ARTICLE IN PRESS
Journal for Nature Conservation 12 (2004) 171—179
www.elsevier.de/jnc
Assessing the risk of potentially invasive plant
species in central Europe
Ewald Weber, Daniel Gut
Swiss Federal Research Station of Fruit-Growing, Viticulture and Horticulture, CH-8820 Wädenswil, Switzerland
Received 10 September 2003; accepted 30 April 2004
KEYWORDS
Biological invasions;
Environmental
weeds;
Invasiveness;
Prediction;
Risk assessment
Summary
A risk assessment system was developed to assess the invasion potential of new
environmental weeds in central Europe. A pre-evaluation step excludes species that
are officially controlled, widespread, or intended for use in protected cultures only.
Species eligible for risk assessment are classified into three categories (high risk,
further observation required, low risk) by rating them according to various
biogeographical and ecological aspects. The rating system was validated by testing
47 well-known invasive plant species of temperate Europe and 193 exotic plants which
have failed to establish themselves in Switzerland. The overall accuracy was 65%.
Accuracy of correctly predicting invasive species was 77%, while accuracy of correctly
predicting non-invasive species was 62%. The proposed risk assessment protocol should
be understood as a first attempt for a European country and needs modifications.
These can only be achieved by applying the system in practice.
& 2004 Elsevier GmbH. All rights reserved.
Introduction
The spread of exotic species into natural communities is threatening native biological diversity and
the functioning of ecosystems, and is occurring at
an alarming rate (Pimentel, Lach, Zuniga, &
Morrison, 2000; Sandlund, Schei, & Viken, 1999;
Weber, 2003). The significance of invasive species
as a global environmental problem is widely
recognised, and article 8(h) of the Biodiversity
Convention asks for measures ‘‘to prevent the
introduction of, control or eradicate those alien
species which threaten ecosystems, habitats or
species’’. As more and more exotic (alien, nonnative) plant species are introduced and become
naturalised in most regions of the world, the
likelihood of new invasion events with subsequent negative ecological impacts on the native
Corresponding author. Geobotanisches Institut, Swiss Federal Institute of Technology, Zürichbergstrasse 38, CH-8044 Zürich,
Switzerland. Tel.: +41-1-632-7457; fax: +41-1-632-1215.
E-mail address: ewald.weber@env.ethz.ch (E. Weber).
1617-1381/$ - see front matter & 2004 Elsevier GmbH. All rights reserved.
doi:10.1016/j.jnc.2004.04.002
ARTICLE IN PRESS
172
communities increases rapidly (Vitousek, D’Antonio, Loope, Rejmánek, & Westbrooks, 1997). Thus,
in addition to already existing invasive species,
managers of protected areas and wildlife refuges
are increasingly facing the need to handle new
plant invasions. They also do not know how to react
to newly appearing exotic species for which
invasion potential is unknown.
To prevent new plant invasions, there is an
urgent need for the development of early warning
systems to determine the likelihood of a given
species becoming invasive and of methods to
conduct rapid assessments of the status of invaders
(Panetta & Scanlan, 1995; Sandlund et al., 1999;
Groves, Panetta, & Virtue, 2001; Wittenberg &
Cock, 2001; Andow, 2003). Preventive measures
would ideally consist of the prevention of entry of a
species, and the restriction of spread once the
species is present (Zamora, Thill, & Eplee, 1989;
Westbrooks, 1991). Both require knowledge on the
invasive characteristics of the species under consideration.
Most of the invasive weeds of a region were
originally introduced intentionally for economic
and other purposes and now their control incurs
high expenses (Pimentel et al., 2000). For example,
more than 70% of the naturalised exotic plants in
Australia had been introduced as ornamentals or for
agricultural purposes (Nairn, Allen, Inglis, & Tanner,
1996). These examples illustrate that it is possible
to avoid damage to native ecosystems by exotic
species and the associated costs if such harmful
species are not used and planted in the first place.
However, this step requires knowledge as to
whether a particular species will become invasive
where it has been introduced but is not yet
widespread, or where it is intended for introduction, e.g., the species must be recognised as
potentially invasive. Although prediction of invasiveness in plants is far from being obvious,
especially because of the absence of key characters, several attempts have been made to assess
the likelihood of invasion. Richardson, Cowling, and
Le Maitre (1990) defined functional groups based on
life-history traits important for invasion in a fireprone area, in order to assess invasive potential in
different species of Pinus. Five functional groups
could be distinguished, representing groups of
different invasion success. Taxa in the group of
highly invasive pines had small and wind dispersed
seeds, a short juvenile period, and fire-resilience.
Scott & Panetta (1993) developed a method of
predicting weed status for plants with southern
African origin naturalised in Australia, using data on
distribution, taxonomy and weed status outside
Australia. Variables with a high predictive value
E. Weber, D. Gut
were weed status, climatic range in the home
country, and the existence of congeneric weeds in
the home country. In order to produce a tool to
screen new introductions of woody plants for
invasion potential, Reichard & Hamilton (1997)
developed a decision tree that allows in a simple
way to allocate species into three classes (high risk,
low risk, further investigation required). The
United States Department of Agriculture (USDA)
uses a weed risk assessment protocol to regulate
species introductions and to prevent new weed
problems (Lehtonen, 1998). A similar scheme is
implemented in weed management in Australia
(Pheloung, 1996). Hiebert & Stubbendieck (1993)
developed a decision-making tool for prioritising
control efforts against exotic plant species by
ranking them according to their ecology, spread
potential, costs, and impacts on the native communities. Tucker & Richardson (1995) developed an
expert system for recognising woody plants potentially invading South African fynbos. In the case of
animals, Smallwood & Salmon (1992) developed a
simple rating system to assess the invasion potential of birds and mammals.
These examples illustrate that although ecological theory has not yet achieved predictive models,
it is possible to recognise potential harmful species
to at least some extent. In fact, sound models that
could be applied to predict invasiveness are needed
urgently, including quick and easy-to-perform
assessment protocols to screen exotic plant species
for their potential invasiveness. Such tools
can be of great help in prohibiting the use of
potential invasive species for example in horticulture and landscape architecture, in regulating
the plant trade, and to set priorities in managing
newly appearing exotic species in protected
areas.
In central Europe, a risk assessment system for
the impacts caused by invasive plant species does
not yet exist, although an ongoing discussion in
Germany on how to deal with invasive plant species
emphasises the necessity of such protocols (Doyle,
1999). The European Plant Protection Organisation
(EPPO) has, however, developed a pest risk assessment scheme that covers any pest organisms
including plants (http://www.eppo.org).
In this paper, we present a rating system to assess
the invasion potential of exotic plant species in a
central European country. We designed the rating
system to meet the specific needs for central
Europe and focused on plant species that may
become invasive in this region. We then validated
the rating system by applying it to exotic plant
species of Europe that had variable degrees of
success in becoming established.
ARTICLE IN PRESS
Risk assessment system for invasive plants
173
Methods
Risk assessment protocol
Existing protocols available from the literature
(Panetta, 1993; Hiebert & Stubbendieck, 1993;
Scott & Panetta, 1993; Pheloung, 1996; Reichard
& Hamilton, 1997) were modified according to the
specific demands of an European country. We chose
a ranking system because it is simple and most
appropriate for the purpose here. It was our aim to
set up a system that is widely applicable and
suitable for any species, and one that relies on
easily available data.
The procedure begins with a pre-evaluation step
(Fig. 1), in accordance with pest regulation
standards of the European Plant Protection Organisation (EPPO). The pre-evaluation step excludes
those species that are not justified for risk
assessment; e.g., species that are already widespread, under official control or listed as a
quarantine species, or unlikely to become natur-
Does the
species
occur in the
risk area?
Confirm species
idendity
START
alised for obvious reasons. Such a step is necessary
in order to ensure that the introduction of species
with considerable economic benefits is not prevented. Plant species considered suitable for risk
assessment include any exotic species that is not
yet present, has a restricted distribution in the risk
area, and is planned to be introduced and
commercially used on a large scale.
The rating system (Appendix A) allocates scores
to the species for biogeographical, ecological, and
experience-linked aspects. The scores of the 12
questions are summed up, and species are classified
into ‘‘high risk’’, ‘‘intermediate risk’’, and ‘‘low
risk’’.
Source of data
The necessary data for the species were obtained
from various sources. Status of the species as a
weed elsewhere was taken from Holm, Pancho,
Herberger, and Plucknett (1979) and Anonymous
(1992). Geographical distribution data for Europe
NO
YES
NO
Is the species
officially controlled
or a quarantine
species?
NO
Is the
species being
introduced intentionally for a special
purpose?
NO
YES
Is the species
widely distributed in the
risk area?
Is the species
planned to be
grown only
inside?
YES
Is there any
evidence that an
accidental introduction is unlikely?
NO
YES
YES
Stop
NO
YES
Stop
Did the species
arrive recently?
NO
Will the species
be grown on a
large scale?
Stop
NO
YES
YES
Evaluate risk
Is the species
solely a weed of
agricultural
areas?
NO
Evaluate risk
YES
Stop
Figure 1. Pre-evaluation of plant species for the risk assessment scheme (see Appendix A). The purpose of the preevaluation is to exclude those species that do not qualify for being assessed.
ARTICLE IN PRESS
174
and the world were obtained from Flora Europaea
online (http://www.rbge.org.uk/forms/fe.html),
the online database from the Germplasm Resources
Information Network (GRIN), National Germplasm
Resources Laboratory, Beltsville, Maryland (http://
www.ars-grin.gov/npgs/tax/index.html), from Meusel
& Jäger (1992) as well as from recent standard
floras. Ecological data were extracted from the
Ecological Flora Database of Great Britain (Fitter &
Peat, 1994), from Holm, Doll, Holm, Pancho, and
Herberger (1997), and from various weed floras and
other publications.
Climatic match (question 1) was decided by
determining the origin of the species and its native
distribution. Data on habitats and the ecology of
the species (question 11) and local abundance
(question 12) were taken from local floras of
Switzerland and other regions.
Validation
We validated the risk assessment scheme by testing
a set of well-known invasive plant species of
temperate Europe (Lambinon, 1997; Kowarik &
Schepker, 1998; Starfinger, Edwards, Kowarik, &
Williamson, 1998), and a set of unsuccessful exotic
species in Switzerland. The latter were taken from
Lauber & Wagner (1996) and consisted of species
which are reported as rarely or occasionally
naturalised, and have a restricted range within
Switzerland.
Accuracies were determined according to Smith,
Lonsdale, and Fortune (1999). Accuracy of correctly identifying invasive species is defined as
Ai ¼ ðIr =It Þ 100;
where Ir is the number of invasive species that were
recognised as such by the system, and It is the total
number of invasive species tested. Thus, Ai gives
the proportion of correct predictions among a set
of known invasive species. Similarly, the accuracy
of identifying non-invaders is
An ¼ ðNa =Nt Þ 100;
where Na is the number of non-invasive species
correctly predicted as such by the system, and Nt
the total number of non-invasive species tested.
The overall accuracy Ao is then calculated as
Ao ¼ ðNa þ Ir Þ=ðNt þ It Þ:
A further useful measure is the likelihood ratio of
the prediction method
LR ¼ ðIr =It Þ=ðNr =Nt Þ;
where Nr is the number of non-invasive species
predicted as being invasive by the system. Va-
E. Weber, D. Gut
lues41 imply that some predictive ability is
present, if LR ¼ 1; then the probability of correctly
identifying invasive species is 50%. For further
details, see Smith et al. (1999).
Results
The risk assessment procedure is given in Appendix
A. The results of the validation tests are summarised in Table 1. Out of the 47 invasive plant
species tested, 36 were recognised as being
invasive in the risk assessment, giving an accuracy
of 76.6% (Table 1). The species with the highest
scores were Ailanthus altissima, Helianthus tuberosus and Reynoutria japonica (Table 2). Species
that were classified by the rating system as
requiring further observation included species that
are serious plant invaders in some countries (e.g.
Rhododendron ponticum and Senecio inaequidens).
Species classified as non-invasive were generally
species with a restricted geographic distribution.
Out of the 193 non-invasive species tested, 119
were recognised as being non-invasive by the risk
assessment, leading to an accuracy of 61.6%.
Species that are recognised as being potentially
invasive by the system include many species with
clonal growth, e.g., Sagittaria latifolia, Pontederia
cordata, Asclepias syriaca and Glyceria striata
(Table 3).
The accuracy of correctly predicting non-invasive
species (61.6%) was less than the accuracy of
correctly predicting invasive species (76.6%). The
overall accuracy was closer to 50% than to 100%
(Table 1). However, the likelihood-ratio was fairly
high (14.8), indicating that the risk assessment has
some predictive character.
Table 1. Accuracy and likelihood ratio of the risk
assessment. For sources of the tested species, see text.
Accuracies were calculated according to Smith et al.
(1999)
Identified as
Invasive plant
species
Non-invasive
plant species
Low risk
Intermediate
High risk
Total number of
species
0
11
36
47
119
64
10
193
(0%)
(23.4%)
(76.6%)
(100%)
(61.6%)
(33.2%)
(5.2%)
(100%)
Accuracy for identifying invasive species: Ai=76.6%
Accuracy for identifying non-invasive species: An=61.6%
Overall accuracy: Ao=64.6%
Likelihood ratio: LR=14.8
ARTICLE IN PRESS
Risk assessment system for invasive plants
Table 2. Invasive plant species of Europe and their
rating as obtained by the risk assessment procedure. For
sources of species, see text
Species
Sum of scores
Risk class
Ailanthus altissima
Helianthus tuberosus
Reynoutria japonica
Reynoutria
sachalinensis
Solidago canadensis
S. gigantea
Arundo donax
Epilobium
adenocaulon
Robinia pseudacacia
Bidens frondosa
Cornus sericea
Heracleum
mantegazzianum
Rudbeckia laciniata
Crassula helmsii
Ludwigia grandiflora
Acer negundo
Elodea canadensis
E. densa
Ludwigia peploides
Lupinus polyphyllus
Pinus strobes
Prunus serotina
Myriophyllum
brasiliense
Parthenocissus
quinquefolia
Paspalum distichum
Rubus laciniatus
Erigeron annuus
Impatiens
glandulifera
Rhus typhina
Rumex longifolius
Oenothera biennis
Rosa rugosa
Veronica filiformis
Lonicera japonica
Rumex confertus
Spiraea douglasii
Amorpha fruticosa
39
39
39
39
III (High risk)
III
III
III
39
39
37
36
III
III
III
III
36
35
35
35
III
III
III
III
35
34
34
33
33
33
33
33
33
33
32
III
III
III
III
III
III
III
III
III
III
III
32
III
32
32
31
31
III
III
III
III
31
31
29
29
29
28
28
28
27
Rhododendron
ponticum
Galinsoga ciliata
Gunnera tinctoria
Senecio inaequidens
Vaccinium
macrocarpon
Cyperus eragrostis
Impatiens parviflora
Physocarpus
opulifolius
Aster squamatus
Lysichiton
americanum
27
III
III
III
III
III
III
III
III
II (further
observation)
II
26
26
26
26
II
II
II
II
25
25
25
II
II
II
24
23
II
II
175
Discussion
The objective of a risk assessment for invasive
weeds is to decide which species should be listed on
national noxious weed lists and to decide which
new species infestations should be controlled or
removed in order to prevent their spread and
associated ecological consequences. Predicting
plant invasiveness is, however, limited due to three
facts: (1) the high ecological and taxonomic
diversity of invasive plants, (2) the lack of
ecological data for most plant species, and (3) the
variation in invasiveness within the range of a
species. Exotic invasive plants form an extremely
heterogenous group of plants comprising of all
lifeforms and many families (Daehler, 1998; Pysek,
1998), thus making generalisations about the
taxonomic position of invasive plant species difficult. The only conclusion that can be drawn is that
certain families are more likely to contribute to
invasive plants than other families (Daehler, 1998;
Pysek, 1998). The limited amount of available data
on the demography and ecology of a species, e.g.,
relative growth rate, seed weight, length of
juvenile period in trees, and crop size, sets serious
constraints to the identification of common characters of invasive plant species. Here, databases of
basic ecological features among plant species are
badly needed, as well as comparative studies under
defined conditions and including a wide array of
different species (Perrins, Williamson, & Fitter,
1992a). Exotic plant species can be invasive in one
part of their introduced range but not in another,
thus complicating the decision whether to grant
invasive status. Also, the perception of a species as
being invasive or not depends on a person’s
perspective, and varies between agriculturists and
conservationists (Perrins, Williamson, & Fitter,
1992b).
Despite these restrictions, the rapidly increasing
number of exotic species urgently requires decision-making tools for the management of plant
invasions and the regulation of plant trade in order
to reduce the risks of new introductions (Sandlund
et al., 1999). Existing risk assessments must therefore rely on surrogates for the species’ ecology, and
on the experience of the species’ colonisation
success in the past. Characters found to be
especially suitable are the extent of the geographic
range, and whether or not the species shows weedy
behaviour elsewhere (Panetta, 1993; Pheloung,
1996).
Based on this background, we developed a simple
rating system for the identification of invasive plant
species in central Europe. Rating systems have
proven to be useful in the recognition of pest
ARTICLE IN PRESS
176
E. Weber, D. Gut
Table 3. Unsuccessful exotic plant species in Switzerland, and their ratings as obtained by the risk assessment.
Species were taken from Lauber and Wagner (1996)
Species
Sum of scores
Sagittaria latifolia
Pontederia cordata
Artemisia dracunculus
Asclepias syriaca
Rorippa austriaca
Glyceria striata
Oenothera parviflora
Rumex longifolius
Aster lanceolatus
Lysimachia punctata
Lonicera tatarica
Mahonia aquifolium
Rubus laciniatus
Cyperus eragrostis
Sisyrinchium montanum
Cynodon dactylon
Erigeron karvinskianus
Gleditsia triacanthos
Lupinus angustifolius
Morus alba
Salvia verbenaca
Sorghum halepense
Xanthium spinosum
Paspalum dilatatum
Typha laxmannii
Verbascum phoeniceum
Ambrosia artemisiifolia
Anethum graveolens
Atriplex hortensis
Beta vulgaris
Chenopodium ambrosioides
Dipsacus laciniatus
Fagopyrum tataricum
Hemerocallis lilio-asphodelus
Malva verticillata
Nicandra physalodes
Nicotiana tabacum
33
32
31
31
31
29
29
29
28
28
27
27
27
26
26
25
25
25
25
25
25
25
25
24
24
24
23
23
23
23
23
23
23
23
23
23
23
mammals (Smallwood & Salmon, 1992) and have
been implemented in Australian weed risk assessments (Pheloung, 1996). The proposed risk assessment uses biogeographical and ecological
characters that should be fairly easy to obtain.
Emphasis was given to the environments in which a
new species are expected to occur.
The rather low overall accuracy in predicting
non-invasives (61.6%) might be due to the low
number of species tested, and the variation in
degree of invasion among these. However, the
likelihood ratio was rather high in our study,
indicating that the procedure has some predictive
value (Smith et al., 1999). The accuracy of
correctly identifying invasive plants of our system
Sum of scores
Opuntia humifusa
Potentilla fruticosa
Rubus phoenicolasius
Rumex patientia
Rumex thyrsiflorus
Silene dichotoma
Sorghum vulgare
Syringa vulgaris
Ulmus laevis
Allium scorodoprasum
Brassica juncea
Diospyros lotus
Lycium barbarum
Allium fistulosum
Amaranthus deflexus
Amaranthus graecizans
Artemisia annua
Avena barbata
Bromus diandrus
Chrysanthemum segetum
Coriandrum sativum
Coronopus didymus
Cotoneaster horizontalis
Cuscuta campestris
Galium rubioides
Lathyrus sativus
Nepeta nepetella
Polypogon monspeliensis
Ranunculus muricatus
Rumex palustris
Salvia officinalis
Scorzonera hispanica
Scutellaria altissima
Tolpis barbata
Torilis nodosa
Tragopogon porrifolius
Tribulus terrestris
23
23
23
23
23
23
23
23
23
22
22
22
22
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
was higher than the accuracy of correctly identifying non-invasive plants. However, since the outcome of any validation depends on the number and
kind of species tested, the risk assessment must be
tested in practice and further developed as
necessary. For example, the answers to the questions of the rating system may be weighted in
various ways, influencing the outcome of the
system and its accuracy. Further studies are
necessary to explore how accuracy changes with
different weightings, and whether the accuracy can
be improved.
The system classified many serious plant invaders
of central Europe into the high risk class.
There were, however, exceptions. For example,
ARTICLE IN PRESS
Risk assessment system for invasive plants
Rhododendron ponticum, although considered as a
highly invasive species in the British Isles (Fuller &
Boorman, 1977) was classified as requiring further
observation. This might be due to the restricted
ranges, both the native and the introduced. Other
exotic species that currently have a very restricted
range in Europe were classified into risk class II
(intermediate risk). These include many species
with clonal growth, e.g., Cyperus eragrostis,
Gunnera tinctoria and Lysichiton americanum.
Among the unsuccessfully established exotic plant
species of Switzerland, species with clonal growth
again achieved high scores. Whether these species
pose potential problems must be evaluated by
studying the ecology of these species.
It was not the aim to test the practicability of the
rating system. Clearly, such a system must be
applied and its practicability evaluated both by
scientists and practical experts involved in the
management of invasive plants.
Acknowledgements
This study was partially funded by the Swiss Plant
Protection Foundation. We thank two anonymous
referees for valuable comments on the manuscript.
Appendix A.
Risk assessment scheme for assessing the
invasion potential of environmental weeds in
central Europe.
Risk assessment
Answer the following questions and sum up the
scores given on the right side. If not otherwise
indicated, only one answer applies.
1. Climatic match
Does the known geographical distribution of the
species include ecoclimatic zones similar with
those of the risk area?
No
0
Yes
2
2. Status of species in Europe
Is the species native to Europe?
Yes
0
No
2
3. Geographic distribution in Europe
In how many countries does the species occur?
Species occurs in 0 or 1 country
1
Species occurs in 2–5 countries
2
Species occurs in45 countries
3
177
4. Range size of global distribution
How is the size of the global range (native and
introduced)?
Range is small, species is restricted
0
to a small area within one continent
Range is large, extending over more
3
than 151 latitude or longitude in one
continent or covers more than one continent
5. History as an agricultural weed elsewhere
Is the species reported as a weed from somewhere
else?
No
0
Yes
3
6. Taxonomy
Does the species have weedy congeners?
No
0
Yes
3
7. Seed viability and reproduction
How many seeds do the species approximately
produce?
If the species is present in the risk area,
this question refers to plants within the risk area.
If the species is present in Europe, this question
refers to plants within the European range. If the
species is not present in Europe, this question
refers to the native or introduced range of the
species.
Few seeds or no viable seeds
1
Many seeds
3
Do not know
2
8. Vegetative growth
Allocate species to one of the following. If more
than one statement applies, take the one with the
highest score.
Species has no vegetative growth that leads 0
to lateral spread
If a tree or shrub, species has the ability to 2
resprout from stumps or stem layering, or
stems root if touching the ground
Species has bulbs or corms
1
Species has well developed rhizomes and/ 4
or stolons for lateral spread
Species fragments easily, fragments can be 4
dispersed and produce new plants
Other or do not know
2
9. Dispersal mode
Allocate species to one of the following. If more
than one statement applies, take the one with the
highest score.
Fruits are fleshy and smaller than 5 cm in 2
diameter
Fruits are fleshy and larger than 10 cm in 0
length or diameter
Fruits are dry and seeds have well devel- 4
oped structures for long-distance dispersal by
wind (pappus, hairs, wings)
ARTICLE IN PRESS
178
Fruits are dry and seeds have well-devel- 4
oped structures for long-distance dispersal by
animals (spikes, thorns)
Species has mechanisms for self-dispersing 1
Other or do not know
2
10. Lifeform
What is the lifeform of the species?
Species is a small annual (o 80 cm)
0
Species is a large annual (480 cm)
2
Species is a woody perennial
4
Species is a small herbaceous perennial
2
(o 80 cm)
Species is a large herbaceous perennial
4
(480 cm)
Species is a free floating aquatic
4
Other
2
11. Habitats of species
Allocate species to one of the following. If more
than one statement applies, take the one with the
highest score.
Riparian habitats
3
Bogs/swamps
3
Wet grasslands
3
Dry (xeromorphic) grasslands
3
Closed forests
3
Lakes, lakeshores, and rivers
3
Other
0
12. Population density
What is the local abundance of the species?
If the species is present in the risk area, this
question refers to plants within the risk area. If the
species is present in Europe, this question refers to
plant within the European range. If the species is
not present in Europe, this question refers to the
native or introduced range of the species.
Species occurs as widely scattered indivi- 0
duals
Species forms occasionally patches of high 2
density
Species forms large and dense monocul- 4
tures
Total score
Identify risk class
3–20
Low risk — Species is unlikely to pose a
threat to natural communities.
21–27
Intermediate risk — Species requires
further observation.
28–39
High risk — Species is likely to become a
threat to natural communities if naturalised.
References
Andow, D. A. (2003). Pathways-based risk assessment of
exotic species invasions. In Ruiz, G. M., & Carlton, J.
E. Weber, D. Gut
T. (Eds.), Invasive species: vectors and management strategies (pp. 439–455). Washington: Island
Press.
Anonymous. (1992). Important crops of the world and
their weeds. Business Group Crop Protection, Bayer
AG, Leverkusen, Germany.
Daehler, C. C. (1998). The taxonomic distribution of
invasive angiosperm plants: ecological insights and
comparison to agricultural weeds. Biological Conservation, 84, 167–180.
Doyle, U. (1999). Alien organisms in Germany. Federal
Environmental Agency (Umweltbundesamt) Research
Report 298 82 779/05, Berlin, Germany.
Fitter, A. H., & Peat, H. J. (1994). The ecological flora
database. Journal of Ecology, 82, 415–425.
Fuller, R. M., & Boorman, L. A. (1977). The spread and
development of Rhododendron ponticum L. on dunes
at Winterton, Norfolk, in comparison with invasion by
Hippophae rhamnoides L. at Saltfleeby, Lincolnshire.
Biological Conservation, 12, 83–94.
Groves, R. H., Panetta, FD., & Virtue, J. G. (2001). Weed
risk assessment. Collingwood, Australia: CSIRO Publishing.
Hiebert, R. D., & Stubbendieck, J. (1993). Handbook for
ranking exotic plants for management and control.
Natural Resources Report 93/08, US Department of
the Interior, National Park Service, USA.
Holm, L., Doll, J., Holm, E., Pancho, J., & Herberger, J.
(1997). World weeds. Natural histories and distribution. New York: Wiley.
Holm, L., Pancho, J. V., Herberger, J. P., & Plucknett, D.
L. (1979). A geographical atlas of world weeds. New
York: Wiley.
Kowarik, I., & Schepker, H. (1998). Plant invasions in
northern Germany: human perception and response.
In Starfinger, U., Edwards, K., Kowarik, I., & Williamson, M. (Eds.), Plant invasions: ecological mechanisms
and human responses (pp. 109–122). Leiden: Backhuys
Publishers.
Lambinon, J. (1997). Introduction of non-native plants
into the natural environment. Nature and environment, No. 87. Strasbourg, France: Council of Europe
Publishing.
Lauber, K., & Wagner, G. (1996). Flora Helvetica. Bern,
Switzerland: Paul Haupt.
Lehtonen, P. (1998). Pennisetum clandestinum. Weed risk
assessment. USA: US Department of Agriculture.
Meusel, H., & Jäger, E. J. (1992). Vergleichende
Chorologie der zentraleuropäischen Flora, Vol. 3.
Jena: Gustav Fischer.
Nairn, M. E., Allen, P. G., Inglis, A.R., & Tanner, C. (1996).
Recent incursions of quarantine concern. In Australian
quarantine: a shared responsibility (pp. 218–237).
Department of Primary Industries & Energy, Canberra,
Australia.
Panetta, F. D. (1993). A system of assessing proposed
plant introductions for weed potential. Plant Protection Quarterly, 8, 10–14.
Panetta, F. D., & Scanlan, J. C. (1995). Human involvement in the spread of noxious weeds: what plants
ARTICLE IN PRESS
Risk assessment system for invasive plants
should be declared and when should control be
enforced? Plant Protection Quarterly, 10, 69–74.
Perrins, J., Williamson, M., & Fitter, A. (1992a). Do
annual weeds have predictable characters? Acta
Oecologica, 13, 517–533.
Perrins, J., Williamson, M., & Fitter, A. (1992b). A survey
of differing views of weed classification: implications
for regulation of introductions. Biological Conservation, 60, 47–56.
Pheloung, P. C. (1996). Predicting the weed potential of
plant introductions. 11th Australian Weeds Conference Proceedings (pp. 458–461).
Pimentel, D., Lach, L., Zuniga, R., & Morrison, D. (2000).
Environmental and economic costs of nonindigenous
species in the United States. BioScience, 50, 53–65.
Pysek, P. (1998). Is there a taxonomic pattern to plant
invasions? Oikos, 82, 282–294.
Reichard, S. H., & Hamilton, C. W. (1997). Predicting
invasions of woody plants introduced into North
America. Conservation Biology, 11, 193–203.
Richardson, D. M., Cowling, R. M., & Le Maitre, D. C.
(1990). Assessing the risk of invasive success in Pinus
and Banksia in South African mountain fynbos. Journal
of Vegetation Sciences, 1, 629–642.
Sandlund, O. T., Schei, P. J., & Viken, A. (1999). Invasive
species and biodiversity management. Dordrecht:
Kluwer Academic Publishers.
Scott, J. K., & Panetta, F. D. (1993). Predicting the
Australian weed status of southern African plants.
Journal of Biogeography, 20, 87–93.
179
Smallwood, K. S., & Salmon, T. P. (1992). A rating system
for potential exotic bird and mammal pests. Biological
Conservation, 62, 149–159.
Smith, C. S., Lonsdale, W. M., & Fortune, J. (1999). When
to ignore advice: invasion predictions and decision
theory. Biological Invasions, 1, 89–96.
Starfinger, U., Edwards, K., Kowarik, I., & Williamson, M.
(1998). Plant invasions. Ecological mechanisms and
human responses. Leiden: Backhuys Publishers.
Tucker, K. C., & Richardson, D. M. (1995). An expert
system for screening potentially invasive alien plants
in south African fynbos. Journal of Environmental
Management, 44, 309–338.
Vitousek, P., D’Antonio, C. M., Loope, L. L., Rejmánek,
M., & Westbrooks, R. (1997). Introduced species: a
significant component of human-caused global
change. New Zealand Journal of Ecology, 21, 1–16.
Weber, E. (2003). Invasive plant species of the world: a
reference guide to environmental weeds. Wallingford:
CABI Publishing 548pp.
Westbrooks, R. G. (1991). Plant protection issues. I. A
commentary on new weeds in the United States. Weed
Technology, 5, 232–237.
Wittenberg, R., & Cock, M. J. W. (Eds.). (2001). Invasive
alien species: a tool kit of best prevention and
management practices. Wallingford, UK: Global Invasive Species Programme and CAB International.
Zamora, D. L., Thill, DC., & Eplee, R. E. (1989). An
eradication plan for plant invasions. Weed Technology,
3, 2–12.