating of their records, since there is a suggestion that the need for alternative control

tactics has reduced (Bourdôt et al., 2007). An additional agent Apion onopordi (Kirby)

(Coleoptera: Brentidae) has been released (2003, 2007), since the release of T. horridus. The

impacts on nodding thistle populations from T. horridus are therefore considered here to be

moderate in magnitude (Table 1), as the decline in this thistle has been localised.

3.2 Ragwort flea beetle Longitarsus jacobaeae (Coleoptera: Chrysomelidae)

This beetle was released in New Zealand in 1981 against ragwort, Jacobaea vulgaris Gaertn.

(1754), syn. Senecio jacobaea L. (Syrett et al., 1984). Reductions of 90–100% in plant

density have been recorded at many sites (Gourlay et al., 2005). Gourlay (2007) reported that

ragwort disappeared after four years at a site on Otago, while 98% reductions were attributed

to the beetle at another site in the region. At sites further north, a 94% reduction in ragwort

plants was evident after only two years, and elsewhere a 90% reduction was evident after

three years, leaving only widely scattered plants after four years. There are some areas where

the flea beetle has not been successful. The main exceptions to the successful establishment

of the flea beetle and the reduction in ragwort populations are on the West Coast and southern

regions of the South Island, and in western parts of the North Island, where the insect appears

to be limited by high rainfall (Gourlay et al., 2005). A formal economic assessment has not

yet been done for New Zealand, but could be close to US$5M/y (S. Fowler pers. comm.). In

Oregon a similar program cited by Gourlay (2007) claimed regional benefits of US$5M/y due

to reduced livestock losses, increased pasture productivity and reduced herbicide use. The

cost of biological control (US$5/ha) reportedly gave a 13:1 return on investment there. The

impacts on ragwort are considered here to be massive (Table 1) as the beetle has contributed

9

to the control of ragwort throughout much of New Zealand where the weed has been

established as a pest.

3.3 Lesser St John’s wort beetle Chrysolina hyperici (Coleoptera: Chrysomelidae)

Chrysolina hyperici (Forster) was introduced into New Zealand for biological control of St

John’s wort, Hypericum perforatum L. in 1943, which had become a major pasture weed, in

places the dominant plant over many hectares (Hancox et al., 1986). The beetles occasionally

break out into very large populations, causing complete defoliation of host plants. There are

no longer reports of areas in New Zealand where St. John's wort is a problem, and Syrett

(1997) concluded that successful biological control with insects was as least partially

responsible for this change. Retrospective host range testing was been conducted, suggesting

some risk of non-target impacts, but this has never been recorded in the field (Groenteman et

al., 2011).The benefit is considered to be massive, because there has been a measurable long-

term benefit in control of St John’s wort in a type of ecosystem that is widespread in New

Zealand. The weed is not now listed on regional pest management strategies (Giera and Bell,

2009).

3.4 White smut fungus Entyloma ageratinae on Mist Flower (Ageratina riparia)

The white smut fungus Entyloma ageratinae R. W. Baretto and H. C. Evans (Tilletiales:

Tilletiaceae), developed damaging infestations on the environmental weed mist flower at all

nine of the sites where it was released in November–December 1998. It spread rapidly and

thoroughly across apparently all the North Island areas where mist flower occurs. It reduced

the mean percentage cover of mist flower from 81% to 1.5% over 5 years (Barton et al.,

2007). This decline was accompanied by increased species richness and percentage cover of

native plants, with only a weak ‘replacement weed’ effect. Maximum mist flower plant

height also declined significantly. The beneficial effect of mist flower biological control is

seen as major, because of the long-term benefit to localized ecosystems.

10

3.5 Distribution of Benefits

Benefits, as outlined in Table 1, were matched to the regulatory agency criteria for minimal

to massive degrees of magnitude, with declining frequency occurring with higher magnitudes

of benefits (Fig. 2). Two cases were found that matched the criteria for massive benefit

having been achieved, with sustainable long-term reductions in the target pest. These were C.

hyperici (Forster), the lesser St John’s wort beetle and L. jacobaeae (Waterhouse), the

ragwort flea beetle. The mist flower fungus E. ageratinae discussed above, offered a major

benefit, while there were five cases of moderate benefit from release of weed biological

control agents (Agasicles hygrophila Selman & Vogt (Coleoptera: Chrysomelidae), alligator

weed beetle; Trichosirocalus horridus (Panzer) (Coleoptera: Curculionidae); Arcola malloi

(Pastrana) (Lepidoptera: Pyralidae) = Vogtia malloi Pastrana, alligator weed moth, nodding

thistle crown weevil; Lochmaea suturalis (Thomson) (Coleoptera: Chrysomelidae), heather

beetle and Urophora solstitialis (L.) (Diptera: Tephritidae), nodding thistle gall fly).

Benefits were assessed as minor for two additional agents (Mexican devil weed gall fly

Procecidochares utilis Stone (Diptera: Tephritidae), and broom seed beetle (Bruchidius

villosus (Fabricius) (Coleoptera: Chrysomelidae) = Bruchidius ater, Bruchidius fasciatus).

The latter case could warrant an upgrade in future, if predictions are correct for increased

impact on seed rain from reduced pollination by wild bees because of mortality from varroa

mite (Paynter et al., 2010b).

Minimal benefits were found for an additional twelve cases, with only localised short-term

(such as seasonal) impact. These included cases cited by Bourdôt et al. (2007) with low to

occasionally high damage to plants or plant parts but no significant suppression of weed

populations (Tyria jacobaeae (Linnaeus), Exapion ulicis (Forster), Tetranychus lintearius

Dufour, Cydia succedana (Denis & Schiffermüller), Zeuxidiplosis giardi (Kieffer),

Rhinocyllus conicus Fröhlich, Urophora cardui (Linnaeus)).

11

A further two species were assigned to negligible, as they were cited by Bourdôt et al. (2007)

as having insignificant damage to individual plants and plant parts (agent usually too rare)

and no impact on weed populations, Botanophila jacobaeae Hardy (Diptera: Anthomyiidae),

and Chrysolina quadrigemina (Suffrian), Coleoptera: Chrysomelidae). In addition, gorse

thrips (Sericothrips staphylinus (Haliday), Thysanoptera: Thripidae) and old man’s beard leaf

fungus or Phoma leaf spot (Phoma clematidina (Thumen) Boerema Pleosporales:

Pleosporaceae) were assigned to negligible benefit on the basis of lack of evident impact.

The benefits were found to be negligible for 12 species that had been released, including

eight that failed to establish self-sustaining populations. The poor establishment of insect

biological control agents has been recognised in the past (Cameron et al., 1993), and is an

obvious area for improvement. Fowler et al. (2010) reported that the establishment rate for

weed biological control agents had increased from 44 to 78%, partly as a result of research on

propagule release number, but there is room for further improvement. Most of the costs of an

expensive process will have been incurred by the time of release, so it makes sense to

maximise establishment.

Fig. 2 here

4 DISCUSSION

While notoriously difficult because of the need for many assumptions, economic analysis of

the cost of weeds to New Zealand has been attempted, and includes defensive investment and

production losses to the pastoral economy of the order of $1.2B/y (Bourdôt et al., 2007; Giera

and Bell, 2009). Where individual cases have been analysed, the market economy benefit

appears to align with the nominal dollar ranges in Table 1 (Anonymous, 2012). A simple

summation of the market economy benefits from the distribution of magnitudes of successful

biological control cases (Fig. 2), based on the nominal dollar ranges in Table 1, would

suggest net benefits in the range of $11-217M pa, but there is no way to corroborate or add

12

precision to this estimate without better information. The comparison does suggest a positive

cost-benefit ratio. The analysis of cases where weeds are worsening warrants consideration,

along with a large number of sleeper weeds that are emerging (Bourdôt et al., 2007). Some

analysis of this is warranted where weed distributions can be tracked. This analysis does not

take into account worsening of weed problems over time, which requires estimation of

benefits from successfully forestalling such exacerbating developments.

5 CONCLUSIONS

Looking ahead, it is clear that the issues raised by advocates for increased caution in classical

biological control (McCoy and Frank, 2010; Simberloff, 2011b) need to be addressed. New

Zealand, which operates under the precautionary principle for new organism introductions,

has been able to achieve significant net value while managing the risks from weed biological

control agents adequately, with few exceptions. While the precautionary principle may fall

short of providing a guide for action (McCoy and Frank, 2010), it does not have to impede

progress. With greater understanding of the causes of non-target effects in weed biological

control, including improved host range testing (Paynter et al., 2004), ecosystem-level effects

from introduced species through subtle mechanisms such as apparent competition between

organisms (Carvalheiro et al., 2008), or reduced benefit from predation of agents (Paynter et

al., 2010a), will come a greater emphasis on predicting such effects before releases are made.

Additional research to reduce the proportion of releases that establish but make no

contribution to suppressing the target is also needed (Fowler et al., 2010). Much of the risk in

classical biological control of weeds appears to be in failure to achieve benefits rather than

from unpredicted non-target impacts, which can be managed by careful adherence to best

practice (Paynter et al., 2004). Improved clarity around risk and benefit could help to improve

the quality of debate on biological control, and the five-step scale used in New Zealand may

prove more widely useful elsewhere. Biological control programmes are long-term efforts at

13

a geographical scale, and appropriate analysis at this scale is needed to identify risks,

successes and failures in correct proportion. Increased attention to the risks should be

balanced by increased attention to the benefits of biological control.

ACKNOWLEDGMENTS

This assessment was conducted during a fellowship supported by the Organisation for

Economic Cooperation and Development (Cooperative Research Programme: Biological

Resource Management for Sustainable Agricultural Systems) at the European Biological

Control Laboratory (USDA-ARS), Montpellier. Dr Simon Fowler (Landcare Research) and

Dr Graeme Bourdôt (AgResearch) provided helpful comments on the manuscript. Landcare

Research kindly gave permission to use the before and after photos in the graphical abstract.

The author was a member of the board of the Environmental Risk Management Authority of

New Zealand (2003-2011), and a signatory on decisions to approve or decline many new

organism introductions over this period. He was on study leave from the Environmental

Protection Authority at the time of writing this review.

REFERENCES

Anonymous, 2012. Saving biodiversity and money: why biocontrol is so effective. Available:

<http://www.landcareresearch.co.nz/publications/casestudies/case_details.asp?Highlig

ht_ID=14>.

Barton, J., Fowler, S.V., Gianotti, A.F., Wilks, C.J., de Beurs, M., Arnold, G., Forrester, G.,

2007. Successful biological control of mist flower (Ageratina riparia) in New

Zealand: Agent establishment, impact and benefits to the native flora. Biological

Control 40, 370-385.

Bloem, K.A.B., S.,Carpenter, J. E., Hight, S., Floyd, J., Zimmermann, H., 2007. Don't let

cacto blast us: development of a bi-national plan to stop the spread of the cactus moth

Cactoblastis cactorum in North America. In: Vreysen, M.J.B.R., A. S., Hendrichs, J.

(Ed.) Area-wide control of insect pests. From research to field implementation.

Springer, Dordrecht, The Netherlands, pp. 337-344.

Bourdôt, G.W., Fowler, S.V., Edwards, G.R., Kriticos, D.J., Kean, J.M., Rahman, A.,

Parsons, A.J., 2007. Pastoral weeds in New Zealand: Status and potential solutions.

New Zealand Journal of Agricultural Research 50, 139-161. doi:

10.1080/00288230709510288.

14

Cameron, P.J., Hill, R.L., Bain, J., Thomas, W.P., 1989. A review of biological control of

invertebrate pests and weeds in New Zealand 1874-1989., CAB International,

Wallingford, U.K., 424 pp.

Cameron, P.J., Hill, R.L., Bain, J., Thomas, W.P., 1993. Analysis of importations for

biological control of insect pests and weeds in New Zealand Biocontrol Science and

Technology 3, 387-404.

Carvalheiro, L.G., Buckley, Y.M., Ventim, R., Fowler, S.V., Memmott, J., 2008. Apparent

competition can compromise the safety of highly specific biocontrol agents. Ecology

Letters 11, 690-700.

Chalak-Haghighi, M., Van Ierland, E.C., Bourdôt, G.W., Leathwick, D., 2008. Management

strategies for an invasive weed: A dynamic programming approach for Californian

thistle in New Zealand. New Zealand Journal of Agricultural Research 51, 409-424.

doi: 10.1080/00288230809510471.

Ehrenfeld, J.G., 2010. Ecosystem consequences of biological invasions. Annual Review of

Ecology Evolution and Systematics 41, 59-80.

EPA, 2011. Decision Making: A Technical Guide to Identifying, Assessing and Evaluating

Risks, Costs and Benefits. Enviromental Protection Authority, Wellington, New

Zealand, p. 56pp.

Ferguson, C.M., A., M., Barratt, B.I.P., Hill, R.L., Kean, J.M., 2007. BCANZ - Biological

Control Agents introduced to New Zealand. Available from <http://b3.net.nz/bcanz>.

Fowler, S.V., Paynter, Q., Hayes, L., Dodd, S., Groenteman, R., 2010. Biocontrol of weeds in

New Zealand: an overview of nearly 85 years. In: Seventeenth Australasian Weeds

Conference "New Frontiers in New Zealand: Together we can beat the weeds"

Christchurch, New Zealand.

Fowler, S.V., Syrett, P., Hill, R.L., 2000. Success and safety in the biological control of

environmental weeds in New Zealand. Austral Ecology 25, 553-562. doi:

10.1046/j.1442-9993.2000.01075.x.

Gassmann, A., Louda, S.M., 2001. Rhinocyllus conicus: initial evaluation and subsequent

ecological impacts in North America, Wajnberg, E., Scott, J.K., Quimby, P.C. (Eds.),

147-183 pp.

Giera, N., Bell, B., 2009. Economic cost of pests to New Zealand, MAF Biosecurity New

Zealand Technical Paper No: 2009/31. Nimmo Bell, Wellington, New Zealand, p. 80.

Gourlay, A.H., Fowler, S.V., Rattray, G., 2005. Abundance of Ragwort Flea Beetle

(Longitarsus jacobaeae) at Five Sites on the West Coast, South Island, New Zealand.

LC0405/131. Landcare Research Lincoln.

Gourlay, H., 2007. Ragwort Flea Beetle. Available from

<http://www.landcareresearch.co.nz/research/biocons/weeds/book/documents/Ragwor

t_Flea_Beetle.pdf>.

Groenteman, R., 2010. Understanding the implications of size differences in nodding thistle

crown weevil, Trichosirocalus horridus, for the successful biological control of

nodding thistle (Carduus nutans). Landcare Research Lincoln, New Zealand.

Groenteman, R., Fowler, S.V., Sullivan, J.J., 2011. St. John’s wort beetles would not have

been introduced to New Zealand now: A retrospective host range test of New

Zealand’s most successful weed biocontrol agents. Biological Control 57, 50-58. doi:

10.1016/j.biocontrol.2011.01.005.

Hancox, N.G., Syrett, P., Scott, R.R., 1986. Biological control of St. John’s wort Hypericum

perforatum in New Zealand: A review. Plant Protection Quarterly 1, 152-155.

Hoffmann, J.H., Moran, V.C., 2007. Assigning success in biological weed control: what do

we really mean? In: Bargeron, C.T., Wallace, R.D., Sumner, K., Markin, G.P.,

15

Reardon, R.C. (Eds.), XII International Symposium on Biological Control of Weeds;

USDA Forest Service. Morgantown, WV. FHTET-2011-03. .

Hulme, P.E., 2012. Weed risk assessment: a way forward or a waste of time? Journal of

Applied Ecology 49, 10-19. doi: 10.1111/j.1365-2664.2011.02069.x.

Jessep, C.T., 1989. Introduction of the crown weevil (Trichosirocalus horridus) as an

additional biocontrol agent against nodding thistle Proceedings of the NZ Weed and

Pest Control Conference 42, 52-54.

Louda, S.M., D., K., Connor, J., Simberloff, D., 1997. Ecological effects of an insect

introduced for the biological control of weeds. Science 277, 1088-1090.

Louda, S.M., Pemberton, R.W., Johnson, M.T., Follett, P.A., 2003. Nontarget effects—the

Achilles' heel of biological control? Retrospective analyses to reduce risk associated

with biocontrol introductions. Annual Review of Entomology 48, 365-396. doi:

doi:10.1146/annurev.ento.48.060402.102800.

McConnachie, A.J., Hill, M.P., Byrne, M.J., 2004. Field assessment of a frond-feeding

weevil, a successful biological control agent of red waterfern, Azolla filiculoides, in

southern Africa. Biological Control 28, 25-32.

McCoy, E.D., Frank, J.H., 2010. How should the risk associated with the introduction of

biological control agents be estimated? Agricultural and Forest Entomology 12, 1-8.

doi: 10.1111/j.1461-9563.2009.00446.x.

McFadyen, R.E.C., 1998. Biological control of weeds. Annual Review of Entomology 43,

369-393.

Moss, S.R., 2010. Integrated weed management (IWM): will it reduce herbicide use ?

Communications in Agricultural and Applied Biological Sciences 75, 9-17.

Page, A.R. and Lacey, K.L. 2006. Economic Impact Assessment of Australian Weed

Biological Control. Technical Series no.10, CRC for Australian Weed Management,

Adelaide, Australia. 151 pp.

Palmer, W.A., Day, M.D., Dhileepan, K., Snow, E.L., Mackey, A.P., 2004. Analysis of the

non-target attack by the lantana sap-sucking bug, Aconophora compressa and its

implications for biological control in Australia. In: Proceedings of the 14th Australian

Weeds Conference. Weed Society of New South Wales, Sydney, pp. 341-344.

Paynter, Q., Fowler, S.V., Gourlay, A.H., Groenteman, R., Peterson, P.G., Smith, L., Winks,

C.J., 2010a. Predicting parasitoid accumulation on biological control agents of weeds.

Journal of Applied Ecology 47, 575-582.

Paynter, Q., Fowler, S.V., Gourlay, A.H., Haines, M.L., Harman, H.M., Hona, S.R., Peterson,

P.G., Smith, L.A., Wilson-Davey, J.R.A., Winks, C.J., Withers, T.M., 2004. Safety in

New Zealand weed biocontrol: a nationwide survey for impacts on non-target plants

New Zealand Plant Protection 57, 102-107

Paynter, Q., Main, A., Hugh Gourlay, A., Peterson, P.G., Fowler, S.V., Buckley, Y.M.,

2010b. Disruption of an exotic mutualism can improve management of an invasive

plant: varroa mite, honeybees and biological control of Scotch broom Cytisus

scoparius in New Zealand. Journal of Applied Ecology 47, 309-317. doi:

10.1111/j.1365-2664.2010.01784.x.

Pemberton, R.W., 2000. Predictable risk to native plants in weed biological control.

Oecologia 125, 489-494.

Pemberton, R.W., Liu, H., 2007. Control and persistence of native Opuntia on Nevis and St.

Kitts 50 years after the introduction of Cactoblastis cactorum. Biological Control 41,

272-282. doi: 10.1016/j.biocontrol.2007.02.002.

Pimentel, D., McNair, S., Janecka, J., Wightman, J., Simmonds, C., O’Connell, C., Wong, E.,

Russel, L., Zern, J., Aquino, T., Tsomondo, T., 2001. Economic and environmental

16

threats of alien plant, animal, and microbe invasions. Agriculture Ecosystem and

Environment 84, 1-20.

Russell, J.C., Clout, M.N., McArdle, B.H., 2004. Island biogeography and the species

richness of introduced mammals on New Zealand offshore islands. Journal of

Biogeography 31, 653-664.

Sheppard, A.W., Hill, R., DeClerck-Floate, R.A., McClay, A., Olckers, T., Quimby Jr., P.C.,

Zimmermann, H.G., 2003. A global review of risk-benefit-cost analysis for the

introduction of classical biological control agents against weeds: a crisis in the

making? Biocontrol News and Information No. 24 91N -108N.

Simberloff, D., 2011a. How common are invasion-induced ecosystem impacts? Biological

Invasions 13, 1255-1268. doi: 10.1007/s10530-011-9956-3.

Simberloff, D., 2011b. Risks of biological control for conservation purposes. BioControl, 1-

14. doi: 10.1007/s10526-011-9392-4.

Simberloff, D., Stiling, P., 1996. How risky is biological control? Ecology 77, 1965-1974.

doi: 10.2307/2265693.

Swanton, C.J., Murphy, S.D., 1996. Weed science beyond the weeds. Weed Science 44, 437-

445.

Syrett, P., 1997. Biological control of St John's wort (Hypericum perforatum) in New

Zealand. Plant Protection Quarterly 12, 88-90.

Syrett, P., Scheele, S.M., Philips, B.A., 1984. Renewed activities in biological control of

ragwort. Proceedings of the NZ Weed and Pest Control Conference 37, 37-41.

Van Wilgen, B.W., de Wit, M.P., Anderson, H.J., Le Maitre, D.C., Kotze, I.M., Ndala, S.,

Brown, B., Rapholo, M.B., 2004. Costs and benefits of biological control of invasive

alien plants: Case studies from South Africa. South African Journal of Science 100,

113-122;

Van Wilgen, B.W., De Lange, W.J. 2011. The costs and benefits of biological control of

invasive alien plants in South Africa. African Entomology 19:2 (Special Issue), 504-

514.

Wilcove, D.S., Rothstein, D., Dubow, J., Phillips, A., Loso, E., 1998. Quantifying threats to

imperiled species in the United States. Bioscience 48, 607-615.

Williams, P., Timmins, S., 2002. Economic impacts of weeds in New Zealand. In: Pimentel,

D. (Ed.) Biological Invasions, Economic and Environmental Costs of Alien Plant,

Animal, and Microbe Species. CRC Press, London, pp. 175-184.

Withers, T.M., Hill, R.L., Paynter, Q., Fowler, S.V., Gourlay, A., 2008. Post-release

investigations into the field host range of the gorse pod moth Cydia succedana Denis

& Schiffermüller (Lepidoptera: Tortricidae) in New Zealand. New Zealand

Entomologist 31, 67-76.

Figure legends

Fig. 1. Weed biological control introductions into quarantine or released in New Zealand by

decade, including fungi, mites and insects (data from Ferguson et al., 2007 and EPA 2012).

17

Fig. 2. Histogram of beneficial effects assigned to individual weed biological control

introductions into New Zealand from 1920 to 1990 (Table 1). Data from Ferguson et al.

(2007) were categorized according to Table 1 (see case studies).

Table 1. The New Zealand Environmental Protection Authority risk assessment framework

for beneficial effects that are potentially relevant to biological control of weeds (not including

health effects). Dollar values are nominal/y and indicative.

Descriptor Environment Economy Social

Minimal

Highly localised and

contained environmental

impact, affecting a few

individual members of

communities of flora or

fauna, no discernible

ecosystem impact

Local/regional short-term beneficial

economic effects on small

organisations (business/individuals),

temporary job creation

<$5000

No social effect

Minor

Localised and contained

reversible environmental

impact, some local plant or

animal communities

temporarily benefited, no

discernible ecosystem

impact

Regional beneficial economic effects

on small organisations

(business/individuals), temporary job

creation

$5,000-49,999

Minor localised community benefit

Moderate

Measurable benefit to local

plant and animal

communities, expected to

pertain to medium term

Medium term (1-5 y) regional

beneficial economic effects with

some national implications, medium-

term job creation

$50,000-499,999

Local community and some

individuals beyond immediate

community receive social benefit

Major Long-term benefit to

localised ecosystem

Measurable beneficial effect in GDP,

some long-term (>5 y) job creation

$500,000-4,999,999

Substantial social benefit to

surrounding community and

individuals in wider community

Massive

Long-term widespread

benefits to species and/or

ecosystems

Significant ongoing effect beneficial

on GDP, long-term job creation on a

national basis

$5,000,000-49,999,999

Major social benefit affecting wider

community

Table 2. Fate of biological control agents for weeds imported into New Zealand

quarantine, 1920-2012 (data from Ferguson et al. (2007) and EPA (2012)).

Fate Number

Released and evaluated 33

Too early to assess 16

Not released (some pending) 32

Total 82

1

Highlights 1

33 weed biological control agents were assessed for benefit on a five-step scale. 2

Suppression of ragwort and St John’s wort were considered to be massive in benefit. 3

Other agents offered major (n=1), moderate (n=5) or minor (n=2) benefits. 4

Failure to achieve benefits from biological control is a considerable risk. 5

Improved communication about risk and reward is needed. 6