Assessment of Invasive Naturalized Plants in South-East Queensland
Total Page：16
File Type：pdf, Size：1020Kb
Plant Protection Quarterly Vol.17(1) 2002 27 Queensland, most natural areas have experienced some form of disturbance Assessment of invasive naturalized plants in (e.g. grazing, altered fire regimes, nutrient south-east Queensland enrichment and partial clearing) and therefore the list includes some species which benefit from disturbance. Also, George N. Batianoff and Don W. Butler, Queensland Herbarium, some invasive plants are ‘broad spectrum’ Environmental Protection Agency, Brisbane Botanic Gardens, Mt. Coot-tha weeds. For example, lantana (Lantana Road, Toowong, Queensland 4066, Australia. camara) and green panic (Panicum maxi- mum) can be categorized as suburban, ag- ricultural and environmental weeds. Summary Methods A list of 200 invasive naturalized species anecdotal or inadequate, ranking systems Relevant literature (Kleinschmidt and was prepared as a working document for need to be pragmatic and utilize botanical Johnson 1977, Hobbs and Huenneke 1992, the compilation of an environmental information currently available. The ob- Parsons and Cuthbertson 1992, Crawley et weed list for south-east Queensland. ject of this project was to identify which of al. 1996, Kleinschmidt et al. 1996, Laz- Species were selected from the 1060 natu- the 1060 plant species the Queensland arides et al. 1997, Anon 2000, Thorp and ralized taxa and were ranked based upon Herbarium has recorded as naturalized in Lynch 2000, Groves et al. 2000, Randall invasiveness and frequency in non-agri- south-east Queensland could be catego- 2000, Groves et al. 2001 and Williamson cultural areas and/or remnant natural ar- rized as ‘environmental’ weeds. This list 1999, 2001) was consulted to assist our eas. The list was compiled using records provides a working base from which the understanding of naturalized plant inva- of 1413 vegetation sites, 10 163 her- Southeast Queensland Environmental siveness and current categorizations of barium specimens and ranking scores of Weeds List will be compiled in consulta- weeds. Common names used mostly fol- thirteen members of an expert assess- tion with the wider community. low Shepherd et al. (2001). ment panel. Environmental weeds are defined as The study area (Figure 1) was defined One third of the 200 species listed invasive plants occurring in relatively un- by the Southeast Queensland Environ- were categorized as highly invasive. disturbed, mostly native vegetation. The mental Weeds Strategy Working Group Ninety percent were intentionally intro- Australia-wide weed prioritization Weeds and was based on Local Government Ar- duced for landscaping, agriculture and of National Significance (WONS) identi- eas. The task of distinguishing weeds of aquariums. One half of the list consist of fied and used two criteria, invasiveness disturbance (ruderal) from weeds of natu- herbs, including 34 species (17%) of and impact (Thorp and Lynch 2000). In ral areas (environmental) was addressed grasses and 19 species (9%) of aquatic this assessment, evidence of invasiveness by categorizing species invasiveness and weeds. Woody weeds (shrubs and trees) or potential invasiveness was the primary referring to a panel of thirteen weed scien- are a problem in south-east Queensland, criterion for selection, and frequency data tists and field botanists. The expert assess- representing 37% (75 species). The ten used as a secondary criteria. Invasiveness ment panel included members of four or- highest ranked invasive species in de- is related to a plant’s vigour, ability to es- ganizations. Tom Anderson, Trevor scending order were lantana (Lantana tablish and colonize existing vegetation Armstrong and Dane Panetta are weed camara), groundsel bush (Baccharis and to develop substantial populations scientists at Alan Fletcher Research Sta- halimifolia), mother-of-millions (Bryo- (Randall 2000). Assessment of all 1060 tion, Department of Natural Resources phyllum delagoense), cat’s claw creeper species for environmental impact was and Mining. George Batianoff, Anthony (Macfadyena unguis-cati), Madeira vine considered to be impractical because of Bean, Paul Forster, Ailsa Holland, Bill (Anredera cordifolia), ornamental aspara- insufficient data. McDonald, Sue Phillips and Kathy gus (Asparagus africanus), Chinese celtis In this study ‘natural’ vegetation is de- Stephens are botanists at Queensland Her- (Celtis sinensis), camphor laurel (Cinna- fined as plant communities where more barium, Environmental Protection Agency. momum camphora), broad-leaf pepper than 70% of the cover of vegetation is com- Michael Olsen and John Swarbrick are en- tree (Schinus terebinthifolius) and sal- prised of indigenous species and vegeta- vironmental consultants representing pri- vinia (Salvinia molesta). tion structure approaches that of remnant vate enterprise. Paul Grimshaw is a tech- communities. Natural areas threatened by nical officer representing Queensland Introduction invasive species include remnant vegeta- Parks and Wildlife Service. Risk assessments and development of tion in nature reserves, natural wildlife The panel members were presented methodologies for prioritizing weedy spe- corridors, wetlands, waterways, seashore with a list of 1060 naturalized species cies are recognized as useful weed man- and roadside native communities. Rej- based on specimens held at the Queens- agement tools (Thorp and Lynch 2000, mánek (1989) suggested that ‘there seems land Herbarium. The questionnaire in- Randall 2000, Groves et al. 2001, Tye 2001). to be no community without some degree cluded four effective categories of inva- Because information on distribution, fre- of natural disturbance’ and few without siveness, plus a category for early records quency, autecology and impact is often signs of human disruptions. In south-east as outlined in Table 1. The weed scientists Table 1. Categories of plant invasiveness offered to the assessment panel. Score Description Notes 5 highly invasive, forms monocultures e.g. Lantana camara, Celtis sinensis, Cabomba caroliniana 4 generally invasive escaping from cultivation and spreading to natural areas e.g. Asparagus spp., Ochna serrulata, Pinus elliottii 3 common, invasive? (needs disturbance) weeds of suburban and agricultural disturbance e.g. Alternanthera pungens, Conyza spp., Medicago polymorpha 2 infrequent, poorly invasive e.g. Aloe arborescens 1 poorly known and/or early records only scores of 1 were not included in calculation of average scores 28 Plant Protection Quarterly Vol.17(1) 2002 and botanists in the panel estimated plant Water weeds were poorly represented Investigations of the source of introduc- invasiveness based on field observations in vegetation sites and therefore their tion indicate that 64% of the 200 listed spe- and understanding of the biological per- ranking was primarily based upon the cies were ornamental garden plants, 21% formance of the species. The scores were opinions of four experts. Prickly and large were agricultural plants, 5% were averaged and species with average scores plants (e.g. palms) were also under-repre- aquarium plants and 10% were of un- of 3.0 or less were not considered further, sented in the specimen data. known source. Relatively few species leaving 443 species. were considered extremely invasive by Three aspects of distribution were also Results and discussion the assessment panel. Table 2 illustrates considered: (i) frequency as specimens Distribution of naturalized species records that about one third of the 200 species had collected from the study area (HERBRECS presented in Figure 1 indicates a pattern average scores greater than 4, and only 27 = Queensland Herbarium Records Data- of high density of records within areas of species (13.5%) were regarded as ‘highly base System), (ii) frequency in detailed high development e.g. around Beaudesert, invasive’ (>4.5). Table 2 indicates that vegetation sites from the study area Toowoomba, Brisbane and Nambour and woody weeds (shrubs and trees) are a (CORVEG = Queensland Herbarium eco- coastal areas. The sub-coastal areas from problem in south-east Queensland, repre- logical site database, sites with compre- Esk to Mundubbera and further north are senting 37% (75 species) of the 200 listed hensive species lists only) and (iii) number poorly sampled and/or represent areas invasive species. However, the largest life of south-east Queensland sub-regions (ten with relatively low numbers of weed spe- form category of 100 species (50%) be- divisions, as per Young and Dillewaard cies. Similarly, taxa records in one degree longed to herbs. These herbs include 34 1999) within which a species had been re- grids across Queensland identified strong (17%) species of grasses and 19 species corded in HERBRECS or CORVEG. Site correlation between high number of natu- (9%) of aquatic herbs. The frequency of the data was predominantly restricted to na- ralized species records and high density of life forms was broadly similar to that of tive vegetation. Whereas Queensland people (Batianoff et al. 2000). the combined native and naturalized flora Herbarium specimen data also included The 200 more invasive species and their of the neighbouring area of Port Curtis records from heavily disturbed areas. ranks are presented in the Appendix. District (Batianoff and Dillewaard 1988). Invasiveness and the three aspects of distribution were combined into a single prioritization. The three aspects of distri- bution were combined into a frequency index by summing the three ranks and di- # Gladstone viding the sum for each species by the S S