Presentation on theme: "Japanese knotweed biocontrol"— Presentation transcript:
1Japanese knotweed biocontrol Progress to dateLindsey Norgrove, Dick Shaw, René Eschen, Ghislaine Cortat, Alex Brook
2CBC activity in Europe Insect BCA history Weed BCA history CountryRecipientSourceAustria48Finland5France111Germany46Greece29Italy71Portugal18Spain9Sweden3UK141Total381In Europe for Insects there have been more than 300 releases of more than 150 predators and parasitoids with very little regulationEuropean settlers not only have a long history of starting invasions around the world but also of doing something about them as shown by this very unbalanced table of weed biocontrol releases
3Weed CBC - Long and extensive history Over the past 100 years, more than 400 different biocontrol agents have been used against around 150 target plants, totalling over 1,300 introductions around the globe.Weed biocontrol is not a novel approach and over the past century there has been a lot of activity
4This is a typical shot of the unruly knotweed ignoring the rules of the road and as one of the worst weeds in the world it was an excellent candidate for the first target for this approach in the EU. It costs GB a little over £165 million (188 million euro) and is famed for damaging the built environment
5It costs GB a little over £165 million (188 million euro) and is famed for damaging the built environment
6Cost of Japanese knotweed to GB 92% of the £166 million annual costs are experienced by the construction and development industry.
7A consortium of Sponsors came together in 2003 to sponsor the programme After many years of project development activity a consortium of sponsors came together to fund the biocontrol programme and CABI won the tender. British Columbia are also sponsors but don’t have a logo
8As is always the case, It soon became clear that Japanese knotweed had many more natural enemies in its native range than here and our job was to find the safest and most effective member of that group
9Many insects feeding on most parts 186 species of phytophagous arthropod recorded from Japanese knotweed in Japan.Of the many species of arthopods which are found damaging the plant in Japan there were almost no root feeders surprisingly but most other parts of the plant were attacked to some degree.
11A process of elimination Biocontrol comes down to a process of elimination whereby most potential agents are dismissed in the early stages and don’t ever make it to the laboratory study stage. Because of time constraints this presentation will not cover the years of research that led to us dismissing potential agents even at the late stage.
12In the interests of safety, the testing work was carried out in our Defra-licensed level 3 quarantine facilities alongside field work carried out by our collaborators in Japan
13I won’t deal with fungi in the short time I have today but as you can see from our lead pathogist’s face there were some exciting species found and the inset mycosphaerella leafspot remains a very damaging and promising potential agent it was just a little harder to work on but it should be noted that fungi have an excellent track record in weed biocontrol
14Aphalara itadoriAll of that painstaking research narrowed our options down to the sap-sucking psyllid A. itadori which is named after its host – always a good sign
15The adult is really just an egg-laying machine and it is her offspring, the nymphs that do the damage as they pass through their 5 developmental stages, sucking the sap and generating the white wax you can see in this slide
16Test Plant List 90 species and varieties representatives from 19 families.37 plants natives including all native Polygonaceae23 species introduced to the UK,3 species native to Europe,13 ornamental10 economically important UK speciesNo means of agreeing the test plant list in advance!Drawing up the test plant list based on phylogeny resulted in 89 species being tested including some American test plant species. This is a very long test plant list and includes some species that have some cultural value despite being quite distantly related
18The 78 spp. that did not receive eggs are excluded This graph summarises 4 years of work following over 145,000 eggs. This shows the results of the multiple choice oviposition studies and only includes those species which ever received any eggs. Note the rapid drop off from the 430 or so eggs laid on the target plant. To the right of the red line no adults were produced and to the left only invasive knotweeds, varieties or hybrids could support complete development from egg to adult. Therefore we can conclude that the psylid is incapable of maintaining a population on anything other than the target weed(s)Bar chart showing mean egg count on those plants that did receive eggs in multiple choice oviposition tests. (+/- 1SE). Development only successful to the left of red line
23Licensing: The two processes (England) Licence to release into the wild under the Wildlife and Countryside Act 1981Licence to free it from a Plant Health Quarantine license as “an organism likely to be injurious to plants in the UK” - EU Standing Committee on Plant Health were kept informedNo for the complex but effective licensing process which became more straightforward once it was dettermined that Plant health Legislation was appropriate. These two pathways had to be cleared simultaneously baciase without one licence granted the other would not permit release into the wildSee also: Hunt et al (2010) An international comparison of invertebrate biological control agent regulation: what can Europe learn? REBECA.
24W&C Act application for release Based on Eppo template Pest Risk AnalysisW&C Act application for releaseBased on Eppo templateBrand new version for Wales & EnglandInternal Govt iterative reviewACRE Committee reviewExternal Peer reviewPublic consultation (3 months)Chief Scientist adviceMinisterial decision for Sec. of StateRelease from PH quarantine licenceW&C license to releaseThe EPPO – based PRA was the best tool for the job as it is widely recognised despite not being designed for a “beneficial pest”. As can be seen there was intense consultation including a full public consultation alongside peer reviews and it was useful that a peer-reviewed paper supported the application
252o & 3o and community level effects? Choice tests with commercially available generalists showed no feeding preferenceNative coccinellids fed exclusively on psyllids fared worse than when fed on aphids
26One requirement of the license was that a 5 year monitoring plan was agreed and fully funded. The plan agreed represents one of the most thorough post- release monitoring plan in weed biocontrol history and is designed to detect any negative impact as well as record the efficacy of the agent over time. There was also a requirement for a contingency plan which requires the treatment of any affected sites with insecticide and a follow-up herbicide treatment should any significant adverse effects be observed (this is the first time such a contingency has been used to our knowledge). A combination of fixed quadrats and repeat sampling with vortice and reverse leaf-blowers is being used to record all arthropods whilst vegetation surveys cover the flora. In Phase 1 Fallopia dumetorum (red data book sp.) and F. convolvulus plants + and – insecticide were planted out at release and any impact recorded.
27Caged no-choice & Choice experiment Oviposition and development of A. itadori and non-target impact on F. dumetorumAs a result of the late frost in may which removed the knotweed leaves at release sites and therefore most of the eggs, we decided to carry out caged studies alongside applying for additional release at one site in higher numbers. This would ensure useful data was generated outside the comfort of the lab. This involved the use of the most important native Fallopia sp. And requent measuring, production of cages and sleeves very labour-intensive. Ca. three-four people for three days every fortnight for three months.
28Caged no-choice experiment Many eggs, some nymphs, limited developmentF. dumetorumIn the extreme example of no-choice tests plenty of eggs were laid on the non-target in the absence of its normal host but as expected very few nymphs developed and none made it to adult.Grey bars eggs, black bars nymphs
29Comparison with pre-release quarantine multiple-choice tests Patterns in oviposition similarNo complete development on any non-target speciesVery similar to published studies1.9%0.2%1.6%0.4%0 (2)%0 (1)%QuarantineCagedOpen fieldRedrawn from Shaw (2009)Eggs two (four) weeks after startIn the more realsitic multiple choice study with both non-target Fallopia spp. we found almost identical results as we generated in quarantine. This is one of the first times such a comparison has been made that was not retrospective.
30Host-specificity testing QuarantineCagedOpen fieldNo-choiceMultiple choiceFundamental host rangeRealised host rangeOur experiments span the whole range of specificity tests done in other weed biological control programmes and all show that the risk to native ecosystems is very lowArtificialNaturalHost-range tests reliably predict non-target attack(Pemberton 2000; Fowler et al. 2000; Barton 2004; Briese 2005)Non-target attack either “predicted” or ephemeral
31SummaryCaged and open-field studies confirm the host-specificity of Aphalara itadoriNo impact of the psyllid on non-target plant speciesNo impact of A. itadori on native vegetation or invertebrate communityRisk of non-target impact on native vegetation and invertebrates very lowIn summary
32No A. itadori found in winter sampling Typical sample from evergreens contained 100s native psyllids, but no A. itadoriSpecies sampled included: yew, Pinus spp., Leyland cypress, etc.Disappointing but it is like finding a needle in a haystack at this stageLow abundance of A. itadori at release sites larger releases required for establishment
33Successful overwintering!! The latest news is that in early April during our first visit to the field we saw adult Aphalara at the site which had received the most adult psyllids (6,000) which means that they had successfully overwintered
34Mass-producing psyllids Two production cycles:First in cages inside Controlled environmentWe are currently reaching the end of the mass production facility and have begun our Phase 2 widespread releases thanks to the mass-rearing that has been going on over the past 3 months.
35Mass-producing psyllids Data loggers allow checks on developmentInsects to be ready for releases in last two weeks in May
36What can we expect? If successful: Establishment of the agent Spread to JKReduced plant vigourReduced control costsRecovery of native speciesControl not eradication!Biocontrol is a long term solution to very large scale problems and as such is not a quick fix. But if released we would hope to see damaged and less invasive knotweed within a few years. Even if its rate of spread is reduced then it would have been worthwhile. A 1% reduction in costs in GB will have paid for the project in less than a year.
37EU opportunities Sheppard, Shaw & Sforza - Weed Research 2006 SpeciesFormOriginEU distributionGenus native?ConflictBC historyBuddleja davidiiPhChinaTemperateNobOYesFallopia japonicaGeJapanNoAcacia dealbataAustraliaMediterraneanYesdAzolla filiculoidesHyN AmericaTemp/MedAilanthus altissimaImpatiens glanduliferaHeIndiaRhododendron ponticumS EuropeRobinia pseudoacaciaFSenecio inaequidensS AfricaAmbrosia artemisiifoliaThC AmericaCarpobrotus edulisChHeracleum mantegazzianumW AsiaSolanum elaeagnifoliumS AmericaTem/MedBaccharis halimifoliaHydrocotyle ranunculoidesLudwigia peploidesCrassula helmsiiAustralasiaElodea canadensisMyriophyllum aquaticumSolidago canadensisThere are many other targets for this approach in Europe and we can provide pdf copies of this paper if requested.
38Hydrocotyle ranunculoides In GB we are actively researching other targets such as Floating PennywortPhoto – T. Renals
39For which a South American weevil looks to be very promising
40Impatiens glandulifera Himalayan balsam is another excellent target and extensive research in the Himalayas has revealed a rust fungus
41Puccinia rustWhich is currently undergoing late stage testing in our quarantine facility
42Thank you to all involved Dr Harry Evans (CABI),Dr Marion Seier & Dr Rob ReederRob Tanner (CABI)Djamila Djeddour (CABI)Dr Carol EllisonDrs Murphy, Cock and Holderness (CABI)Ghislaine Cortat (CABI)Dr Rene EschenAnna HarrisSonal VariaCorin PrattAlex BrookDr Esther GerberValérie Coudrain & Sarah Bryner (CABI tudents)Sasha WhiteDr Paul Cannon and Dr Alan Buddie (CABI)Linda Birken (Imperial College student)Gareth Martin (Imperial College student)James Broom (Imperial College student)Dr John Bailey and Kat Pashley (Leicester University)Dr Lois Child (Loughborough University)Dr Andy Polaszek & others (NHM)Professor Masami Takagi (Kyushu University)Dr Daisuke Kurose (Kyushu University)Dr Narutu Furuya (Kyushu University)Dr Naoki Takahashi (Kyushu University)Yuko Inoue (Kyushu University)Dr Fritzi Grevstad (University of Washington)Dr Bernd Blossey (Cornell University)Dr Rob Bouchier (AAFC Canada)Dr Brian Van Hezelwink (AAFC Canada)Victoria Nuzzo (Independent Consultant)Mic Julien (CSIRO)Dr Andy Sheppard (CSIRO)Dr Simon Fowler (Landcare Research NZ)Drs Ted Centre & Gary Buckingham (Florida Uni)Profs Mick Crawley &Tim Coulson (Imperial College)Dr Willie Cabrera Walsh (SABCL)Dr Jonathan Newman (CEH)Dr Usha Dev (NBPGR)Dr Ravi Kheterpal (NBPGR)Dr Robin Adair (DPI Queensland)Drs John Ireson & Richard Holloway (Utas)Lindsay Smith (Landcare Research)ACREFERA – many especially Dr Claire SansfordPesticide Safety DirectorateThe Non Native Species SecretariatThe Project Board and sponsors for funding and guidanceAs with any long term international research there are a lot of people to thanks.
43Thank You Any Questions? A multilingual Thank You slide is available in the CABI Resources site (www.cabi.org/commercial)