Thirty-seven alien plant species, pre-identified by horizon scanning exercises were prioritised for pest risk analysis (PRA) using a modified version of the EPPO Prioritisation Process designed to be compliant with the EU Regulation 1143/2014. In Stage 1, species were categorised into one of four lists – a Residual List, EU List of Minor Concern, EU Observation List and the EU List of Invasive Alien Plants. Only those species included in the latter proceeded to the risk management stage where their priority for PRA was assessed. Due to medium or high spread potential coupled with high impacts twenty-two species were included in the EU List of Invasive Alien Plants and proceeded to Stage 2. Four species (Ambrosia trifida, Egeria densa, Fallopia baldschuanica and Oxalis pes-caprae) were assigned to the EU Observation List due to moderate or low impacts. Albizia lebbeck, Clematis terniflora, Euonymus japonicus, Lonicera morrowii, Prunus campanulata and Rubus rosifolius were assigned to the residual list due to a current lack of information on impacts. Similarly, Cornus sericea and Hydrilla verticillata were assigned to the Residual List due to unclear taxonomy and uncertainty in native status, respectively. Chromolaena odorata, Cryptostegia grandiflora and Sphagneticola trilobata were assigned to the Residual List as it is unlikely they will establish in the Union under current climatic conditions. In the risk management stage, Euonymus fortunei, Ligustrum sinense and Lonicera maackii were considered a low priority for PRA as they do not exhibit invasive tendencies despite being widely cultivated in the EU over several decades. Nineteen species were identified as having a high priority for a PRA (Acacia dealbata, Ambrosia confertiflora, Andropogon virginicus, Cardiospermum grandiflorum, Celastrus orbiculatus, Cinnamomum camphora, Cortaderia jubata, Ehrharta calycina, Gymnocoronis spilanthoides, Hakea sericea, Humulus scandens, Hygrophila polysperma, Lespedeza cuneata, Lygodium japonicum, Pennisetum setaceum, Prosopis juliflora, Sapium sebiferum, Pistia stratiotes and Salvinia molesta).

Biodiversity, ecosystem services, Europe, impact, non-native, risk management

Trade liberalisation and rapid globalisation have led to the increased spread of invasive alien species (IAS) around the world (van Kleunnen et al. 2015). IAS (plants, animals, fungi or micro-organisms) are recognised as one of the greatest threats to biological diversity by inflicting irreversible damage to the ecosystems they invade (Wilcove et al. 1998). In Europe, there are an estimated 12,000 alien species with 10-15 % considered invasive and it is these species that cost the EU around €12-billion per year (European Commission 2014, Kettunen et al. 2008).

Established invasive alien plant species are one of the largest groups of IAS both in terms of species numbers and the area they occupy (Sheppard et al. 2006). There are an estimated 3,749 naturalised alien plant species in Europe of which 1,780 are alien to Europe, with the remaining being native to parts of Europe (Pyšek et al. 2009). When alien plants invade regions, they can outcompete native plant species through direct (Daehler 2003) or indirect competition (Murrell et al. 2011). Impacts, because of habitat modification and displacement of native plant species can cascade to higher trophic levels impacting at an ecosystem scale (Tanner et al. 2013, Daniel et al. 2003, Levine et al. 2003). Although impacts on ecosystem services are less studied, examples show negative effects on provisioning (Kasulo 2000, Eagle et al. 2007), regulating (Chittka and Schürkens 2001, Prater et al. 2006) and cultural services (Chilton et al. 2002, McFarland et al. 2004).

To mitigate the threat of IAS to the European Union (EU), the European Commission adopted the EU Regulation (No. 1143/2014) ‘on the prevention and management of the introduction and spread of invasive alien species’ which came into force on the 1^st January 2015 (EU 2014, Genovesi et al. 2015, Tollington et al. 2015). The EU Regulation, hereafter referred to as the IAS Regulation, aims to primarily address the negative impact of IAS on biodiversity and ecosystem services, while impacts on human health and the economy are considered as aggravating factors. The IAS Regulation is centred around three main themes (1) prevention, (2) early warning and rapid response, and (3) management. The IAS Regulation will restrict the use, trade and transport of certain IAS and will be underpinned by a list of IAS of Union concern. At present the Union List contains 37 IAS, of which 14 species are invasive alien plants (European Commission 2016).

The IAS of Union concern will be subject to stringent enforcements including a ban on sale and preventative actions such as a ban on import (see Genovesi et al. 2015). Member States will be obliged to prevent the spread and conduct eradication and management measures for species on the list and already present in Member States (EU 2014). In theory, such measures would go a long way to mitigating entry and impacts of invasive, or potentially invasive alien plants in the EU, especially when considering two thirds of established alien plant species have been introduced intentionally for horticulture or agricultural purposes (Keller et al. 2011).

The IAS Regulation places an emphasis on prevention as opposed to cure, and as such the focus should be on species with a limited regional distribution within the Union, and species that are currently absent but pose a potential threat in the future. Many European countries and regional organisations have produced species lists and conducted horizon scanning studies which have identified priority species (Gallardo et al. 2015, Roy et al. 2015). However, for a species to be included in the list of Union concern a risk assessment is required to technically and objectively evaluate scientific and economic evidence to determine the level of risk associated with a species. It should be noted that the European and Mediterranean Plant Protection Organization (EPPO) always combines risk assessment with risk management, resulting in a risk analysis and hereafter referred to as a pest risk analysis (PRA).

A PRA can be a time-consuming process requiring significant finances and high levels of species specific expertise. When presented with a large pool of invasive, or potentially invasive alien plants, prioritizing species for PRA is an essential prerequisite to focus limited resources. High priority species would be those that have the highest negative impact and can be prevented from entering, or cost effectively managed in the European Union (Kumschick et al. 2012, Branquart et al. 2016).

Several schemes have been developed for different countries or regions to prioritise alien plants (Austria-Germany: Essl et al. 2011, Belgium: D’hondt et al. 2014, central Europe: Weber and Gut 2004). The scheme by Brunel et al. (2010) was designed to assess alien plants under the Plant Health Regulation. However, in the context of the IAS Regulation, more emphasis is required on impacts on biodiversity and ecosystem services. Due to this shift in the regulatory process of invasive alien plants, a new prioritisation scheme was designed to ensure that species prioritisation was compliant with the IAS Regulation (Branquart et al. 2016). What the new prioritisation process allows is to (1) prioritise species based on their impacts and spread, (2) to exclude species unsuited for PRA due to a lack of scientific information and (3), include the effectiveness of potential risk management measures for a given species in the prioritisation process. Thus, the prioritisation process deals with both risk assessment and risk management (i.e. risk analysis).

The objective of this study was to produce a list of alien plant species that comply with the definitions and criteria of Article 4 of the IAS Regulation, i.e. alien species that would be capable of causing major detrimental impacts on biodiversity and associated ecosystem services after establishment and spread within the EU territory, and to determine which of these have the highest priority for PRA at the European level.

In March 2016, a three-day workshop was held at EPPO in Paris (FR), with the purpose of prioritising a list of invasive alien plants for PRA as part of a LIFE funded project ‘Mitigating the threat of invasive alien plants in the EU through pest risk analysis to support the EU Regulation 1143/2014’ (LIFE15 PRE FR 001) (see, www.IAP-risk.eu). Eight experts from the EPPO Panel on Invasive Alien Plants, the NERC Centre for Ecology and Hydrology and the EPPO Secretariat attended the workshop.

EPPO prioritisation process compliant with the IAS Regulation

The prioritisation scheme used for this study was an amended version of the EPPO prioritisation process for Invasive Alien Plants (Brunel et al. 2010, EPPO standard PM5/6), specifically adapted within the remit of the LIFE project to be fully compliant with the IAS Regulation. A full description of the process is given in Branquart et al. (2016) and depicted in figure (1). The prioritisation process was designed to meet the requirements of Article 4 (IAS Regulation) where the highest priority for performing a PRA at the European level is given to species that satisfy the following criteria: (i) they are alien to the territory of the EU excluding the outermost regions, (ii) they are capable of establishing a viable population and spreading rapidly in the environment in the EU (excluding the outermost regions), (iii) they are capable of causing major detrimental impacts to biodiversity and the associated ecosystem services, (iv) actions can be taken to effectively prevent, minimise or mitigate their adverse impact, which involves that they are moved from country to country primarily by human activities and they still have a significant area suitable for further spread within the EU (EU 2014).

The first stage of the process, the preliminary risk assessment stage, categorises each species into one of four lists (Residual List, EU List of Minor Concern, EU Observation List and the EU List of Invasive Alien Plants) by addressing pre-determined criteria (questions). To proceed to any of the three EU lists, each species needs to meet the requirements of questions A1, A2, A3, A5 and A6, i.e. a positive (yes) answer is required. If a negative (no) answer is recorded, the species is included in the Residual List of species that do not qualify. Reasons a species (including subspecies, varieties, hybrids and cultigens, hereafter collectively called species) may be included in the Residual list include uncertainty in taxonomy and nomenclatural (question A1. Fig. 1), or a lack of current scientific information (question A3. Fig. 1). Only those species included in the EU List of Invasive Alien Plants (those species which have the highest potential spread capacity and high negative impacts on biodiversity or ecosystem services) proceed to the risk management stage.

Figure 1.

Decision scheme for the EU prioritisation process for alien plants (Taken from Branquart et al. 2016).

Within the second stage, the preliminary risk management stage, priority for a PRA at the EU level is evaluated based on the feasibility and cost-effectiveness of mitigating impacts with management measures and/or preventative actions. The output of stage two is to define the species into one of two categories:

(1) the plant species is included in a List of Priority Invasive Alien Plants for performing an EU level PRA,

(2) the plant species is included in a List of Invasive Alien Plants that are not considered as a priority to conduct a EU level PRA.

Modelling the potential occurrence of species

To support question A6, ‘based on ecoclimatic conditions, could the species establish in at least 3 EU Member States (excluding the outermost regions)’, maps of potential species occurrence were developed. A key criterion in evaluating the risk of a species to the EU is to assess if the species can establish under current climatic conditions. This is especially important for species which are currently absent from the region but have been highlighted as a risk through horizon scanning exercises.

However, modelling the potential distributions of alien species presents challenges, including the non-equilibrium nature of the distribution, presence of casual records representing failed introductions and spatial biases in recording effort (Václavik and Meentemeyer 2012). Substantial effort is usually required to develop accurate models that account for these effects, prohibiting the use of such models for rapid multi-species PRA prioritisation exercises. Therefore, we adopted a simple but precautionary approach based on delimiting a ‘climate envelope’ of each species that can be projected onto a map of Europe.

To delimit climate envelopes, we used the 19 standard bioclimatic variables gridded at 10 arcminute resolution (0.167 × 0.167 decimal degrees) from WorldClim (Hijmans et al. 2005). ‘Climate space’ was summarised by taking the first two axes of a principal components analysis (PCA) on centred and scaled bioclimatic variables, with log-transformed precipitation variables. These axes captured 77.5 % of the variation in climate. For each species, georeferenced occurrence data was obtained from the Global Biodiversity Information facility (GBIF: www.gbif.org). Data points were filtered according to expert opinion (Figure 2A). The species occurrences were then plotted in climate space, by extracting the PCA axis scores for occurrence locations (Figure 2B). To delimit a climate envelope for each species, bivariate density kernels were fitted to the occurrences in climate space using the kernelUD function of R package adehabitat with automatic selection of the smoothing parameter (Calenge 2006). From these models, 95 % kernel density polygons were extracted for each species. These bound the region of climate space containing 95 % of the smoothed occurrence density of each species. Finally, the climate envelopes were projected onto the EU by identifying the grid cells whose PCA axis scores fell inside the species’ climate envelope (Figure 2C). The resulting maps were critically appraised by the working group panel, using their expert knowledge to consider the accuracy of the estimates and the potential for non-climatic factors such as habitat availability to limit establishment.

Figure 2.

An example of the distribution maps and potential occurrence in Europe – Ambrosia confertiflora. A Global occurrence locations were obtained from the Global Biodiversity Information Facility B The global climate was summarised as two principal components analysis (PCA) axes on the 19 WorldClim layers (Hijmans et al. 2005). Species occurrences were plotted in this climate space and a bivariate normal kernel density model (Calenge 2006) was used to estimate ‘climate envelopes’ at different percentiles C These envelopes were then projected onto geographic space in the EU. Shading indicates these percentiles, with smaller numbers indicating higher density of occurrences.

We emphasise that this method does not provide a definitive estimate of the potential for further species establishment, but rather a way of rapidly assessing if a species is worthy of further consideration in full PRA. We also note that the 95 % density kernels may be overly generous and exceed the climatic tolerances of the species. However, while a lower percentage threshold could have been used to constrict the envelopes, a precautionary approach is desirable for our purpose, given that invasive species may not have fully filled their climate niche space and because many species can invade outside of their native climatic niche (Bocsi et al. 2016).

The 37 alien plant species prioritised in this study include representatives from 23 families where Asteraceae (5 species) and Poaceae (4 species) are most represented (Table 1). In total, the list contained 6 aquatics and 31 terrestrial species. Terrestrial species included 4 perennial grasses, 10 vines, 6 tree species, 7 woody shrubs and 4 perennial herbs. Almost half of the species (43 %) were native to Asia, followed by South America (18 %), North America (13 %), Africa (8 %), Australia (5 %) and pan-global species (8 %).

Globally, numerous prioritisation schemes have been specifically designed to address specific taxonomic groups (Brunel et al. 2010, Worner et al. 2013), regions or habitats (Dawson et al. 2015), pathways (NOBANIS 2015) or requirements of specific regulations (see McGeoch et al. 2016). With the implementation of the IAS Regulation, the European Commission has placed a clear focus on mitigating the negative impacts of IAS on biological diversity and ecosystem services, coupled with an underlying requirement to focus efforts on prevention rather than cure. Often, risk management components are lacking in prioritisation schemes (Heikkilä 2011), even though there is a clear advantage of incorporating such aspects to prioritise species that can be effectively controlled over other more difficult species (Hulme 2009). The current EU prioritisation scheme has been specifically designed to incorporate the requirements of the IAS Regulation and to the best of our knowledge, this is the first-time invasive alien plant species have been prioritised using a scheme compliant with the IAS Regulation.

This study identified 19 globally invasive alien plant species with a high priority for a PRA at the EU level. As shown in our results, all 19 species comply with the IAS definition and criteria of art. 4 of the IAS Regulation, i.e. alien species being capable of causing major detrimental impacts to biodiversity and the associated ecosystem services after establishment and spread within the territory of the EU. Within the first stage of the prioritisation scheme, four species (A. trifida, E. densa, F. baldschuanica and O. pes-caprae) were assigned to the EU Observation List highlighting that at the current time the species are likely to cause only a moderate detrimental impact to biodiversity and associated ecosystem services. A. trifida has become a major weed of annual crops in the US (Weaver 2003) and is a threat for the economy where it has established in Europe, especially in SW France (Chauvel et al. 2015), while O. pes-caprae has impacts on livestock. It should be noted that the placement of species in the three lists is not a definitive placement and each species should be reviewed as and when new information comes to light. This is particularly important for any species included in the EU list of Minor Concern and EU Observational List.

In the first stage of the prioritisation scheme, eleven species were assigned to the Residual List and thus did not qualify for further assessment. Having a clear understanding of the taxonomic identity of a species is an essential component in any prioritisation, and subsequent PRA. This is important to ensure that the assessment is performed on a distinct organism (IPPC 2016) but also to ensure that information used in the assessment is relevant to the organism under consideration (Elith et al. 2012). In our initial list of 37 species, the taxonomy of one species, Cornus sericea, was identified as being uncertain as in Europe naturalised plants belong to a complex of hybrids of C. sericea and C. alba (Q-Bank 2016). It has also been suggested that C. alba is conspecific with C. sericea (National Botanic Garden of Belgium 2016). We suggest that further research is carried out on the exact identity of the species within Europe before any PRA is conducted to reduce uncertainty.

If an invasive plant is native to part of the European Union, this would preclude its inclusion on the list of species of Union concern. Although Hydrilla verticillata is often considered non-native to Europe, there is some uncertainty to the status of the species and Lansdown (2013) details the species as native to Belarus, Ireland, the United Kingdom (southern Scotland) and the Russian Federation. There is additional uncertainty of its native status in Latvia and Poland (Cook and Lüönd 1982). In the absence of a pan-global biogeographical molecular study the uncertainty of native populations within Europe will remain (Zhu et al. 2015). It should be noted that provisions are made within the IAS Regulation (Article 11) for species native to the Union, where their inclusion on national lists can be used to enhance regional cooperation.

Most species included in the Residual List (75 %) warrant their place due to the lack of current information on the species. A PRA is only as robust as the scientific information which is used to compile the assessment and even though uncertainty ratings can go some way to capturing data gaps, or conflicting information, without some baseline data consideration is needed to whether a PRA is warranted. Based on the lack of quantitative impact studies, and to some extent information on the biology and ecology of the species (at a global scale), we considered Albizia lebbeck, Clematis terniflora, Euonymus japonicus, Lonicera morrowii, Prunus campanulata and Rubus rosifolius are not suited for a PRA at this time. We do suggest that a comprehensive literature review is conducted periodically for each species in the Residual List, including those species where there is uncertainty in potential for establishment (C. odorata, C. grandiflora and S. trilobata). If new scientific information comes to light that may change the outcome of the prioritisation, the species should be re-evaluated.

Impact studies can be biased to species which are widespread and/or high-profile species to particular sectors of society (Hulme et al. 2013). When considering species which are absent from the EU, there is a clear need to use the invasion history from another region as a proxy (Gallardo et al. 2015). As already mentioned, most species assessed under question A9, impacts on ecosystem functions and services, received a medium score (69 %) with a high level of uncertainty (55 %). This is in contrast with the previous question on impacts on native species where 84 % of the species received a high score with medium uncertainty (68 %). This is not a surprise as impacts can be ambiguous to define in relation to ecosystem services and impacts can be inconspicuous in many studies conducted over a short timeframe (Eviner et al. 2012). It is however fair to note that our understanding of the effects of invasive plants on ecosystem services is growing (Vilà et al. 2010), and with the prominence of ecosystem services in the IAS Regulation further studies will undoubtedly follow.

All 22 species evaluated under stage 2 have potential for further spread, though three species, namely Euonymus fortunei, Ligustrum sinense and Lonicera maackii were not considered a priority for an European level PRA due to being widely cultivated within the region without showing any signs of invasive behaviour. E. fortunei has been cultivated within the region since the late 1800s where it is grown in parks and gardens (Personal Communication, John David, Royal Horticultural Society, UK, 2016). It has however, been identified in the eastern USA as a species spreading into native plant communities (Missouri Botanical Garden 2016), and research has shown it causes native species decline (Bauer and Reynolds 2016, Mattingly et al. 2016). We recommend that Member States monitor these species, e.g. considering the possibility to join a network of sentinel gardens (to detect as soon as possible any sign of potential invasiveness) (Visser et al. 2014).

In the prioritisation scheme, questions B3-B5 focus on the cost effectiveness of prevention and management measures and assess if the introduction and spread of the species can be reduced by trade restrictions, other preventative actions (pathway management) or cost-effective management in the field (Branquart et al. 2016). A positive answer to any of the three questions indicates that a full PRA may identify actions to mitigate entry or spread. The risk from the majority of species (84 %) could be mitigated with trade restrictions as most are either traded or absent from the EU. However, for A. confertiflora, A. virginicus, and C. jubata it was considered that pathway management would be ineffective in detecting and preventing the incurrence of small plant propagules. As all species are relatively large in form, management in situ should be a feasible cost-effective option for isolated incurrences, particularly as management options exist for each species (see Panetta 2015).

It should be noted for Andropogon virginicus, Prosopis juliflora and Cortaderia jubata, trade in these species is predominately via internet sites such as eBay and Amazon and as such any trade restrictions may be ineffective in the absence of greater enforcement of existing regulations (Lenda et al. 2014). The volume of movement of these species is likely to be low along this pathway, but not necessarily insignificant, as has been shown for other species where a small number of introductions have resulted in invasive populations (for example Polygonum perfoliatum in the USA (IPANE 2016)).

The EU prioritisation scheme does not consider potential impacts which may be realised because of climate change scenarios, or indeed the potential for further spread and establishment because of future climatic projections. Of course, the effect of climate change on a species is a key consideration in any subsequent PRA but we suggest that the detailed analysis needed to address this issue is not suited to a prioritisation scheme. We reiterate the point made in Branquart et al. (2016) that a prioritisation scheme is no substitution for a comprehensive PRA but instead acts as a valuable tool to filter out those species where a PRA is not currently needed allowing efforts to focus on those species where a detailed RA is required.

In conclusion, in utilising the EU prioritisation process for alien plants, 19 species have been identified as high priorities for PRA. These species present a prominent risk to the EU, either now or in the future and thus warrant a full PRA. Whether these species are eventually included on the list of Union concern remains to be seen and will depend, in part, on the outcome of the subsequent PRA and decision makers of the Member States.

This research was funded in part by the European Commission, DG Environment LIFE funding under the project LIFE15 PRE-FR 001: Mitigating the threat of invasive alien plants in the EU through pest risk analysis to support the Regulation (EU) No 1143/2014.