Mark A. Hamilton, Hillary Cherry and Peter J. Turner
Pests and Weeds Team, New South Wales National Parks and Wildlife Service, Office of Environment and Heritage, PO Box 1967, Hurstville, New South Wales 1481, Australia.
Hamilton, M.A., Cherry, H. and Turner, P.J. (2015) Hawkweed eradication from NSW: Could this be ‘the first’? Plant Protection Quarterly 30(3), 110-115.
Eradication is the complete and permanent elimination of all wild populations of an organism from a defined area in a given timeframe. There are currently no documented weed eradications from New South Wales. Hawkweeds (Hieracium spp.) are perennial herbs native to Eurasia that are serious weeds in many temperate and sub-alpine areas of the world. Four hawkweed species are present in Australia, and modelling indicates at least 27 million ha of south east Australia is susceptible to invasion by orange (H. aurantiacum L.) and mouse-ear hawkweed (H. pilosella L.). In New South Wales, small infestations of the Class 1 noxious weeds, orange hawkweed and mouse-ear hawkweed are present in sub-alpine and alpine regions of Kosciuszko National Park, the only known location in New South Wales. Since 2003, 8.21 ha of orange hawkweed has been found in a remote and rugged area of the Great Dividing Range that extends across approximately 8 951 ha.
The New South Wales National Parks and Wildlife Service, with a range of partners, are working to eradicate hawkweeds from the state. This involves strategic surveillance and rapid response to prevent reproduction, control infestations and repeated follow-up control activities. Surveying treated sites is critical because orange hawkweed seed viability is up to 5 years and, in the past, control efficacy proved variable due to limited herbicide translocation through stolons. However, through extensive monitoring, evaluation and adaptive management, a suite of control methods are now resulting in efficient and effective hawkweed control. Concerted efforts have reduced orange hawkweed from 1.36 ha extant in 2010 to a total of 0.02 ha extant in 2015. These efforts have been supported by surveys several times each season of all known sites across the 8.21 ha management area to ensure no plants re-emerge and reproduce.
The key to hawkweed eradication now lies in finding the very last plant, which means the remaining large, remote areas must be surveyed. New and innovative tools are being employed to ensure delimitation, such as using unmanned aerial vehicles to detect plants over large areas, and training scent detection dogs to locate the hard-to-find plants. These techniques have reliably detected orange hawkweed in trials and will be operational in the 2015/16 field season. This paper details the New South Wales hawkweed eradication program and how these new tools may allow us to make orange and mouse-ear hawkweed the first documented weed eradications from New South Wales.
Eradication is the complete and permanent elimination of all wild populations of an organism from a defined area in a given timeframe (Bomford and O’Brien 1995). It is usually only attempted when there is little or no chance of reinvasion (Bomford and O’Brien 1995). For eradication of plant species, the key operational objectives are to delimit the infestation, halt reproduction, control all above-ground matter, and completely exhaust the seedbank. Eradication can only be declared successful when the species is not detected for a period equal to or greater than its seed longevity.
Eradication is a weed management strategy that is particularly appealing because other alternatives such as containment or impact reduction require permanent, ongoing investment of resources (Grice et al. 2013). This is supported by the Natural Resources Commission’s recent review of weed management in New South Wales which highlighted a need to improve responses to new weed incursions, as early and effective responses can be the difference between successful eradication and ongoing management (NRC 2014).
Orange hawkweed (Hieracium aurantiacum L.) and mouse-ear hawkweed (Hieracium pilosella L.) are Eurasian, stoloniferous, perennial herbs in the Asteraceae family that have recently invaded natural vegetation in the Victorian and New South Wales Alps (Williams and Holland 2007). Under the New South Wales Noxious Weeds Act 1993, all hawkweed species are Class 1 State Prohibited noxious weeds. Orange hawkweed is also on the National Alert List for Environmental Weeds (Australian Government 2015), a list of 28 non-native plants at the early stages of establishment with high potential impacts and feasibility of eradication, and recognised nationally as an Agricultural Sleeper weed (Cunningham et al. 2003). Both species produce wind-dispersed seeds that may travel large distances, and may occasionally be dispersed by animals, vehicles and humans (Rinella and Sheley 2002). Stergios (1976) suggests long-distance dispersal is a rare event, finding that 83% of seed was deposited within 2 m of parent plants. Modelling by Williams et al. (2008) confirms this, but shows long distance dispersal, although rare, is possible: in very rare instances seed can travel up to 5 km from the parent source. It is postulated that Victorian populations of orange hawkweed 1–2 km distant from parent sources have arrived through wind dispersal (Cousens et al. 2012). In addition to occasional long distance dispersal, hawkweeds can become locally abundant through stolon growth that can form a dense mat, exclusive of other ground cover (Morgan 2000, Espie 2001).
These biological traits have contributed to hawkweeds becoming major weeds in the United States of America, Canada, Japan and New Zealand. In New Zealand, hawkweeds have invaded more than 6 million ha of the South Island, particularly on the Canterbury Plains, and have significantly decreased carrying capacity over large areas (Espie 2001). In Australia, hawkweeds are at the early stages of establishment, with small populations found in Tasmania, Victoria and New South Wales, but if left unchecked, invasion of south east Australia could have disastrous ecological (Worboys and Good 2011) and economic impacts. For example, 27 million ha are susceptible to invasion by orange hawkweed (Cunningham et al. 2003), and economic modelling shows a conservative potential annual loss to the grazing sector of $66 million (Brinkley and Bomford 2002, and adjusted to account for Consumer Price Index [Reserve Bank of Australia]).
In New South Wales, H. aurantiacum only occurs in nine distinct locations in sub-alpine regions of central Kosciuszko National Park, centred on the Jagungal Wilderness. Hieracium pilosella occurs as a small population on the Main Range of Kosciuszko National Park (Figure 1); this population is not linked to the H. aurantiacum populations. Kosciuszko National Park is the largest conservation reserve in New South Wales and is significant for its unique landscape and biodiversity values. Since detection of orange hawkweed in New South Wales in 2003, the New South Wales National Parks and Wildlife Service has invested heavily in surveillance, treatment, trialling herbicide techniques, increasing awareness and understanding the ecology of the weed (Caldwell and Wright 2014). The Victorian government is also working towards eradication of orange, mouse-ear and king devil (H. praealtum Vill. ex Gochnat) hawkweeds (Curran and Primrose 2012), and Victoria and New South Wales sit on the National Hawkweed Working Group that coordinates efforts towards national eradication.
Notwithstanding many eradication attempts, unlike pest animals, there are few documented examples of successful weed eradications (Simberloff 2003). In New South Wales, there are no documented cases of statewide eradications, and very few examples exist Australia wide (Panetta 2009). In New South Wales, hawkweeds (Hieracium species) are the subject of an eradication program that the authors believe has a high likelihood of success. This paper outlines the National Parks and Wildlife Service eradication program for orange and mouse-ear hawkweeds, the progress to date, and the innovative control and detection techniques currently being employed to achieve eradication success. Note, although this paper largely deals with the longstanding orange hawkweed eradication program, the eradication progress to-date of the recently discovered mouse-ear hawkweed infestation will also be covered.
Eradication attempts require significant commitment and resourcing. Panetta et al. (2011) suggested that the average time to eradicate a Class 1 weed from Queensland was 18.2 years, and the average total cost of eradication was estimated at $2.997 million weed-1. Thus, there is a great need to evaluate the progress towards the eradication objective (Panetta and Brooks 2008).
From 2003, initial National Parks and Wildlife Service control efforts were devoted to treating H. aurantiacum infestations at Round Mountain, the first infestation to be discovered. As new invasion foci were discovered at Ogilvies Quarry, Ogilvies Airstrip and Cool Plain, and the nature and scale of the problem became clear, a dedicated program coordinator was employed in 2009, and the program became increasingly well-resourced. This greater coordination and resourcing enabled increased delimitation efforts (including significant surveillance conducted by volunteers), and greater frequency of return to sites to ensure reproduction was prevented and herbicide efficacy was monitored (Caldwell and Wright 2012).
The current National Parks and Wildlife Service hawkweed eradication program is comprised of six key activities: 1) treating known infestations; 2) revisiting known infestations at regular intervals to ascertain further control requirements and to remove reproductive material; 3) searching for possible new infestations using wind dispersal data and habitat suitability, with the assistance of staff and volunteers; 4) undertaking monitoring to determine herbicide efficacy and to adaptively manage infestations; 5) collaborating with researchers to understand hawkweed biology; and 6) improving public understanding and support for the eradication program.
A total of nine distinct orange hawkweed locations have now been identified across an extent of 8 951 ha (the management area currently totals 8.21 ha). This extent was determined using the minimum convex polygon method (Mohr 1947), a simple technique used to determine a species’ spatial extent or home range based on available point locations. For orange hawkweed, this represents a rough approximation of the area over which the species may occur, and is useful in directing surveillance efforts. However, the likelihood of the species occurring across this extent is variable, due to dispersal pathways, and habitat preferences, including disturbed areas associated with past infrastructure.
At the end of the 2014/15 season, the area of extant orange hawkweed amounted to 0.02 ha. This is a notable achievement and represents a 98.5% reduction in hawkweed area since 2010/11 season (Figure 2). In addition, the cumulative area of orange hawkweed infestations (being the area of all infestations discovered since 2003, despite most being controlled), totalling 8.21 ha, has increased very little since the 2010/11 season (Figure 3). This plateau of cumulative infested area over time is one of the best measures of whether delimitation is occurring (Panetta and Brooks 2008), and has occurred despite a concomitant increase in surveillance effort. The corollary of this is that, due to a potential 5-year seedbank, the effort required to revisit all sites at sufficient frequency throughout the season has increased substantially.
The National Parks and Wildlife Service has utilised volunteers, field staff and contractors to conduct extensive ground surveillance for orange hawkweed. Approximately 1 197 ha has been searched for orange hawkweed since 2010 with the assistance of over 236 volunteers. Importantly, as surveillance efforts have increased, the detection rate has decreased. As more ground is searched, fewer hawkweed sites are being discovered. Figure 4 illustrates this trend with a decreasing proportion of new sites being discovered over time. Almost half of all orange hawkweed sites are now classified as ‘inactive’ (Figure 4). These require ongoing monitoring for hawkweed re-emergence but no treatment as no above-ground hawkweed is present. The proportion of ‘inactive’ sites is generally increasing over time and the 2014/15 season was the first year that a small proportion of sites (4%) were classified as ‘inactive [for] 5 years’. These sites are where hawkweed is thought to be locally eradicated, meaning absent for 5 years or more (the period equal to seed longevity). This trend of an increasing proportion of sites being locally eradicated is expected to continue, with the greater herbicide efficacy that the program is now achieving. The greater kill rates result in fewer sites with re-emerging hawkweed following treatment.
The progress towards eradication outlined above is in part due to the increased herbicide efficacy that has been achieved. As with many new and highly invasive weeds, early National Parks and Wildlife Service control efforts resulted in variable plant kill rates. This was due to the inability of herbicides to translocate through orange hawkweed stolons, limited control options in sensitive semi-aquatic areas, and the species’ ability to remain undetected in environments with dense tussock vegetation (Caldwell and Wright 2014). Herbicides trials in a range of vegetation and semi-aquatic communities have resulted in picloram-based Tordon® Granules being used as a follow up control to ensure sustained herbicide activity; the clopyralid-based and broad-leaf selective Lontrel® being used in semi-aquatic areas; and greater spray buffers being applied to hawkweed plants to ensure herbicide application to all rosettes and stolons. National Parks and Wildlife Service research into alternative herbicides and control methods has improved control success (Caldwell and Wright 2014), thus improving the chances of eradication.
In December 2014, a small infestation of mouse-ear hawkweed was discovered by a bushwalker near Blue Lake, on the Main Range (2033 m elevation) in Kosciuszko National Park. Like the Victorian mouse-ear hawkweed population, the infestation is thought to have been inadvertently introduced by bushwalkers on clothing, boots or camping gear, after visiting New Zealand or Victoria. Knowing the devastation this weed has caused in New Zealand, within six days of identification verification the National Parks and Wildlife Service had controlled the infestation with Grazon Extra® foliar spray, surveyed the surrounding area, quarantined the area to prevent potential vertebrate dispersal, and established monitoring to determine the species’ response to herbicides. Preliminary results suggest that seven weeks after treatment 60% of monitoring plots recorded no mouse-ear hawkweed and the remainder experienced cover reductions but live plants persisted. Monitoring will continue to ascertain the species’ full response to herbicide.
A detection and control strategy has been put in place by the National Parks and Wildlife Service. The population is thought to have been present for five or more years, and plants had seeded before their discovery (M. Hamilton personal observations), meaning secondary dispersal to other areas may have occurred. Using tools developed for orange hawkweed, a search of other high risk areas was undertaken in February 2015, with over 112 ha searched. To date, the original infestation remains the only known location. It covered 0.015 ha, which although controlled, will require follow-up treatment as seeds may germinate from a seed bank. Again, seeds are viable in the soil for up to five years, although most seeds are only viable for two years, with 2.3% viable after this time (Roberts 1986). Continued surveillance of high priority areas will be needed to ensure plants do not establish elsewhere.
Improved surveillance method
As described above, the suite of herbicides and control methods now available are considered sufficient to extirpate all known orange hawkweed infestations. However, due to the size and remote and rugged nature of the search area, greater efforts are required to delimit the infestation. Until recently, the eradication program relied on humans undertaking ground surveillance, which is resource intensive over large areas. Cindy Hauser (unpublished data) determined that humans searching for non-flowering orange hawkweed in heath vegetation achieved very low detection certainties; in addition, unassisted ground searching becomes less effective when the weed population becomes sparse (Chandler 2014). More efficient surveillance techniques involving unmanned aerial vehicles (UAVs) and specially trained dogs are now being trialled for hawkweed. The combination of these novel surveillance techniques with aerial insertion of search crews should enable larger areas to be surveyed at lower cost and with greater accuracy.
Unmanned aerial vehicles
Trials of unmanned aerial vehicles or drones to detect orange hawkweed are currently underway in New South Wales (Hung and Sukkarieh 2015). The process involves the drone collecting aerial images and the development of a machine learning algorithm, to analyse the image data and identify potential hawkweed infestations. Results suggest hawkweed flowers (with a diameter of around 15 mm) can be detected reliably at 15 m elevation using a Falcon 8 UAV with a Sony Nex 7 camera (Hung and Sukkarieh 2015). Whilst promising and providing proof of concept, a higher flying UAV with higher resolution sensing and a faster frame rate may achieve detection from higher elevations, therefore enabling greater coverage per flight. Further trials to determine the feasibility of orange hawkweed detection using other plant characteristics may also improve detection.
Dogs have long been used in the detection of illicit substances by police and border security forces around the world but their olfactory abilities are increasingly being used for conservation work. Goodwin et al. (2010) showed dogs outperformed people in detection of rare weeds, both more accurately and more efficiently over large areas. In 2014, the Victorian Government began trialling the use of a hawkweed detector dog. Initial findings suggest a dog can be trained to detect orange, mouse-ear and king devil hawkweed rosettes, stolons and rhizomes, with no false positives or false negatives (Hannigan and Smith 2014). Using the same professional dog trainer, the National Parks and Wildlife Service has begun training two dedicated detection dogs for orange and mouse-ear hawkweed. The dogs are being trained to detect hawkweed in field situations, and to distinguish hawkweed from other plant species. Preliminary results indicate that the dogs successfully detect orange hawkweed in both controlled environments and in situ in Kosciuszko National Park. It is envisaged that detection dogs will more efficiently and reliably find hawkweed to accelerate eradication of hawkweeds from New South Wales.
Aerial crew insertions
The final surveillance method the National Parks and Wildlife Service is employing is the use of aerial insertion of crews (volunteers, National Parks and Wildlife Service staff and contractors) in remote areas. Due to the constraints posed by the remoteness of the area, the rugged terrain and, in places, dense vegetation, helicopter insertion of surveillance crews has increasingly been employed to great effect. Helicopter insertion surveys have resulted in the discovery of a disjunct orange hawkweed location close to the Tumut River, and the survey of 70 ha for mouse-ear hawkweed.
Depending on the results of further UAV and detector dog trials, it is envisaged UAVs may assist detection across large areas whilst dogs may be used to improve localised detection and to verify hawkweed absence in an area.
The ‘case’ for eradication
Much research into the variables that affect eradication success has occurred since the seminal study of Rejmánek and Pitcairn (2002). However, many subsequent studies (Myers et al. 2000, Panetta and Timmins 2004, Pluess et al. 2012, Panetta 2014) agree that a combination of the following factors are important to eradication success: i) the ability to prevent re-invasion; ii) biological factors such as time to maturation and seed longevity of the target weed; iii) total gross area of the infestation; iv) number of infestations; v) availability of effective controls; vi) target must be detectable at relatively low densities; and vii) socioeconomic factors that include clear lines of authority, commitment and sufficient program resourcing.
If resourced adequately, orange and mouse-ear hawkweed are at a stage where eradication is feasible. We describe below how most factors above have been met for orange and mouse-ear hawkweeds in New South Wales.
Ability to prevent reinfestation
The chance of reinvasion of hawkweeds to New South Wales is considered low: their deliberate or accidental importation into Australia is now prohibited, the sale of hawkweeds is prohibited in New South Wales, and there is little evidence that hawkweeds are grown ornamentally. The Victorian Government has, and continues to invest heavily in their eradication from Victoria and hygiene protocols are in place in New South Wales to prevent accidental hawkweed spread during management.
Rinella and Sheley (2002) state that orange hawkweed can mature from seed in approximately 5 months. Though a longer time to reproduction period would be desirable, the reduced growing season in the Australian Alps means that at most only one full life cycle (from germinating seed to a fruiting plant) could be achieved per year. However, time to maturation through rosettes produced from stolons is much shorter than 5 months (G. Wright unpublished data). To prevent reproduction in the National Parks and Wildlife Service hawkweed eradication program, sites are revisited 3–4 times per season.
There is considerable uncertainty regarding the long-term viability of soil-stored seeds of Hieracium species. Hawkweed seed longevity has been quoted as being as high as seven years for H. pratense (Williams and Holland 2007), although Bear (2009) points out that this seems to be an erroneous citing of Panebianco and Willemson (1976), who state that ‘hawkweed seeds are probably not viable in soil for a long period of time’. Bear et al. (2012) found little evidence to suggest a long-lived soil seedbank, which is supported by Rowland (2012), who inferred that orange hawkweed soil seedbanks in Kosciuszko National Park were likely to be transient or short-term. However, unpublished National Parks and Wildlife Service research suggests that some orange hawkweed seeds could be viable for up to five years, though this is based on field-based experiments (where windblown seed contamination could not be ruled out) and not laboratory-controlled conditions. As stated previously, mouse-ear hawkweed seeds are viable in the soil for up to five years, although most seeds are only viable for two years, with 2.3% viable after this time (Roberts 1986).
The total gross area of infestation or the area requiring searching for the target species amounts to 45.3 ha for orange hawkweed, being the search area (a 20 m radius area) around each infestation found since 2003. However, despite the orange hawkweed extent appearing to be delimited (Figure 3), there is the possibility the total gross area of infestation is greater. To overcome this issue, targeted surveillance for orange hawkweed has occurred across 1 197 ha since 2010. Despite the large area surveyed, there are still significant areas yet to search. The likelihood of hawkweed presence in these areas should be lower as the highest priority areas have already been searched. Additional outlier priority surveillance areas will be targeted in the coming seasons, and the ability to search a greater area more efficiently should become available with the assistance of UAVs and dogs.
Orange hawkweed infestations occur in nine distinct locations within the 8 951 ha extent. Infestations appear strongly associated with areas that have experienced prior disturbance (e.g. Snowy Hydro Electric Scheme activity) and areas immediately to the east of infestations, suggesting dispersal on the prevailing westerly winds. Though present across a large extent, known orange hawkweed infestations mostly exhibit clumping centred on the nine distinct locations.
Effective control and surveillance
As discussed above, there is now a suite of herbicides and control methods available that are ensuring much greater plant kill rates. In terms of hawkweed detectability at low densities, much of the surveillance effort occurs in the flowering season, when hawkweeds are detectable with higher confidence. To increase detection confidence, self-auditing of surveillance efforts occurs and volunteers and staff are trained to detect hawkweeds when only rosettes are present, resulting in many infestations being found during the non-flowering stage. This does not discount that humans conducting ground surveillance can produce false negatives, as found by Cindy Hauser (unpublished data). National Parks and Wildlife Service drone and detector dog trials suggest greater detection certainty, and that using both may assist in evaluation of the detection certainty of human-conducted ground surveillance.
For both orange and mouse-ear hawkweed, the lines of authority for management are clear, with infestations only being found on National Parks and Wildlife Service estate. Strong partnerships are also in place with surrounding land managers that include local weeds officers, natural resource management staff, the community and park user groups. Recently there has been a significant amount of publicity concerning hawkweeds, resulting in increased community awareness, the National Parks and Wildlife Service program and general hygiene practices. Orange and mouse-ear hawkweed eradication are the most critical priority weed eradication programs for the National Parks and Wildlife Service. The National Parks and Wildlife Service has shown commitment to drive the New South Wales eradication program with funding assistance from the New South Wales Department of Primary Industries (Weeds Action Program) and the Australian Government. However, therein lies the key to eradication success, program resourcing.
With the efforts of the National Parks and Wildlife Service, Victorian Government agencies, volunteers and other partners, there is an opportunity to eradicate hawkweeds from mainland Australia. Hawkweed eradication from New South Wales can only be achieved when every last individual plant is located, controlled and the weed seed bank exhausted. The total area of extant orange and mouse-ear hawkweed in the 2014/15 season was 0.022 ha and 0.015 ha respectively. The current area of extant orange hawkweed demonstrates a decline in area of 98.5% since 2010/11. The development of innovative new tools that may enable more exacting delimitation will become operational in 2015/16. For the reasons outlined above, the probability of eradication is considered relatively high, so, could this be ‘the first’?
The authors would like to thank Jo Caldwell and Greg Cullen for their valuable comments on the manuscript and for their hard work on the orange and mouse-ear hawkweed eradication program. Many thanks also to the myriad volunteers that have contributed their time to the ‘Hunting Hawkweed’ program and hence contributed to the progress described in this paper. The work described above was partially funded by the New South Wales Department of Primary Industries (Weeds Action Program), the Australian Government, and UAV trials are funded through the Northern Tablelands Local Land Services.
Australian Government (2015). National Environmental Alert List. http://www.environment.gov.au/biodiversity/invasive/weeds/weeds/lists/alert.html (accessed 26 August 2015).
Bear, J. (2009). The seed ecology of invasive Hieracium species, Hons thesis. University of Melbourne.
Bear, J.L., Giljohann, K.M., Cousens, R.D. and Williams, N.S.G. (2012). The seed ecology of two invasive Hieracium (Asteraceae) species. Australian Journal of Botany 60, 615-24.
Bomford, M. and O’Brien, P. (1995). Eradication or control for vertebrate pests. Wildlife Society Bulletin 23, 249-55.
Brinkley, T.R. and Bomford, M. (2002). Agricultural sleeper weeds in Australia: What is the potential threat? (Bureau of Rural Sciences, Canberra). 33 pp.
Caldwell, J. and Wright, G. (2012). Orange hawkweed control program report 2011/12. (Office of Environment and Heritage, New South Wales National Parks and Wildlife Service, Hurstville).
Caldwell, J. and Wright, G. (2014). Orange hawkweed control program report 2013/14. (Office of Environment and Heritage, New South Wales National Parks and Wildlife Service, Hurstville).
Chandler, B. (2014). Innovation vs. invasion: innovative techniques to assist the eradication of weed populations. Proceedings of the 19th Australasian Weeds Conference, ed. M. Baker, pp. 259-62. (Tasmanian Weed Society, Hobart).
Cousens, R., Heydel, F., Giljohann, K., Tackenberg, O., Mesgaran, M. and Williams, N. (2012). Predicting the dispersal of hawkweeds (Hieracium aurantiacum and H. praealtum) in the Australian Alps. Proceedings of the 18th Australasian Weeds Conference, ed. V. Eldershaw, pp. 5-8. (Weed Society of Victoria, Melbourne).
Cunningham, D.C., Woldendorp, G., Burgess, M. and Barry, S.C. (2003). Prioritising sleeper weeds for eradication: Selection of species based on potential impacts on agriculture and feasibility of eradication. (Bureau of Rural Sciences, Canberra). 24 pp.
Curran, I. and Primrose, K. (2012). Falls Creek hawkweeds eradication project annual report 2011/2012. Unpublished report. (Alpine National Park, Parks Victoria, Melbourne).
Espie, P. (2001). ‘Hieracium in New Zealand: ecology and management’. (AgResearch Ltd, Mosgiel, New Zealand). 66 pp.
Goodwin, K.M., Engle, R.E. and Weaver, D.K. (2010). Trained dogs outperform human surveyors in the detection of rare spotted knapweed (Centaurea stoebe). Invasive Plant Science and Management 3, 113-21.
Grice, A.C., Clarkson, J.D., Murphy, H., Fletcher, C.S. and Westcott, D.A. (2013). Containment as a strategic option for managing plant invasion. Plant Protection Quarterly 28, 62-5.
Hanigan, L. and Smith, N. (2014). Determining the feasibility or training a dog to detect Hieracium species. Proceedings of the 19th Australasian Weeds Conference, ed. M. Baker, pp. 374-6. (Tasmanian Weed Society, Hobart).
Hung, C. and Sukkarieh, S. (2015). Using robotic aircraft and intelligent surveillance system for orange hawkweed detection. Plant Protection Quarterly 30, 100-2.
Mohr, C.O. (1947). Table of equivalent populations of North American small mammals. American Midland Naturalist 37, 223-49.
Morgan, J.W. (2000). Orange hawkweed Hieracium aurantiacum L.: a new naturalised species in alpine Australia. The Victorian Naturalist 117, 50-1.
Myers, J.H., Simberloff, D., Kuris, A.M. and Carey, J.R. (2000). Eradication revisited – dealing with exotics. Trends in Ecology and Evolution 15, 316-20.
NRC, Natural Resources Commission (2014). Weeds – Time to get serious. Review of weed management in NSW. Final report and recommendations. Natural Resources Commission, Sydney. http://www.nrc.nsw.gov.au/weed-management (accessed 27 July 2015).
Panebianco, R. and Willemsen, R.W. (1976). Seed germination in Hieracium pratense, a successional perennial. Botanical Gazette 137, 255-61.
Panetta, F.D. (2009). Weed eradication – an economic perspective. Invasive Plant Science and Management 2, 360-8.
Panetta, F.D. (2014). Weed eradication feasibility: lessons of the 21st century. Weed Research 55, 226-38.
Panetta, F.D. and Brooks, S.J. (2008). Evaluating progress in weed eradication programs. Proceedings of the 16th Australian Weeds Conference, eds R.D. van Klinken, V.A. Osten, F.D. Panetta and J.C. Scanlan, pp. 418-20. (Weed Society of Queensland, Cairns).
Panetta, F.D, Csurhes, S., Markula, A. and Hannan-Jones, M. (2011). Predicting the cost of eradication for 41 Class 1 declared weeds in Queensland. Plant Protection Quarterly 26, 42-6.
Panetta, F.D. and Timmins, S.M. (2004). Evaluating the feasibility of eradication for terrestrial weed invasions. Plant Protection Quarterly 19, 5-11.
Pluess, T., Jarošík, V., Pyšek, P., Cannon, R., Breukers, A. and Bacher, S. (2012). Which factors affect the success or failure of eradication campaigns against alien species? PLoS ONE 7, e48157.
Rejmánek, M. and Pitcairn, M.J. (2002). When is eradication of exotic pest plants a realistic goal? In ‘Turning the tide: The eradication of island invasives’. eds C.R. Vietch and M.N. Clout, pp. 249-53. (International Union for Conservation of Nature, Gland, Switzerland).
Rinella, M. and Sheley, R. (2002). Orange and meadow hawkweed. (Montana State University Extension Service, Bozeman, Montana, United States of America).
Roberts, H.A. (1986). Seed persistence in soil and seasonal emergence in plant species from different habitats. Journal of Applied Ecology 23, 639-56.
Rowland, D. (2012). Mechanisms of invasion of Hieracium aurantiacum and Leucanthemum vulgare in Kosciuszko National Park, Hons thesis. University of Wollongong, Wollongong.
Simberloff, D. (2003). Eradication – preventing invasions at the outset. Weed Science 51, 247-53.
Stergios, B. (1976). Achene production, dispersal, seed germination, and seedling establishment of Hieracium aurantiacum in an abandoned field community. Canadian Journal of Botany 54, 1189-97.
Williams, N.S.G. and Holland, K.D. (2007). The ecology and invasion history of hawkweeds (Hieracium species) in Australia. Plant Protection Quarterly 22, 76-80.
Williams, N.S.G., Hahs, A.K. and Morgan, J.W. (2008). A dispersal-constrained habitat suitability model for predicting invasion of alpine vegetation. Ecological Applications 18, 347–59.
Worboys, G.L. and Good, R.B. (2011). Caring for our Australian Alps catchments: Summary report for policy makers. (Department of Climate Change and Energy Efficiency, Canberra).