Are Japanese Knotweed (Fallopia japonica) Control and Eradication Interventions Effective? (systematic review)
Japanese knotweed (Fallopia japonica) is an invasive plant listed as one of the IUCN’s top 100 invasive species of global concern. It is a vigorously competitive plant that regenerates readily, and is difficult to control. Japanese knotweed control and eradication is undertaken using a wide suite of mechanical and chemical techniques. Many statutory and non-statutory agencies publish guidelines detailing the effectiveness of various methods, but a critical appraisal of empirical evidence regarding the effectiveness of different control and eradication methods has not previously been undertaken. The need to evaluate control and eradication methods under a variety of circumstances and time periods has been identified by UK stakeholders, but has global relevance.
To systematically collate and synthesise published and unpublished evidence in order to address the question: “Are Japanese knotweed (Fallopia japonica) control and eradication interventions effective?”
The secondary objective was: “To investigate whether effectiveness of control and eradication treatments for Japanese knotweed is influenced by the following factors:
- Environmental and geographical factors;
- Operational level variables; and
- Hybridisation and species variety.”
Electronic searching was completed using the following databases, catalogues and web-engines: Agricola, CAB Abstracts, Digital Dissertations Online, Index to Theses Online, ISI Web of Knowledge (including ISI Web of Science and ISI Proceedings searches), JSTOR, Science Direct, Scirus (all journal sources), Scopus, AllTheWeb, Dogpile, Google Scholar, Scirus (all web sources), Blackwell Synergy, ConservationEvidence.com, Copac, Directory of Open Access Journals, English Nature’s “Wildlink” library catalogue, Elsevier, European Nature Information System database V2 (EUNIS), iSpecies, and SpringerLink. Publication searches on 49 statutory and non-statutory organisation websites were conducted. Specialist searching was completed on 14 invasive species websites. Bibliographies of articles accepted into the review, traditional literature reviews, and relevant literature lists were searched for additional articles. Personal contact with researchers was used to retrieve further data.
Any studies in any habitat that examined the impact on abundance of any management intervention used to control or eradicate any subspecies, variety or hybrid of Japanese knotweed were included. Appropriate spatial or temporal controls were a prerequisite for studies to be included in quantitative analysis. Studies of biological control were not included as no data are yet available on effectiveness.
Data collection and analysis
The inclusion criteria were met by 74 articles, and these included information for Japanese knotweed and the hybrids Bohemian knotweed (F. x bohemica) and back- crossed F. japonica var. ‘Crimson Beauty’. Multivariate synthesis was used to identify broad patterns in the effectiveness of all management interventions, using data extracted from 64 of the included articles only. Data suitable for meta-analyses were extractable from only 11 articles. Meta-analyses were used to examine the following six management techniques only (due to lack of suitable data on any other techniques): the herbicides glyphosate and imazapyr used alone and in combination, cutting applied alone, cutting followed by filling the stem with glyphosate herbicide, and cutting followed by spraying regrowth with glyphosate.
All six interventions investigated by meta-analysis produced statistically significant decreases in knotweed abundance in the short-term, except for cutting used alone. However, the ecological significance of the impacts of these treatments is uncertain, and there is no robust evidence available regarding their long-term effectiveness. Uncertainty is exacerbated by the small number of individual effect sizes, the limitations of the pooled studies (particularly confounded baselines and short timescales), and the high heterogeneity among included studies. The meta-analyses demonstrate that existing available evidence is insufficient to derive generic evidence- based management guidance for these particular techniques. These conclusions are supported by multivariate analysis of lower quality data from a wider range of sources. Variation in effectiveness was evident both within and between treatments, but this variation could not be linked to any ecological or intervention-related variables.
Timing of control influences the effectiveness of glyphosate application, with application later in the year appearing to have a more significant effect on knotweed abundance. However, the effect is no longer significant when considered alongside the duration of control. This relationship should be treated cautiously, as it could be confounded by one of the many variables that differed between the included studies.
No conclusive evidence was found for differences in effectiveness of management techniques due to taxonomic variation.
Available evidence suggests that applications of the following six control methodologies will not eradicate Japanese or hybrid knotweed in the short-term: the herbicides glyphosate and imazapyr used alone and in combination, cutting applied alone, cutting followed by filling the stem with glyphosate herbicide, and cutting followed by spraying regrowth with glyphosate. The review highlights a lack of readily-available, long-term, robust, controlled experiments assessing the effects of the full range of management techniques used against Japanese and hybrid knotweed. As such, it emphasises important deficiencies within the current body of evidence. The authors of this review are aware of control methods in use other than the six methods analysed in this review. However, as some monitoring results are not made readily available, the effectiveness of the full range of control and eradication methods currently implemented cannot be tested. Readers must therefore put the evidence presented here into a broader context of poor data accessibility.
This review recommends further research into methods used to control and eradicate Japanese knotweed. This research should focus on long-term collection of monitoring data, adequate replication, and investigation of the impacts of treatments on rhizomes. A large, well-replicated experiment or monitoring programme examined over a long time period could be used to test a range of different factors that may influence Japanese knotweed control and eradication. It is recommended that collaboration between stakeholder groups across and between countries be used to achieve a multi- site aspect to this research. Considering the substantial amount of money that is already invested in knotweed control, it would be worthwhile to provide funding for developing more effective ways of managing the problem under different circumstances.
Japanese knotweed (Fallopia japonica*) was first introduced into Britain from Japan in 1825 (Beerling et al., 1994). Originally used as an ornamental or as cattle fodder, the first escape was reported in the late 19th century, with naturalisation occurring in the early 20th century (Bond and Turner, 2005). Due to the subsequent spread of this invasive weed, F. japonica is now established alongside railways, canals, rivers and streams, roadsides, and areas of human disturbance such as waste tips. It is widespread in the British Isles, occurring in the south and as far north as the Outer Hebrides and Orkney. The plant has also become a problem in the USA, Canada, many European countries, Australia and New Zealand (Beerling et al., 1994; Australian Weeds CRC, 2005). The species has been described as one of IUCN’s top 100 invaders (Lowe et al., 2000).
Spread in Great Britain is by vegetative means via rhizomes, as most individuals in Britain are female, and therefore the setting of fertilised F. japonica seed is not possible (Seiger, 1992). Within its native range, reproduction is assisted by wind- and water-dispersed seeds. Successful invasion has been facilitated mainly by the deposition of plant material or cartage of soil containing plant fragments; therefore, it can be transported great distances and released accidentally in a relatively easy manner (Child et al., 2001). However, F. japonica hybridises with F. sachalinensis, the hybrid F. xbohemica (also known as Reynoutria xvivax) producing viable seed and being highly invasive (Bimova et al., 2001). There is some indication that this hybrid is more invasive than its parents, as it is more difficult to control and has a higher regeneration rate (Bimova et al., 2001, 2003). Hybridisation with F. baldschuanica to produce fertile F. xconollyana has also occurred (Bailey, 2001). Both species are widespread in Europe, and F. xbohemica is also widespread in the USA (known as Polygonum xbohemicum) (Zika and Jacobson, 2003).
Japanese knotweed can grow up to 3m high (Beerling et al., 1994). This perennial plant forms tall thickets that exclude all other vegetation (Bond and Turner, 2005). A dense leaf canopy is created, and when the leaves are shed in autumn they decompose, forming a thick mulch that prevents germination of other plants (Kidd, 2000). Native plants therefore cannot compete with this invasive, and local plant biodiversity is reduced (Seiger, 1992).
Various methods for control and eradication of Japanese knotweed have been used. Mechanical methods such as grazing, cutting, mowing, pulling and digging have been used to control outbreaks, but without intensive applications over long periods of time, these will not eradicate the weed (Seiger, 1992; Soll, 2004; Bond and Turner, 2005). Herbicides are frequently used to control or eradicate the weed over several years, and several herbicides (selective and non-selective) and application methods (e.g. stem injection) are in use (Seiger, 1992; Soll, 2004). Large-scale excavation is usually the only method of rapid eradication (Japanese Knotweed Alliance website; Brian Tollitt, United Utilities, pers. comm. 2006). Geosynthetic textiles and thick polythene sheeting have been utilised to reduce the spread of Japanese knotweed. These can be used to encase excavated material or cover infestations and then buried, or they can be placed into the ground beside infestations to act as a vertical barrier (REC Ltd. website; Renals and Rene, 2001; Wreford Ltd. website). Combinations of the various control methods are also used. Investigations are being conducted into possible biological controls for Japanese knotweed (Japanese Knotweed Alliance website).
Defra (2003) estimated it would cost £1.56 billion to eradicate Japanese knotweed across all infestation in Britain, with £52 million of that estimate related to removal of the weed from riparian areas. Fasham and Trumper (2001) stated that £160,000 had been spent on controlling Japanese knotweed in the Swansea area since 1992, with estimated costs of full control in the area at nearly £8,000,000. The threat of this plant to native flora has been recognised and there is now legislation surrounding its handling and planting (Fasham and Trumper, 2001). It is therefore necessary to be aware of the success of the measures taken to control or eradicate Japanese knotweed.
Using systematic review methodology, the interventions used to control or eradicate Japanese knotweed will be critically appraised. The review will consider the best available evidence of the effectiveness of different control and eradication methods in different situations (Table 1). The review will limit bias through the use of comprehensive literature searching (both published and unpublished), specific inclusion criteria, and formal assessment of the quality and reliability of the studies retrieved. Subsequent data synthesis (qualitative and/or quantitative) will summarise evidence, guiding the formulation of appropriate evidence-based management guidelines and highlighting gaps in research evidence.
The review should be of use to staff carrying out, or advising on, Japanese knotweed control and eradication for environmental organisations (statutory and non-statutory) and local authorities, as well as landholder stakeholder groups. It will have value at regional and national scales, informing local management or policy decisions, and has the potential to influence national policy and guidelines on Japanese knotweed management.