- Email: [email protected]
Australian plant biosecurity surveillance systems
Crop Protection 100 (2017) 8e20
Contents lists available at ScienceDirect
Crop Protection journal homepage: www.elsevier.com/locate/cropro
Australian plant biosecurity surveillance systems Chris Anderson a, Samantha Low-Choy b, c, Peter Whittle c, Sharyn Taylor b, € cker a, Richard Davis e, Cherie Gambley d, Linda Smith d, Peter Gillespie a, Holger Lo Bernie Dominiak a, * a
Biosecurity and Food Safety, NSW Department of Primary Industries, Locked Bag 21, Orange, New South Wales, 2800 Australia Plant Biosecurity CRC, University of Canberra, Bruce, Australian Capital Territory, 2617 Australia c Mathematical Sciences, Queensland University of Technology, GPO Box 2434, Brisbane, Queensland, 4001 Australia d Queensland Department of Agriculture and Fisheries, Applethorpe Research Station, GPO Box 501, Stanthorpe Queensland 4378 Australia e Northern Australia Quarantine Strategy, Department of Agriculture and Water Resources, PO Box 96, Cairns International Airport, Queensland 4870 Australia b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 December 2016 Received in revised form 24 May 2017 Accepted 24 May 2017
Australia is an island nation and a primary producer of agricultural and horticultural products. There is a large diversity of plant biosecurity threats which could adversely impact on Australia's production and exports. Surveillance has traditionally been used to monitor pests and optimise production. Increasingly surveillance is being used for early detection of exotic incursions, demonstration of eradication of incursions and pest freedom from exotic or endemic pests. These newer uses of surveillance utilise general and specific surveillance: surveillance data is maintained in electronic databases. Specific surveillance is a targeted surveillance search used by industry or state regulators for a specific pest to support pest freedom or other trade standards. The plant biosecurity surveillance cycle shows the flow of surveillance operations. In this paper, this cycle is demonstrated by case studies including pre border and the northern Australian at-border surveillance for the Australian-Asian interface. Within Australia, the multiple plant pest surveillance program was established in most capital cities where there are high flows of population and produce. As an industry example, the cotton industry surveillance program, particularly for cotton leaf curl, demonstrates how plant biosecurity surveillance operates within an industry. Asiatic citrus canker is another example of industry pertinent surveillance. Finally, surveillance for the purpose of declaring pest freedom areas is reviewed using fruit flies and currant lettuce aphid as examples. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Market access Freedom Detection Monitoring
Contents 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
Introduction - a diversity of threats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Why do we do surveillance? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pest records to support pest status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . How do we do surveillance? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . General surveillance? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Specific surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The plant biosecurity surveillance cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case studies: successful surveillance systems in action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . Case study. Pre border - surveillance North of Australia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case study. Border e Northern Australia quarantine strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case study. Post border e the multiple plant pest surveillance program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case study. Post border e Australian cotton disease surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
* Corresponding author. E-mail address: [email protected] (B. Dominiak). http://dx.doi.org/10.1016/j.cropro.2017.05.023 0261-2194/© 2017 Elsevier Ltd. All rights reserved.
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. 9 . 9 . 9 11 11 12 12 13 14 14 15 15
C. Anderson et al. / Crop Protection 100 (2017) 8e20
13. 14. 15. 16.
Case study. Post border e early detection of cotton leaf curl disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case study. Post border e emergency surveillance for Asiatic citrus canker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case study. Post border e maintaining pest free areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Concluding comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction - a diversity of threats Invasive species cost an estimated US$314 billion annually, with over 120,000 plant, animal and pest species believed to have invaded Australia, Brazil, India, South Africa, the United States of America and the United Kingdom alone (Pimentel et al., 2001). Recognising the need for international standards to facilitate trade of plants and plant produce around the world, while minimising the movement of plant pests, the International Plant Protection Convention (IPPC) develops International Standards for Phytosanitary Measures (ISPMs). These standards include surveillance, defined as ‘an official process that collects and records data on pest occurrence or absence by survey, monitoring or other procedures’ (IPPC, 2010). As such, plant biosecurity surveillance provides a scientific basis for the assessment and management of risks posed to plant production and the environment by plant pests. The term ‘pest’ refers to “… any species, strain or biotype of plant, animal or pathogenic agent, injurious to plants or plant products” (IPPC, 2010). The modern age of globalisation means that the threat from unwanted pests is greater than ever before (Hulme, 2009; Stanaway et al., 2001). Global pathways of entry through tourism, immigration and trade are open and largely unregulated in many areas (Hulme, 2009). Data published on the IPPC website indicates that Australia has reported on the status of 49 new plant pests comprised of 17 fungi, 12 invertebrates, eight viruses and virus like organisms, five bacteria and seven weeds (Table 1 covers the period from 13 November 2006 to 21 January 2014). These likely conservative numbers demonstrate the diversity of biosecurity threats confronting managers, scientists and persons designing and implementing surveillance systems. Of these pests, five have been eradicated, and a further 10 are under eradication or official control. The remaining detections were considered not feasible or cost-beneficial to eradicate, reflecting the fact that they were of minor concern or, more significantly, that detection did not occur until post-border establishment and spread. Moreover, many of these plant pests are cryptic, being vectored within or on other organisms and having symptoms that are hard to distinguish from endemic and established pests or abiotic constraints. Because of this they are difficult to detect through traditional surveillance activities. 2. Why do we do surveillance? Plant biosecurity is a suite of risk management activities deployed across a continuum of jurisdictions, from pre-border, at border, to post-border activities that minimise the impact of plant pests on natural and managed ecosystems (Magarey et al., 2009; Nairn et al., 1996). Surveillance is a fundamental component of any plant biosecurity system, as knowledge of pest status is the basis for managing risk. Plant biosecurity surveillance systems are designed and implemented to assess pest status for many different reasons (McMaugh, 2005). These fall broadly into three categories:
9
16 17 17 18 19 19
1. surveillance to facilitate trade (e.g. area freedom, areas of low pest prevalence, pest free places of production); 2. surveillance to support pest management and control (e.g. commodity specific pest surveillance and monitoring of endemic pests); and 3. surveillance to facilitate early detection, and support and validate responses to pest incursions. Surveillance systems often serve more than one of these purposes, which are intrinsically linked. For example in south eastern Australia, routine surveillance in the Fruit Fly Exclusion Zone (FFEZ) (Anon, 1996) was carried out systematically to validate area freedom (Category 1), and support pest management and control (Category 2) by ensuring early detection (Category 3) so that measures could be rapidly implemented to prevent establishment of Queensland fruit fly (QFly) (Bactrocera tryoni (Froggatt)) and provide early detection of Mediterranean fruit fly (Medfly) (Ceratitis capitata (Weidemann)) (Dominiak and Daniels, 2012). Such measures thereby facilitated market access for fruit grown within the FFEZ on the basis of area freedom. 3. Pest records to support pest status Surveillance is a tool used to generate pest records, which in turn support determinations of pest status, through sufficient evidence in the form of accurate and reliable pest records (IPPC, 1998; van Halteren, 2000). The accuracy and reliability of a pest record is directly impacted by the design and implementation of the surveillance system through which it was generated. A pest record ‘provides information concerning the presence or absence of a pest, the time and location of the observations, host(s) where appropriate, the damage observed, as well as references or other relevant information pertaining to a single observation’ (IPPC, 1998). Substantial effort is directed towards ensuring individual reliability of pest records. The reliability of pest records ‘is based on consideration of the data in regard to the collector/identifier, the means of technical identification, the location and date of the record, and the recording/publication of the record’ (IPPC, 1998). Further, the reliability of pest records will be improved where nationwide or international database and metadata standards are in place to provide consistency and coherency in pest record management. Hill (2013) noted the importance of verifying old records with contemporary surveillance to ensure the validity of historic records in databases. The incorporation of fully documented voucher pest specimens in institutional repositories is imperative to support surveillance activities. Pests may impact several categories including economic, social and environmental categories. Economic pests impact on individual producers, groups of producers and regional/international market access. This category of pest usually drives a funded response or eradication. Australia conducts surveillance for fruit flies to demonstrate freedom from particular species or to detect incursions (Dominiak and Daniels, 2012; Dominiak and Worsley, 2016). Pests such as Asian gypsy moths (Lymantria spp.) cause
10
C. Anderson et al. / Crop Protection 100 (2017) 8e20
Table 1 Plant pest reports in Australia and their status. Information published on IPPC website - correct as of February 2014 (IPPC, 2011). Dates of incursions range from 13 November 2006 to 21 January 2014. Note that Vegetable leaf minor was originally detected in this period and was detected again in 2017 and the table includes this more recent detection for completeness. Common name
Scientific name
Host Commodity
Taxon
Status
Date reported
Alternaria
Alternaria sp.
Fungus
Present: in all parts of the area
16 November 2006
Angelonia flower break virus Canna yellow mottle badnavirus Siam weed Mango Fruit Borer
Angelonia flower break virus
Parsley (Petroselinum crispum) Verbena, Angelonia and Phlox spp. Canna spp.
Virus
Absent: intercepted only
30 April 2009
Virus
Present: at low prevalence
17 May 2007
Weed Insect
Present: but managed Present: only in some areas
11 November 2013 3 December 2009
Weed Mite Insect Fungus Fungus Virus
Present: under eradication Present: only in some areas Absent: pest eradicated Present: under eradication Present: only in some areas Present at low prevalence
25 23 21 21 10 30
Fungus
Present: only in some areas
28 October 2011
Koster's curse New Hebrides coconut mite Cocoa pod borer Chestnut blight Citrus leaf spot Dracaena mottle badnavirus Black scab of jojoba
Canna yellow mottle badnavirus Eupatorium odoratum Citripestis eutraphera
Clidemia hirta Colomerus novahebridensis Conomorpha crameralla Cryphonectria parasitica Cryptosporiopsis citri Dracaena mottle badnavirus
Mango (Mangifera indica), cashew (Anacardium occidentale)
Mango Malformation Disorder Garlic allexivirus Potato Cyst Nematode
Fusarium mangiferae
Coconut (Cocos nucifera) Cocoa (Theobroma cacao) Chestnut (Castanea sativa) Citrus spp. Happy Plant (Dracaena spp.) Jojoba (Simmondsia California) Mistflower (Ageratina riparia) Mango (Mangifera indica)
Garlic allexiviruses Globodera rostochiensis
Garlic (Allium sativum) Potato (Solanum tuberosum)
Virus Nematode
Willow black canker Tobacco Slug Beetle
Physalospora miyabeana Hemileia wrightiae Lema bilineata
Willow (Salix spp.) Wrightia pubescens Prairie Ground Cherry, evaluated as a biocontrol agent for jimsonweed
Fungus Fungus Insect
Yellow Burrhead Vegetable Leaf Miner
Limnocharis flava Liriomyza sativae
Rice blast
Magnaporthe grisea
Miconia vine
Mikania vine Narcissus Late Season Yellows Virus Teak leaf rust
Miconia calvescens, M. nervosa, M. racemosa Mikania micranthia Narcissus Late Season Yellows Virus Olivea tectonae
Panicum mosaic virus
Panicum mosaic virus
Grapevine leaf rust Branched broomrape
Solenopsis Mealybug
Phakopsora euvitis Phelipanche ramose/P. mutelii (formerly Orabanche ramosa/O. mutelii) Phenacoccus solenopsis
Banana freckle
Phyllosticta cavendishii
Phytophthora elongata Impatiens downy mildew
Phytophthora elongata Plasmopara obducens
Mango leaf gall midge Asian strain of bacterial canker of kiwi fruit Spider mite Red imported fire ant
Procontarinia pustulata Pseudomonas syringae pv. actinidiae Schizotetranychus baltazari Solenopsis invicta Striga asiatica
White smut fungus
Elsinoe australis (new pathotype) Entyloma ageratinae
Aphid Tomato Torrado Virus
Takecallis arundinariae Tomato torrado virus
Myrtle rust Sugarcane smut
Uredo rangelii Ustilago scitaminea Valsa malicola
February 2010 April 2009 January 2014 September 2010 October 2012 April 2009
Fungus
Present: only in some areas
25 November 2010
Fungus
Present: under eradication Present: only in some areas Absent from WA: Eradicated; Present Victoria: Under official control Present: only in some areas Present: only in some areas Present: only in some areas
22 February 2008; 14 January 2010 17 January 2012 11 November 2008; 14 January 2010
Weed
Present: Present: control Present: control Present:
Daffodil (Narcissus sp.)
Weed Virus
Present: only in some areas Present: only in some areas
5 March 2010 11 January 2009
Teak (Tectona grandis)
Fungus
31 October 2006
Buffalo grass (Stenotaphrum secundatum cv. Palmetto) Vitis spp.
Virus Fungus Weed
Present: in all parts of the area where host crop is grown Present: only in some areas; Present: at low prevalence Absent: Eradicated Present: under official control
11 March 2008 25 February 2010
Insect
Present: only in some areas
18 March 2010
Fungus
4 October 2013
Oomycete Oomycete
Present: only in some areas; under eradication Present: only in some areas Present: only in some areas
25 November 2010 17 May 2007
Insect Bacteria
Present: only in some areas Present: only in some areas
11 November 2008 12 July 2011
Insect Insect Weed
Present: Present: Present: control Present: Present:
only in some areas under eradication subject to official
26 July 2012 24 February 2010 22 July 2013
only in some areas only in some areas
23 April 2009 10 October 2008
Solanaceae, Cucurbitaceae, Fabaceae, Asteraceae Rice (Oryza sativa)
Cotton (Gossypium hirsutum) Banana (Musa sp.) Cavendish sub group Eucalyptus spp. Impatiens spp. Mango (Mangifera indica) Kiwi fruit (Actinida chinensis) Citrus spp. Sugarcane (Saccharum officinarum) Bambusa spp. Tomato (Solanum lycopersicum) Species in Myrtaceae Sugarcane (Saccharum officinarum) Apple (Malus domestica)
Aquatic weed Insect
16 November 2006 25 November 2010 23 April 2009
Fungus
Insect Virus Fungus Fungus Fungus
under eradication subject to official
5 March 2010 24 April 2017
subject to official
22 August 2011
only in some areas
5 March 2010
Present: only in some areas Present: established in all parts where host is grown Unknown
6 July 2010
5 May 2010 13 Nov 2006 1 February 2011
C. Anderson et al. / Crop Protection 100 (2017) 8e20
11
Table 1 (continued ) Common name
Scientific name
Invasive ant High Plains Virus Citrus canker
Wasmannia auropunctata Wheat high plains virus Xanthomonas citri subsp. citri Xanthomonas fragariae
Angular leaf spot Mahogany Angular Leaf Spot Yucca bacilliform badnavirus
Xanthomonas sp. Yucca bacilliform badnavirus
Host Commodity
Taxon
Status
Date reported
Wheat (Triticum aestivum) Citrus spp.
Insect Fungus Bacteria
Present: under eradication Present: only in some areas Absent: eradicated
11 January 2012 21 December 2012 23 January 2009
Bacteria
Absent: eradicated
30 November 2012
Bacteria
Present: only in some areas
5 February 2009
Virus
Present: at low prevalence
17 May 2007
Strawberry (Fragaria x ananassa) African mahogany tree (Khaya senegalensis) Yucca elephantipes
health concerns to residents (Dominiak et al., 2013). Red imported fire ants (Solenopsis invicta Burren) would adversely impact Australia's outdoor lifestyle and surveillance is conducted for early detection (Dominiak et al., 2010). Red imported fire ants and yellow crazy ants (Anoplolepis gracilipes (Smith)) impact the environment removing a range of biota: the latter was detected and eradicated in northern New South Wales (Dominiak et al., 2010, 2011a). Recent incursions of amenity tree pests could reduce tree canopies and this is undesirable in Australia where skin cancer and sun exposure are health concerns (Dominiak et al., 2008; Lefoe et al., 2014). 4. How do we do surveillance? Data to support claims of pest status e.g. that a pest is “known not to occur” can be gathered through general surveillance (also called passive surveillance) and specific surveys (also called active or targeted surveillance) (IPPC, 1998; van Halteren, 2000). General surveillance is defined as a process whereby information on particular pests that are of concern for an area is gathered from many sources (IPPC, 1997). Specific surveys are defined as procedures by which National Plant Protection Organisations (NPPOs) obtain information on pests of concern on specific sites in an area over a defined period of time (IPPC, 1997). Specific surveys can be expensive due to the costs of sending specially trained teams into the field, but can potentially provide more reliable and comprehensive pest records and therefore more precise quantitative estimates of confidence in pest status compared to general surveillance. However, general surveillance can be a very powerful tool for the reporting of new pest incursions, especially where budgetary limitations restrict the level of specific surveys conducted. For example, the majority of new pest reports in New Zealand occur through general surveillance rather than through the deployment of specific surveys (Wilson et al., 2004). In most cases, the capacity of a surveillance system to detect pests will improve when both strategies are deployed at all stages of the plant biosecurity surveillance cycle. Surveillance is often done to a standard. Fruit flies are a major economic pest in Australia and the world. Fruit flies are amongst the easiest of insects to monitor due to their attraction to lures (Suckling et al., 2016). Queensland fruit fly is not detected in Western Australia but that state maintains trapping to demonstrate continuing absence (Dominiak and Daniels, 2012). Similarly decades of fruit fly monitoring have demonstrated that Bactrocera neohumeralis (Hardy) is not detected south of Sydney (Dominiak and Worsley, 2016). On the margins of insect range, populations decline until permanent populations are difficult to maintain. Standards such as areas of low pest prevalence can be clearly declined for species such as Queensland fruit fly (Dominiak et al., 2015). These standards may be extrapolated to other species such as B. neohumeralis (Dominiak and Worsley, 2016). In areas where populations are eradicated, the pest status can be demonstrated by
regular surveillance (Dominiak and Daniels, 2012) or by statistical analysis of the likelihood of eradication (Dominiak et al., 2011a). 5. General surveillance? General surveillance includes: (1) data mining and analysis of information about specific pests from multiple sources; and (2) engagement with stakeholders to increase awareness of target pests and thereby encourage widespread reporting through official channels. Sources of information for data mining include NPPOs, research institutions, scientific publications, consultants, and industry bodies. Data collected may include the results of specific surveys, as well as the records of diagnostic laboratories, scientific collections, and the results of pest research. Examples of Australian general surveillance tools that use these sources of information together include: the National Plant Surveillance Reporting Tool (NPSRT) which collates records of government and industry funded surveys across a range of commodities (Plant Health Australia, 2011a); the Australian Plant Pest Database (APPD) which provides online access to specimen records in 14 scientific collections (Plant Health Australia, 2011b); Australia's Virtual Herbarium which provides a central point for access to plant records in each Australian state (AVH, 2011); and the Biosecurity Surveillance Incident Response and Tracing (BioSIRT) tool which is a national system for collating and analysing results of specific surveys completed in response to plant pest incursions. It has been used by some states but not all. Some states already maintain their own electronic databases that can quickly be mined for trade purposes, particularly to demonstrate pest freedom (Dominiak et al., 2007). Other states keep paper records or multiple spreadsheets and these are difficult to quickly data mine. Another form of data mining and analysis that is not commonly used in the plant world is syndromic surveillance through which real-time data sets like web searches and sales figures of chemicals may be interrogated. These data could reveal patterns to provide early warning of pest and disease outbreaks. Such a technique may be useful in the plant world to predict for example patterns of fruit fly distribution in fruit fly free regions or states (Dominiak and Daniels, 2012). An important aspect of general surveillance is the fostering of “stakeholder” or “grass-roots” surveillance, through education and awareness campaigns encouraging the public to report potential pest detections. A good example of this is the Northern Australian
12
C. Anderson et al. / Crop Protection 100 (2017) 8e20
Quarantine Strategy (NAQS) that specifically targets local communities to foster awareness and reporting of biosecurity threats. The use of “engaged” stakeholders was captured in a recent surveillance design targeting exemplar species entering Barrow Island in north western Western Australia (Jarrad et al., 2010). The design (Barrett et al., 2009) accommodated both general surveillance corresponding to potential detection by workers, and targeted surveillance via a variety of methods including trapping and observations in situ. Workers on the island were divided into two categories: engaged workers that were considered “active” since they were sufficiently motivated to attend training sessions, contrasting with other workers considered “passive”. Another recent case where “engaged” stakeholders were utilised to support surveillance activities occurred during the incursion of myrtle rust (Puccinia psidii Winter) in NSW in 2010 (Carnegie and Cooper, 2011). Stakeholders including the nursery and garden industry, state and local government, the Australian tea tree industry and bush regenerators were engaged through a series of formal and informal meetings and training sessions, media releases, weekly email updates, the provision of extension material including leaflets and posters and a website that was regularly updated with information on the operation. Emails, media releases, extension material and the website were all designed to increase awareness of the disease and how to report it. Reporting mechanisms included a biosecurity email address and the national Exotic Plant Pest (EPP) hotline. These channels resulted in several reports of myrtle rust being received through the email address and hotline on a weekly basis and assisted in the targeting of delimiting surveys: the results of which were logged in BioSIRT. In response to the continued public interest, an online reporting tool is under development by the Atlas of Living Australia (ALA) which will enable the public to log reports of any pest on an interactive map (ALA, 2011). This system will also record the general level of awareness of the reporter so as to attribute confidence levels to each pest report. The data generated through the ALA mapping project will be available to the public, and will be utilised by a range of stakeholders, from government to private individuals, to manage any incursion into the future. In addition to specialised surveillance scenarios like the myrtle rust response and the Multiple Plant Pest Surveillance Program, several national plant biosecurity awareness programs in Australia promote public reporting through the EPP hotline. This is a national free-call telephone number that connects the public directly to relevant state plant protection organisations. Both Plant Health Australia's “Spotted anything unusual” campaign, and the Quarantine Domestic program (Quarantine Domestic, 2011) promote the EPP Hotline as the primary point of contact. This simplifies the reporting process for the public. The hotline creates a single national tool through which general surveillance data on plant pests is captured. A similar national hotline operates in New Zealand and also takes reports of environmental and animal pests and diseases (Froud and Bullians, 2010). Hotlines may reduce the face-to-face contact between government and stakeholders traditionally done by face-to-face contacts or farm visits (Kruger et al., 2009). Hotlines should always form part of an integrated engagement program which includes face to face options for increasing stakeholder awareness since the accuracy and reliability of reports to hotlines increases with increased awareness among callers. 6. Specific surveys Specific surveys, sometimes referred to as active or targeted surveillance, are official procedures by which NPPOs obtain information on pests of concern on specific sites in an area over a defined period of time (IPPC, 1998). Specific surveys include
surveillance for target pests as well as commodity or host surveillance (IPPC, 1998). There are three types of specific surveys: detection, delimiting and monitoring (IPPC, 1998). These form the backbone of the plant biosecurity surveillance cycle, are complimentary, and usually interdependent. One form of surveillance may lead to another. When deployed in conjunction with general surveillance, they improve the precision of the surveillance system. A detection survey is a survey conducted in an area to determine if pests are present (IPPC, 1998) and is implemented: (1) when pest status needs to be determined or verified for trade purposes e.g. establishing or maintaining a pest free area, area of low pest prevalence, pest free place of production or pest free production site; (2) for early detection of emergency plant pest incursions; and (3) for compilation of pest and host lists (Evans and McMaugh, 2005). Detection surveys and general surveillance are often deployed simultaneously to improve the probability of pest detection and both may lead to delimiting surveys. Delimiting surveys are conducted in the initial stages of an emergency plant pest incursion to establish the boundaries of the infested and free areas (IPPC, 1998) and to inform decisions about the feasibility of eradication. Monitoring surveys are ongoing surveys that go beyond simply recording presence or absence. Surveys may verify the characteristics of a pest population and to provide ongoing validation of pest status in an area, for example during and following completion of an eradication program, or to establish an area of low pest prevalence for trade purposes (IPPC, 1998; Evans and McMaugh, 2005). There are many methodologies and designs employed in the development of specific surveillance strategies that will not be dealt with here (Low-Choy et al., 2011). More generally, specific surveys rely on careful stratification of an area according to the associated risk of pest entry, establishment and spread, and the consequences of incursion; a process collectively referred to as pest risk analysis (PRA) (Anon, 2007). PRA considers analysis of trade patterns along with pest biology and epidemiology (IPPC, 2004). For example, the Multiple Plant Pest Surveillance Program targets early detection surveys for exotic pests in areas that have been assessed through this process. Following the introduction of the Emergency Plant Pest Response Deed (EPPRD) in Australia in 2005, Plant Health Australia has worked with industries to develop high priority plant pest threat lists through the PRA process, which assist targeting of surveillance and preparedness activities (Plant Health Australia, 2010). Case studies provide specific examples of surveillance and preparedness activities currently in place in Australia as a result of the industry biosecurity planning process. PRA is also the basis for commodity import risk analysis (IRA) that is used to set policy and regulate international and domestic trade (Australian Government Department of Agriculture, Fisheries and Forestry, 2009). 7. The plant biosecurity surveillance cycle The IPPC (1998) describes pest status in an area as unknown, present, absent or transient. Each of these states has several variations dependent upon the level of pest management in place and the availability of reliable records (IPPC, 1998). Pest status in an area could be considered unknown when: (1) there is a lack of accurate and reliable records for the pest e.g. existing records do not have sufficient coverage or are otherwise not sufficiently representative across the area of interest; and (2) where there are reasons to consider that pest status may have changed (IPPC, 1998). Changes in pest status may occur for various reasons including climate
C. Anderson et al. / Crop Protection 100 (2017) 8e20
change, changes in crop management and diversity, new trade patterns that open new pathways for pest entry, or an undetected incursion. When the pest status of an area is unknown, it can be resolved through a cycle of complementary and often interdependent surveillance activities consisting of both general surveillance and specific surveys. We call this process the Plant Biosecurity Surveillance Cycle (Fig. 1). The extent to which phases of the Plant Biosecurity Surveillance Cycle are engaged is determined by the extent to which the target pest has or is likely to have penetrated the biosecurity continuum and the magnitude of risk associated with the target pest. For example, where there is no reason to suspect a pest has ever been detected post-border, a determination to that effect will be based on the results of “detection” activities which may include data mining and the results of specific surveys deployed at ports of entry (Gillespie, 2003). In such a case there would be no need for “delimiting” or “monitoring” (the lower two-thirds of the cycle, Fig. 1). However when a high-risk pest such as Asiatic citrus canker is detected post-border, then all phases of the cycle will apply (Fig. 1). Here we consider the post-border detection of a high risk pest (Fig. 1), which may be detected first through general surveillance, including reports by the public to plant health organisations, or through specific surveys designed to detect the target around high risk points of entry (IPPC, 1998). Often, activities associated with “detecting” may lead directly to a determination of pest absence through repetition of these activities e.g. annual early detection surveys. Alternatively, following detection, the surveillance cycle progresses to the “delimiting” phase (Fig. 1), with specific surveys designed to determine pest prevalence in an area and structured, in part, according to reports gathered through general surveillance. Based on data gathered through delimiting surveys, along with expert advice, managers may determine that it is feasible to proceed with eradication of the pest, or alternatively that eradication is not feasible. Surveillance during eradication often involves a combination of further delimiting and monitoring surveillance to ascertain further
Pest Status: Unknown
DetecƟng
DelimiƟng
N
Feasible to Eradicate?
Y Monitoring
Pest Status: Present Transient Absent Fig. 1. Plant biosecurity surveillance cycle.
13
spread of the pest and monitor eradication/control efforts respectively (Fig. 1). Following an eradication campaign, or if eradication is not feasible, delimiting surveillance will cease and activities will shift wholly to monitoring. The pest status of an area posteradication is likely to be accepted by trading partners only after monitoring for an agreed period. This period of ongoing monitoring is of particular importance when attempting to demonstrate area freedom following an incursion and eradication program (Collins and Vanstone, 2010). If a pest status of ‘absent’ is confirmed by monitoring, the cycle will revert to detection in order to ensure preparedness for future incursions. Where pest status is determined to be ‘present’ or ‘transient’, monitoring may go on indefinitely. Thus, detection, delimiting, and monitoring are complimentary and interdependent components of the plant biosecurity surveillance cycle. When considered together, the components provide robust cumulative data to support determinations of pest status across the biosecurity continuum. 8. Case studies: successful surveillance systems in action The success of a plant biosecurity surveillance system is measured by its capacity to detect pests to a required standard. This standard is often set to reflect a nation's appropriate level of protection (ALOP), but may also be defined by the circumstances at hand, for example a higher standard may be adopted during an eradication program. The efficacy of a surveillance system is directly proportional to its sensitivity, i.e. the probability of detecting the pest, given it is present. System sensitivity is largely influenced by the underlying conceptual and quantitative models (Low-Choy et al., 2011). However, sensitivity is also influenced by three key non-statistical considerations: (1) availability of resources; (2) accurate diagnostics; and (3) levels of stakeholder engagement. Resource availability, including availability of equipment and trained personnel, directly impacts on sensitivity. For example, where a trapping grid is incomplete or inappropriately spaced, or where staff are not competent to identify disease symptoms, the sensitivity of the system will be reduced. This will also be the case where diagnostic labs are ill-equipped, and where diagnosticians do not have access to current techniques or to comprehensive reference collections. Incorrect or inconclusive pest identification can lead to unnecessary expenditure when indigenous species are identified as exotic and vice versa (Holmes et al., 2008). This should be avoided through adequate resourcing of diagnostic laboratories and scientific reference collections, and investment in ongoing training for diagnosticians and taxonomists. The goal of stakeholder engagement should be to foster pest awareness and preparedness so as to enhance general surveillance at all stages of the plant biosecurity surveillance cycle. A two-way flow of information that encourages collaboration, ownership and feedback is always preferable to top-down flow of information without opportunity for interaction (Kruger et al., 2009; Thompson et al., 2009). Engaged stakeholders are active participants in general surveillance and thereby improve overall sensitivity of the system. Feedback mechanisms are also important when it comes to timely reporting of results. Delays in reporting can lead to delays in the implementation of pest management responses and can put at risk the pest status of trading partners by allowing potentially infested produce to move across borders without adequate treatment. The longer a pest remains undetected, the more the chance of successful eradication decreases and the cost to government, industry and the environment increases (Magarey et al., 2009). Thus the design and implementation of successful surveillance systems is of utmost importance in managing the risk of plant pests to the environment and the economy worldwide.
14
C. Anderson et al. / Crop Protection 100 (2017) 8e20
Plant biosecurity surveillance systems in the Australian context are unique as the target areas are vast, the population widely distributed (similar to North America), and the temperatures can range from sub tropical to sub Arctic. Regardless, the underlying reasons for doing surveillance and the principles of successful surveillance design are universal. The following are examples of successful surveillance systems deployed for different purposes at different points along the biosecurity continuum in the Australian context. 9. Case study. Pre border - surveillance North of Australia For over 20 years, plant health scientists from Australia and Papua New Guinea (PNG) have conducted regular joint surveys in parts of PNG close to the Indonesian border. This surveillance aims to provide early warning of new incursions of plant pests of significance to both nations. In recent years, these have been collaborations between PNG's National Agriculture Quarantine and Inspection Authority (NAQIA) and the Australian federal government's Department of Agriculture and Water Resources. Early in this period, a number of similar collaborative surveys were conducted by the same Australian scientists and counterparts from Indonesia's quarantine service in the Indonesian province of Papua, which occupies the eastern half of the island of New Guinea. The surveys involve large teams, comprising counterparts from each country in each of the disciplines of botany, entomology, and plant pathology, and local extension staff. For periods of 2e4 weeks per survey, these teams travel great distances through sometimes highly inhospitable and remote terrain by road, light aircraft and boat. Scientists visit and examine as many different locations and crop plants of economic importance as possible, and occasionally also other plants. Specimens are collected as necessary. Most recently, ‘smart phone’ technology has been employed using GPS photo-tagging software to record field data. Samples and specimens are processed in the field for long term storage, treated to minimise biosecurity risk and returned to Australia under suitable import permit for laboratory analysis. Recent regional movements of certain pathogens of banana and citrus are particularly significant. These commodities are major Australian production industries, each valued at around $500 million in 2010-11 (National Plant Biosecurity Status Report, 2012). They are also crops vital to subsistence production and local commerce in PNG (Gibson, 2000). Moreover, the island of New Guinea is part of the centre of origin of both banana and citrus implying significant threat to germplasm resources (Simmonds, 1962; Swingle, 1967). The surveys of Papua Province in the late 1990s led to the first records on the New Guinea land mass of two key wilt pathogens of banana known to be spreading to the west through Indonesia. These were ‘tropical’ race 4 of Fusarium oxysporum Schlecht. f.sp. cubense (E.F. Smith) Snyder and Hansen (Davis et al., 2000a) and the blood disease bacterium (Ralstonia haywardii subspecies celebensis) (Davis et al., 2000b). These discoveries threaten the banana industries of both PNG and adjacent far north Queensland (Davis et al., 2001). Investigation of any wiltelike problems of banana in PNG therefore became a high priority activity. Surveys were done by destructive internal examination of banana plants showing yellowing and leaf death. Over several years, a widespread and devastating lethal wilt of mostly cooking banana plants was investigated and no symptoms of fusarium or bacterial wilt disease were seen. Instead, a set of characteristic internal symptoms were consistently found that were not those of any known pathogen or disorder. The cause remained unknown for about a decade. Subsequently, a research interest in phytoplasmas associated with monocotyledonous plants developed following an outbreak of a
new lethal wilt disease of coconuts in northern PNG (Kelly et al., 2011). Davis et al. (2012) built up a body of evidence that implicated the phytoplasmas closely related to the new coconut phytoplasma as the causal agent. This would be a unique and completely new wilt disease of banana worldwide if pathogenicity can be confirmed in inoculation tests. Early surveys in Papua Province also focused on one of the worst diseases of citrus trees, huanglongbing disease (HLB), caused by the phloem-limited bacterium, ‘Candidatus Liberibacter asiaticus’. This was because the vector, Asian citrus psyllid Diaphorina citri (Kuwayama, 1908), had been found on the island in 1990 (Aubert, 1990) with later dispersal monitored by the joint surveys. Elsewhere in the world, vector establishment in new locations has preceded disease outbreaks by a few years and HLB was indeed confirmed in Papua, near the PNG border in 1999 (Davis et al., 2000c). Subsequent surveys in PNG led to records of both insect and disease in the border town of Vanimo in 2002 (Weinert et al., 2004). A campaign of quarantine containment followed and delimiting surveys have documented little movement of D. citri and HLB (unpublished data, 2015). 10. Case study. Border e Northern Australia quarantine strategy Border surveillance in Australia is delivered by an array of different government systems, particularly at ports of entry. This case study features a surveillance program that is particularly comprehensive and unusual. Established in 1989, the Australian government's Northern Australia Quarantine Strategy (NAQS) is designed to address some of the unique quarantine and biosecurity challenges facing Australia's northern coastline, from Broome in Western Australia to Cairns in Queensland. The key driver behind NAQS is Australia's close proximity to Indonesia, Timor Leste and especially Papua New Guinea (PNG), the coast of which at its closest point is less than 5 km from Australia's Saibai Island. These northern neighbours have many pests, diseases and weeds not present in Australia and there is a very real risk of movement south, by natural or human-mediated means. Early detection of any such incursions would facilitate eradication before their establishment and resulting negative impact on Australia's natural environment or plant-based industries. However, the extensive coastline, low population density and relative scarcity of commercial plantings pose challenges to effective surveillance outcomes. As a result, NAQS has established a strategy that combines surveillance conducted by experienced plant scientists. The program has a strong public awareness program, encouraging local people and communities to recognise and report unusual problems. A cornerstone of this strategy is the ongoing cooperation and goodwill of the northern indigenous people. This is achieved through their active involvement in NAQS survey activities at several levels, from facilitating access permissions to local ranger groups assisting on the ground. Target lists of plant diseases, insect pests and weeds are developed through assessment of the risks, pathways and hosts. These form a focus for the survey of domestic plantings of likely host plants, community farms and nurseries. NAQS scientists collect and analyse specimens after processing them to ensure no biosecurity risk. This ongoing specimen-based record of plant health status can be used to confirm presence or absence of targeted threats. A typical example of such surveillance in action is outlined below from the Torres Strait: 44,000 km2 of ocean that separates Australia from PNG, with numerous islands, 17 major settlements, and an indigenous population of around 7000 people. The risk of pest and disease movement is actively increased by monsoon weather patterns and traditional movements of people and trading
C. Anderson et al. / Crop Protection 100 (2017) 8e20
of goods permitted under international treaty. These islands have a clear history of acting as natural stepping stones for entry of pests from the north. Examples include spiralling white fly (Aleurodicus disperses Russell 1965) (Grimshaw et al., 2006) and black Sigatoka disease (Mycosphaerella fijiensis Morelet 1963) of banana (Hayden et al., 2003). In August 2008, a NAQS survey team comprising a botanist, an entomologist and a plant pathologist spent a day on the small Torres Strait island of Warraber. The population is approximately 230 and most families on the island are accustomed to annual NAQS surveys. The team was examining backyards, asking permission at each household to examine food and other plants. Upon entering one property, the entomologist's eyes were drawn to a tomato plant showing dramatic signs of leaf mining damage: something not normally found on NAQS surveys. Closer observation revealed Agromyzid flies associated with the affected foliage. Samples of the flies were collected in vials of alcohol and mined leaves were left with Warraber's resident quarantine officer for carefully controlled rearing to facilitate emergence of adult confirmatory specimens. Later expert examination identified the insect as the vegetable leaf miner, Liriomyza sativae (Blanchard), previously only known to occur in distant parts of south-east Asia. Because of the potential risk this insect poses to a broad range of Australian vegetable industries, a chain of notifications and response actions were initiated by both the federal and Queensland governments. This example clearly demonstrates the value of trained and vigilant specialists working with local communities and other government agencies to detect an exotic pest. 11. Case study. Post border e the multiple plant pest surveillance program Australia has strong border quarantine systems and the Australian Department of Agriculture intercepts upwards of 20,000 confirmed exotic pests each year. There will always be a risk of pests entering Australia by both natural and human-mediated means. Thus ongoing post-quarantine border surveillance in urban areas is vital for the rapid detection of pest incursions (Stanaway et al., 2001). To assist with the early detection of plant pests, the Multiple Plant Pest Surveillance Program (MPPS) was initiated by the Australian Office of the Chief Plant Protection Officer and is implemented by state and territories (Wyle et al., 2008). The primary objective of the MPPS Program is to provide early detection of a number of agreed exotic and established relevant pests. In New South Wales the program assesses some pest targets beyond the greater Sydney region, but with the majority within the Sydney basin as it is the major entry point for most passengers and freight entering Australia. The MPPS Program was designed by examining potential incursion pathways for exotic pests to identify the most likely pathways. For most pests, risk pathways include the direct environment in or adjacent to possible entry points like Quarantine Approved Premises (QAPs). Following the pre-border inspections and clearance, QAPs receive recently imported plants or plant products to allow post entry quarantine treatments and inspection. Therefore QAPs may carry some level of risk given the cryptic nature of many plant pests. The existing processes to manage possible risks associated with QAPs are comprehensive; however they provide a documented basis for targeting surveys that go beyond simply looking in the general vicinity of a port of entry. Other possible entry points include areas of high tourist throughput like zoos, botanical gardens and freight hubs like shipping ports and airports. In other cases, potential for unregulated movements of host material into urban environments was considered when selecting surveillance sites. Target exotic pests include tramp ants
15
(e.g. red imported fire ant, yellow crazy ant), aphids (eg. Russian wheat aphid, Diuraphis noxia (Mordvilko)), parasitic bee mites (eg. Varroa jacobsoni (Oudemans)), Asian gypsy moth (Lymantria dispar L.) and Asiatic citrus psyllid (Diaphorina citri (Kuwayama)). Detection methods employed in the MPPS Program are varied and based on the biology and epidemiology of the targeted pest. For example, tramp ants are surveyed by manual collection of samples on hosts or on the ground, particularly near ports. Food lures using fishbased cat food or peanut butter may be used to attract ants to monitoring sites but pesticides are rarely used to kill ants: fruit fly traps use an attractant and a pesticide to capture flies visiting traps (Dominiak et al., 2015). Pheromone traps also are used to target Asian gypsy moth (Dominiak et al., 2013). Diuraphis noxia was detected in southern New South Wales in production areas but not by the MPPS program (New South Wales Department of Primary Industries, 2016). A network of exotic fruit fly pheromone traps across the Sydney basin provide feedback for exotic fruit fly incursions (Gillespie, 2003). This includes a small number of special traps for the Solanum fruit fly (B. latifrons (Hendel)) that are targeted in suburbs where the cultural practices of residents may increase the risk of pest incursion. Apart from traps, visual inspection of suitable host plants accounts for the primary method for many target exotics like Asian citrus psyllid (D. citri) or Asian citrus longicorn beetle (Anoplophora chinensis (Forster)). Importantly the setting of explicit targets for active surveillance targets should be tempered with the inclusion of the inspection of any other targets that ‘look damaged’ to the professional eye. The detection in NSW of two non-target exotic species (Corythuca ciliata (Say) and Chaitophorus sp), albeit of minor or no economic significance (Dominiak et al., 2008), attest to field surveillance skills and taxonomic expertise validating and supporting targeted surveillance activities. The MPPS Program also plays an important role in monitoring the current distribution and host range of a small number of established pests. Under a global warming regime it is predicted that temperature and rainfall changes may potentially alter a pest range, which may impact on pest status between states with implications for trade. Surveys are targeted to areas where pest distributions would likely extend as a result of climate change. For example, the MPPS Program monitors established pests where current distribution is not well understood (e.g. Bactrocera cucumis (French)) and conducts delimiting surveys where there is a need to establish pest status in an area e.g. melon thrips (Thrips palmi Karny), spiralling whitefly (A. dispersus) and silverleaf whitefly biotype B and Q (Bemisia tabaci (Gennadius)). A range of collection methods are employed for established fruit flies, in addition to the trapping methods mentioned previously. For example, visual inspection and collection of adults on hosts, and collection of host fruits for subsequent larval rearing is an important sampling strategy especially for B. cucumis and other rainforest species. Host leaf collection for pest thrips and whitefly targets is standard with identification confirmed by a diagnostic lab. Methodologies follow the principles outlined in McMaugh (2005). The MPPS Program also undertakes general surveillance through the provision of information to the public in a bid to increase quarantine awareness and promote early detection. In the Sydney basin, information brochures about quarantine surveillance are distributed in high risk suburbs including areas adjacent to QAPs. Individuals are encouraged to report suspect detections through the Exotic Plant Pest Hotline. 12. Case study. Post border e Australian cotton disease surveys The methodology described for cotton could be extrapolated to
16
C. Anderson et al. / Crop Protection 100 (2017) 8e20
the full diversity of Australian plant production, both large and small, with modifications as required. Things to take into account are (1) pests and diseases appear in different phases or stages of crop development so monitoring should preferably be conducted during the emergence of main or dominant disease and pest species and (2) a survey of 10% of the crop acreage in a farm or region provides accurate information on the prevalence and development of diseases and pests. In Australia, a number of surveillance programs are supported and conducted by the cotton industry and federal and state governments as part of the surveillance strategies for exotic and endemic pests of cotton. This case study will focus on state surveillance systems that employ general multi-pest surveys to provide an early warning of the incidence and spread of endemic pests, and the incursion of new and exotic pests of cotton. The cotton industry in Australia runs biannual disease surveys conducted by plant pathologists of the New South Wales and Queensland state governments. The objective of these surveys is to monitor the distribution and importance of key endemic pests and record the presence or absence of new or exotic diseases. Seven endemic diseases are routinely assessed. Absence data for several high priority exotic pathogens as identified through the industry biosecurity planning process is also collected and provided to Plant Health Australia for entry into the NPSRT database. Surveys have been conducted in New South Wales since 1984/5 and in Queensland since 2002/3, with surveillance undertaken by experienced plant pathologists. The survey was designed to assess disease among a random sample of representative plants in each field, avoiding edge and row effects that may confound results. Consequently early detection of exotic viral diseases is provided by a separate surveillance system. Fields are selected from a representative set of survey farms chosen from each commercial cotton growing region. Plants are assessed along at least two 200 m diagonal transects in each field, where 10 groups of 10 plants spaced 20 m apart are destructively sampled. Four rows of plants are visually inspected while walking, which when taken over two transects, equates to approximately 4.5 km of crop, or 45,600 plants visually assessed in each field. GPS data is recorded at each survey point to ensure that the same entry point can be revisited. Disease scoring methods vary between early and late season surveys depending upon the disease being assessed. For example, during early surveys, plant stand is assessed in one metre of row, following which 10 plants are dug and assessed for black root rot (Thielaviopsis basicola (Berk & Broome) Ferraris 1912). During late season surveys, plants are first assessed for the presence and severity of Alternaria leaf spot, cotton bunchy top, and the incidence of boll rot caused by various fungi and bacteria. The stem of each plant is then cut and assessed for vascular discolouration, which may be caused by Fusarium, Verticillium or sudden wilt. Discoloured stems are sent to a diagnostic laboratory for confirmation of the presence of endemic, unknown or exotic strains of Fusarium oxysporum f.sp. vasinfectum, (G.F. Atkinson) Snyder & Hansen or Verticillium dahliae Kleb. Above ground symptoms are relatively easy to visually detect and sample, however soil borne disorders are frequently less obvious visually and require more resources to sample (Knox et al., 2006a, 2006b). Similar issues prevail for crops such as potatoes and grapes. Two cotton examples highlighting how routine diagnostics provides early detection of potential disease problems are: 1) the recent identification of Nematospora coryli (Kurtzman 1995), a pathogenic yeast associated with seed and internal boll rot of cotton. In the USA this pathogen has caused up to 60% fibre loss and reduced seed quality. This insect-vectored pathogen is suspected of being present undetected in Australia for at least ninety years prior to its isolation from citrus in 2004 (Shivas
et al., 2005). It was identified for the first time from cotton bolls in 2010 collected during late season surveys. 2) in November 2012 during annual disease surveys, stunted plants were observed in one field. The roots of these plants had numerous swellings. Root samples were collected and reniform nematode (Rotylenchulus reniformis Linford and Oliveira 1940) was confirmed in the roots by nematologist Jenny Cobon (DAF Queensland). This finding established the need for an intensive survey of the region which determined that reniform nematode was prevalent across the district at populations causing up to 30% yield loss. An industry funded research project has commenced to investigate strategies to manage this new issue. This case study demonstrates the usefulness of general multipest surveillance systems in broadacre agriculture in providing (1) an ongoing evaluation of current integrated disease management practices and (2) early detection for a suite of exotic pests and previously unknown pests. 13. Case study. Post border e early detection of cotton leaf curl disease Cotton leaf curl disease (CLCuD) (Begomovirus spp.) causes serious economic losses to cotton production in many regions where it occurs, in particular the Indian subcontinent. For example, CLCuD cost the Pakistan industry an estimated US$5 billion between 1992 and 1997. During the 2009-1010 season losses of up to 100% were observed in many growing regions in India (Rajagopalan et al., 2012). CLCuD is not known to be present in Australia and represents a serious biosecurity risk. The disease was first recorded in Africa in the 1920s, in Pakistan and India during the 1990s and most recently in 2010 in China (Cai et al., 2010). The disease is present in Pakistan, India, Sudan, Egypt, China, Tanzania, Malawi and Nigeria. In Sudan during the 1950s the disease caused yield losses of up to 40% (Tarr, 1957, 1964; Malathi et al., 2003). In preparedness for potential incursions of CLCuD the Australian cotton industry has developed a surveillance system to facilitate early detection of the disease and/or expedite implementation of delimiting surveillance if an incursion occurs. The surveillance system for detecting CLCuD was developed using the endemic Cotton bunchy top disease and Tomato yellow leaf curl virus as model systems. Training staff in the recognition of symptoms of CLCuD and when and how to inspect plants has also been an important component of this research. CLCuD can be detected in cotton crops by visual inspection of plants for disease symptoms. The pattern of disease spread within a crop for diseases caused by insect-transmitted viruses often exhibit an ‘edge effect’ where disease incidence is higher in plants closer to the inoculum source. This is usually a result of the proximity of crop plants to an area of perennial volunteer plants, weeds or other alternative hosts for the pathogens and/or insect vectors, particularly if the crop is located downwind from the inoculum source. This edge effect is normally evident in early infections where disease incidences are low. The recommended method to detect early outbreaks of CLCuD includes concentrated inspections in areas adjacent to potential sources of pathogen inoculum. However, in some instances an inoculum source may not be obvious and survey of all edges of the cotton block should be considered. The surveillance system for detecting CLCuD was developed to facilitate early detection to minimise the potential economic costs to the Australian cotton industry. It was validated in India where CLCuD is endemic and inspection of Australian cotton crops using this system has commenced. The same system will be useful for delimiting surveys during an eradication campaign and to establish pest free status of an area post-eradication or to provide an ongoing
C. Anderson et al. / Crop Protection 100 (2017) 8e20
monitoring system to support disease management if eradication is not feasible. The recommended system for detection of CLCuD is based on using confidence intervals for detecting disease at given incidences as described by Cannon and Roe (1982) and used in the eradication of citrus canker from Queensland from 2004 to 2009 (Gambley et al., 2009). This model suggests if disease is present at an incidence of at least 1%, inspection of 300 individuals will provide a 95% confidence in detecting that disease. This assumes a 100% success in detection of the disease-affected individuals and that individuals are selected randomly for inspection. In practice, a 100% success in disease detection is unlikely, thus the number of individuals inspected is increased to accommodate this. For example, assuming detection success is only 50%, inspection of 600 individuals is required to maintain the same 95% confidence interval. It is also impractical to inspect randomly selected plants. Therefore, the strategy was modified to provide a practical approach to survey cotton crops for CLCuD incorporating both the expected surveillance sensitivity and the expected higher incidence of disease on crop edges. 14. Case study. Post border e emergency surveillance for Asiatic citrus canker Citrus canker is a serious disease of citrus caused by the bacterium Xanthomonas citri subsp. citri. This disease causes major economic losses in southeast Asia where it originated and in many citrus growing countries including the USA (Irey et al., 2006; Gottwald and Irey, 2007), Argentina (Canteros, 2004) and Brazil (Leite and Mohan, 1990). It is absent from Australia but remains a major threat to the Australian citrus industry both from direct crop losses and the potential loss of export markets. In July 2004, citrus canker was detected in the Queensland growing region surrounding the township of Emerald. A Pest Quarantine Area (PQA) and the National Citrus Canker Eradication Program were established to manage containment and eradication (Gambley et al., 2009). The PQA had a radius of about 50 km encompassing the township and all cultivated citrus in the Emerald district. Eradication was accomplished and finally declared in early 2009, after expenditure of $17.6 million and over 200 000 staff hours (Gambley et al., 2009). The cost was well below the net benefit of $70 million to Queensland alone, estimated by the Australian Bureau of Agricultural and Resource Economics. Surveillance was a key component of the containment and eradication program. Surveillance data was used for a range of different decision making processes, several different strategies were used, consistent with International Standards for Phytosanitary Measures. These included delimiting the host destruction area, detection surveys to establish area freedom for growing districts outside the PQA, epidemiology studies, monitoring disease decline and eradication in residual host material, and monitoring for disease freedom on newly planted citrus trees after the eradication period. Delimiting surveys were required at the outset to determine the extent of the incursion in Queensland. A notable learning from the delimiting surveys was that time should be allowed for the disease to develop and spread to a detectable level. Citrus canker was detected on three properties in the Emerald area but it took ten months following the first report of the disease and three rounds of surveillance to confirm its presence on the third property. The disease was probably present on all three properties during the first round of surveillance but at that time the incidence and severity on the second and third properties was below surveillance sensitivity. It wasn't until a period of environmental conditions favourable for rapid disease development and spread occurred that the disease
17
was detected on the subsequent two properties. Delimiting surveillance strategies evolved as the emergency response and eradication program unfolded, covering citrus and putative host species on all properties, public and private, in the entire PQA. Delimiting surveillance was undertaken in growing areas without established links in Queensland and all other states, to underpin area freedom status for these areas. In Queensland, this was designed statistically to give 95% confidence that disease was not present above a 1% incidence and included a total of 457 547 trees. Inspections were done on 329 commercial production properties and 13 citrus nurseries. Epidemiology studies (Gambley et al., 2009) aimed at quantifying the potential inoculum present in the PQA included inspecting all trees surrounding a confirmed positive infection, documenting the disease severity and the location of the disease on those trees. The data from this information provided insight on the amount of citrus canker inoculum present and gave valuable clues to how long the disease was present prior to detection, the extent and direction of its within-block spread, and the factors affecting its spread. The low amount of inoculum detected and the limited spread of the disease on the second and third properties assisted in making decisions pertinent to cost-effective eradication of the disease. For example, the data were used to determine the zone of destruction of native citrus surrounding the infested properties. A second strategy for evaluating inoculum was to survey for potential reservoirs of the bacterium in the absence of living host material. Following tree destruction, a non-host period of 18 months was imposed and infested properties were subjected to multiple rounds of cultivation to enhance the microbial degradation of any remaining citrus material. Towards the end of this period a survey was done of 300 sites, each 0.8 m2 in size, spread across the three infested properties, to inspect for presence of residual citrus material, which if detected, was tested for the bacterium. The survey included inspecting the soil surface at each site and the subsurface to a depth of up to 30 cm. The results from this survey provided confidence that the host-free period was sufficient and there was a negligible risk of emergence of the disease on newly planted hosts. Finally, newly planted citrus had to be monitored to demonstrate success in eradication and to justify a claim for pest area freedom. This involved inspecting young trees in the propagation nursery prior to arrival at the properties, then multiple inspections over time once they were planted. The earliest plantings were monitored for 18 months. The data from all the various surveillance systems used allowed successful initial containment of the disease, its subsequent eradication and provided the statistical basis for establishing pest free area status for the Emerald citrus growing district. Pest Freedom was acknowledged in early 2009. 15. Case study. Post border e maintaining pest free areas As a major producer and importer of plants and plant products, Australia places a high priority on maintaining its ability to export products to pest-sensitive markets. Pest free areas (PFA) are a mechanism for facilitating trade into pest-sensitive markets (IPPC, 1998). They provide assurance to the importing country that a particular pest does not exist in a defined area and that no disinfestation treatment is required prior to market access. PFAs can be small regulated areas, or a state, or an entire country, or a region comprised of several states or countries. Areas of Low Pest Prevalence (ALPP) are an intermediate step between pest free area and endemic areas (IPPC, 2005). ALPP allow low numbers of pests to be present and permit control measures to prevent pest populations reaching populations equivalent to breeding populations
18
C. Anderson et al. / Crop Protection 100 (2017) 8e20
(Dominiak et al., 2015). Endemic areas are deemed to have breeding populations present at all times and all produce must be disinfested before being sent to pest sensitive countries. Fruit flies are commonly considered to be the world's most significant horticultural pest group. Different species are adapted to particular environments and countries. In Australia, the native Qfly is the most economically important fruit fly, particularly for eastern Australia. Qfly is not present in the states of Western Australia and South Australia (Dominiak and Daniels, 2012). By conducting targeted surveillance, these states can demonstrate long periods of absence in which Qfly has not been recorded and therefore easily satisfy PFA requirements. Both states maintain a series of trapping grids to monitor Qfly as a requirement to maintain pest free status. In addition, parts of Victoria, New South Wales and South Australia had formed a PFA named the Fruit Fly Exclusion Zone (FFEZ) (Anon, 1996) which straddled a large area of horticultural production. Within this PFA, surveillance was conducted to demonstrate continuing absence of both Qfly and Medfly (Jessup et al., 2007). Medfly has not been detected for more than 60 years in the FFEZ that is consequently recognized as a Medfly PFA (Dominiak and Daniels, 2012). Similarly Qfly was not established in the FFEZ although on occasions flies entered the area, were detected by the surveillance system and eradicated. Acceptance of the FFEZ as a Qfly PFA required surveillance for one year of demonstrated absence through specific surveys (Anon, 1996). An early detection surveillance program comprised of specific surveys because most trading partners are sensitive to Qfly and Medfly. Specific surveillance was responsible for providing confidence that produce harvested from the FFEZ was not infested with Qfly. Permanently deployed traps were inspected weekly in the summer and fortnightly in the winter with 41 inspections per year. The design of the traps and wick replacement was controlled by the trade agreement (Anon, 1996). Any detections of pest fruit fly in the monitoring grid must be reported to federal and state authorities within 48 h of detection: federal agencies immediately advised importing countries. There were two staged surveillance responses within the trade agreement. The standard early detection surveillance system uses Lynfield traps deployed on a 400 m network in towns and 1000 m array in orchards. When two male flies were detected, a second stage of delimiting surveillance was engaged, whereby the trapping array was intensified. An additional (or supplementary) 15 Lynfield traps and 16 MacPhail traps using a wet food lure were deployed to delimit the incursion and to determine in particular if there was a genuine breeding population present or if the trapped flies were an isolated incursion not associated with a resident population (Meats et al., 2003). A breeding population was a breach of the PFA rules while isolated incursions or transient populations provide no threat to PFA status. An outbreak was declared if five male flies were trapped within 1 km within 14 days and the area within 15 km of the epicenter was declared a quarantine area (Dominiak et al., 2011b; Dominiak, 2012) and the pest free status was suspended until the incursion was eradicated. Similarly if larvae were detected in locally produced fruit or if a gravid female was trapped, an outbreak was declared. All fruit within this quarantine area must then be treated if it was to be sold into fruit fly sensitive markets. Quarantine and eradication measures were deployed and the trapping network utilised to monitor the effectiveness of the eradication (Anon, 1996). Following successful eradication, the intensive trapping array was returned to the standard network. Florec et al. (2013) reported that it was more cost effective for pest free areas to exclude pests compared with the costs of eradication and losses associated with disinfestations and loss of market access based on their pest free status. The engagement of the general public would help support reporting of unusual events via
general surveillance. Various state and national websites advertise the restrictions for movement of produce and through road blocks where vehicles are inspected for potentially infested produce (Dominiak and Coombes, 2010). Community awareness of Qfly is promoted through extension material including pamphlets. Residents of the FFEZ are encouraged to report Qfly detections to state plant protection organisations. Ongoing maintenance of the FFEZ as a pest free area provides access for produce into fruit fly sensitive markets and reduces the cost on industry by negating the need for costly disinfestations. Engaged local communities and a surveillance network is likely to result in early detection. Early detection usually means that pests are contained to a small area and small areas are pivotal to successful eradication (Tobin et al., 2014). Australia was regarded as being a pest free area for B. papayae (Drew and Hancock) (Papaya fruit fly (PFF)). Bactrocera papayae was detected in Cairns (Queensland) in 1995 and all other Australian states regulated susceptible Queensland produce entering their states. These other states were obligated to establish a monitoring grid to demonstrate that the pest was not present in their state (Jessup et al., 2007; Gillespie, 2003) and that their trade restrictions were reasonable. All Australian states successfully demonstrated their pest free status by their targeted surveillance. Queensland could again declare that it was pest free from PFF after a range of control and management techniques were successful (Lloyd et al., 1998) and supported by zero detection in targeted and general surveillance. Until 2000, both Australia and New Zealand were considered a PFA for currant-lettuce aphid (CLA) Nasonovia ribisnigri (Mosley). In 2002, CLA was detected in New Zealand (Stufkens and Teulon, 2003). Australia regulated the movement of New Zealand lettuce products and intensified early detection surveillance in all Australian states to demonstrate its pest free status as a country. CLA was detected in Tasmania (an Australian island state) in 2004 (Hill, 2012). Mainland Australia remained pest free and regulated Tasmanian host produce in an effort to maintain their pest status and maintain their pest free market access. In 2005, CLA was detected in Victoria, a mainland state immediately north of Tasmania. The other mainland states then regulated the movement of susceptible produce from Victoria (and Tasmania) and maintained their CLA free status. There was therefore no limitation to trade between any mainland state except Victoria. By February 2006, CLA was detected in New South Wales and by December 2006, CLA had spread across all other states (Dominiak et al., 2009) and all Australian states lost their pest free CLA status. By 2006, however, improvements in integrated pest management including the introduction of resistant lettuce cultivars had minimised the adverse impacts of the pest. This case study demonstrates how surveillance was implemented to maintain pest freedom status. It also can provide a valuable forewarning of likely incursions and thereby prompt development of management strategies to minimise the impact of a pest before it arrives to maintain pest freedom status and to justify trade restrictions (Dominiak et al., 2009). The demonstration of pest freedom by surveillance also has significant trade advantages. Pest Free Areas remain a cornerstone of trade providing a competitive and financial advantage for those areas or states that can fulfill the requirements of a PFA. Surveillance, both general and targeted, remains pivotal for the maintenance of PFAs. 16. Concluding comments Plant biosecurity surveillance systems collect information about pest status which is used to inform decisions that impact on market access, management of established pests and responses to exotic plant pest incursions. These systems may deploy general surveillance and specific surveys at all stages of the plant biosecurity
C. Anderson et al. / Crop Protection 100 (2017) 8e20
surveillance cycle. Successful surveillance systems are usually based on sound quantitative designs, are adequately resourced, utilise accurate diagnostic services, and involve an element of community and stakeholder engagement to increase the value of general surveillance data. The case studies described in this paper highlight several successful plant biosecurity surveillance systems in Australia that demonstrate these principles in action. Historically, many exotic pests have been detected post-entry and establishment, rather than pre- or at-border, and eradication may not be technically feasible or economically justified. For example, Myrtle rust was probably present in Australia for many months if not some years prior to its detection in cut flower production (Carnegie and Cooper, 2011). Improving the ability of plant biosecurity systems to successfully detect pests before they establish or soon after should therefore be a priority for biosecurity managers across the globe. Detecting and stopping a pest before it is established is far more cost-effective than managing an established pest and reduces the overall impact on production, trade and the environment (Florec et al., 2013). Early detection will optimise the likelihood of eradication as affected areas remain small (Tobin et al., 2014). Acknowledgements Dr Paul Phelong provided mentoring in the early stages of the project. Dr Louise Rossiter reviewed and improved earlier versions of the manuscript. This review did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References ALA, 2011. Atlas of Living Australia. Available at: http://www.ala.org.au/ (Accessed 1 August 2011). Anon, 1996. Code of Practice for Management of Queensland Fruit Fly. Standing Committee on Agriculture and Resource Management. Department of Primary Industries, Canberra. Anon, 2007. Pest Risk Analysis (PRA) Training: Participant Manual. FAO, Rome. Aubert, B., 1990. Integrated activities for the control of huanglongbin-greening and its vector Diaphorina citri Kuwayama in Asia. In: Aubert, B., Tontyaporn, S., Buangsuwoon, D. (Eds.), Proceedings of the 4th International Asia Pacific Conference on Citrus Rehabilitation, pp. 133e144. Chiang Mai, Thailand, 4-10th Feb. 1990. Australian Government Department of Agriculture Fisheries and Forestry, 2009. Import Risk Analysis Handbook 2007. Canberra. AVH, 2011. Australia's Virtual Herbarium. Available at: http://chah.gov.au/avh/ (Accessed 1 August 2011). Barrett, S., Whittle, P., Mengersen, K., Stoklosa, R., 2009. Biosecurity threats: the design of surveillance systems based on power and risk. J. Environ. Ecol. Statistics 17 (4), 503e519. Cai, J.H., Xie, K., Lin, L., Qin, B.X., Chen, B.S., Meng, J.R., Liu, Y.L., 2010. Cotton leaf curl Multan virus newly reported to be associated with cotton leaf curl disease in China. New Dis. Rep. 20, 29. Cannon, R.M., Roe, R.T., 1982. Livestock Disease Surveys: a Field Manual for Veterinarians. Australian Government Publishing Service, Canberra. Canteros, B.I., 2004. Management of citrus canker in Argentina. A review. In: AitOubahou, A., El-Otmani, M. (Eds.), Proceedings of International Society of Citriculture, pp. 448e451. Agadir, Morocco. Carnegie, A.J., Cooper, K., 2011. Emergency response to the incursion of an exotic myrtaceous rust in Australia. Australas. Plant Pathol. 40, 346e359. Collins, S., Vanstone, V., 2010. WA PCN Freedom Claim. Final report for project PT10701. Horticulture Australia, Sydney, Australia. Davis, R.I., Moore, N.Y., Bentley, S., Gunua, T.G., Rahamma, S., 2000a. Further records of Fusarium oxysporum f.sp. cubense from New Guinea. Australas. Plant Pathol. 29, 224. Davis, R.I., Fegan, M., Tjahjono, B., Rahamma, S., 2000b. An outbreak of blood disease of banana in Irian Jaya, Indonesia. Australas. Plant Pathol. 29, 152. Davis, R.I., Jacobson, S.C., Rahamma, S., Gunua, T.G., 2000c. Surveillance for citrus huanglongbing (greening) disease in New Guinea and north Queensland. Australas. Plant Pathol. 29, 226. Davis, R.I., Moore, N.Y., Fegan, M., 2001. Blood disease and Panama disease: two newly introduced and grave threats to banana production on the island of New Guinea. In: Bourke, R.M., Allen, M.G., Salisbury, J.G. (Eds.), Food Security for Papua New Guinea. Proceedings of the Papua New Guinea Food and Nutrition 2000 Conference. PNG University of Technology, pp. 816e821. Lae, 26-30 June
19
2000 Australian Centre for International Agricultural research (ACIAR) Proceedings No. 99, 892p, Canberra. Davis, R.I., Kokoa, P., Jones, L.M., Mackie, J., Constable, F.E., Rodoni, B.C., Gunua, T.G., Rossel, J.B., 2012. A new banana wilt disease associated with phytoplasmas in Papua New Guinea. Australas. Plant Dis. Notes 7, 91e97. Dominiak, B.C., Gillespie, P.S., Tom, G., 2007. Changes in data management and the impediments to change e a case study for fruit fly surveillance. Plant Prot. Q. 22 (3), 95e98. Dominiak, B.C., Gillespie, P.S., Worsley, P., Locker, H., 2008. Survey for sycamore lace bug, Corythucha ciliata (Say) (Hemiptera: Tingidae) in New South Wales during 2007. General Appl. Entomology 37, 27e30. Dominiak, B.C., Links, I.J., Fletcher, M.J., Worsley, P., McDougall, S., 2009. Detection and spread of currant-lettuce aphid Nasonovia ribisnigri (Mosley) (Hemiptera: Aphididae) in New South Wales. General Appl. Entomology 38, 27e30. Dominiak, B.C., Coombes, N., 2010. Review of the impact of the Tristate community awareness program on road travellers e 1999/2000. Plant Prot. Q. 25 (1), 2e8. Dominiak, B.C., Worsley, P., Gillespie, P.S., Loecker, H., Kerr, M., Shattuck, S.O., 2010. Search for red imported fire ants Solenopsis invicta Burren in New South Wales from 2004-2006. Plant Prot. Q. 25 (1), 15e18. Dominiak, B.C., Gott, K., McIver, D., Grant, T., Gillespie, P.S., Worsley, P., Clift, A., Sergeant, E.S.G., 2011a. Scenario tree analysis of zero detections and the eradication of yellow crazy ant (Anoplepis gracilipes (Smith)), in New South Wales, Australia. Plant Prot. Q. 26 (4), 124e129. Dominiak, B.C., Daniels, D., Mapson, R., 2011b. Review of the outbreak threshold of Queensland fruit fly (Bactrocera tryoni Froggatt). Plant Prot. Q. 26 (4), 141e147. Dominiak, B.C., 2012. Review of dispersal, survival, and establishment of Bactrocera tryoni (Diptera: Tephritidae) for quarantine purposes. Ann. Entomological Soc. Am. 105, 434e446. Dominiak, B.C., Daniels, D., 2012. Review of the past and present distribution of mediterranean fruit fly (Ceratitis capitata Weidemann) and Queensland fruit fly Bactrocera tryoni Froggatt) in Australia. Aust. J. Entomology 51, 104e115. Dominiak, B.C., Gillespie, P.S., Subasinghe, R., 2013. Surveillance for Asian gypsy moth (Lymantria dispar asiatica L.) between 2005 and 2012 in New South Wales, Australia. Plant Prot. Q. 28 (1), 12e14. Dominiak, B.C., Wiseman, B., Anderson, C., Walsh, B., McMahon, M., Duthie, R., 2015. Definition of and management for areas of low pest prevalence for Queensland fruit fly Bactrocera tryoni Froggatt. Crop Prot. 72, 41e46. Dominiak, B.C., Worsley, P., 2016. Lesser Queensland fruit fly Bactrocera neohumeralis (Hardy) (Diptera: Tephritidae: Dacinae) not detected in inland New South Wales or South of Sydney. General Appl. Entomology 44, 9e15. Evans, G., McMaugh, T., 2005. Surveying for Plant Pests: ISPM Terminology. Workshop on Sanitary and Phytosanitary (SPS) Awareness. ASEANET, Bogor, Indonesia. Florec, V., Sadler, R.J., White, D., Dominiak, B.C., 2013. Choosing the battles: the economics of area wide pest management for Queensland fruit fly. Food Policy 38, 203e213. Froud, K.J., Bullians, M.S., 2010. Investigation of biosecurity risk organisms for the plant and environment domains in New Zealand for 2008 and 2009. N. Z. Plant Prot. 63, 262e269. Gambley, C.F., Miles, A.K., Ramsden, M., Doogan, V., Thomas, J.E., Parmenter, K., Whittle, P.J.L., 2009. The distribution and spread of citrus canker in Emerald, Australia. Australas. Plant Pathol. 38, 547e557. Gibson, J., 2000. The economic and nutritional importance of household food production in PNG. In: Bourke, R.M., Allen, M.G., Salisbury, J.G. (Eds.), Food Security for Papua New Guinea. Proceedings of the Papua New Guinea Food and Nutrition 2000 Conference. PNG University of Technology, pp. 37e44. Lae, 2630 June 2000 Australian Centre for International Agricultural research (ACIAR) Proceedings No. 99. Gillespie, P., 2003. Observations on fruit flies (Diptera: Tephritidae) in New South Wales. General Appl. Entomology 32, 41e47. Gottwald, T.R., Irey, M., 2007. Post-hurricane analysis of citrus canker II: predictive estimation of disease spread and area potentially impacted by various eradication protocols following catastrophic weather events. Online. Plant Health Prog. http://dx.doi.org/10.1094/PHP-2007-0405-01-RS. Grimshaw, J.F., Waterhouse, B.M., Weinert, M.P., 2006. The value of early detection and internal quarantine boundaries in the management of incursions: some examples in plant protection from northern Australia and Papua New Guinea. In: Pest and Disease Incursions: Risks, Threats and Management in Papua New Guinea. Australian Centre for International Agricultural research (ACIAR), Canberra, p. 199. TV Price p 154-160 Technical Reports No. 62. Hayden, H.L., Carlier, J., Aitken, E.A.B., 2003. Genetic structure of Mycosphaerella fijiensis populations from Australia, Papua New Guinea and the Pacific islands. Plant Pathol. 52, 703e712. Hill, L., 2012. The currant lettuce aphid, Nasonovia ribisnigri arrives in Tasmania: part 1. Vic. Entomol. 42, 29e31. Hill, L., 2013. The common armyworm, Mythima converta (Walker) (Noctuidae: Lepidoptera), a seasonal resident in Tasmania. Plant Prot. Q. 28, 114e119. Holmes, T.P., Bell, K.P., Byrne, B., Wilson, J.S., 2008. Economics aspects of invasive forest pest management. In: Holmes, T.P., Prestemon, J.P., Abt, K.L. (Eds.), The Economics of Forest Disturbancesdwildfires, Storms, and Invasive Species. Springer, Berlin, pp. 381e406. Hulme, P.E., 2009. Trade, transport and trouble: managing invasive species pathways in an era of globalization. J. Appl. Ecol. 46, 10e18. International Plant Protection Convention, 1997. ISPM No. 6: Guidelines for Surveillance. IPPC-FAO, Rome.
20
C. Anderson et al. / Crop Protection 100 (2017) 8e20
International Plant Protection Convention, 1998. ISPM No. 8: Determination of Pest Status in an Area. IPPC-FAO, Rome. International Plant Protection Convention, 2004. ISPM No. 11: Pest Risk Analysis for Quarantine Pests Including Analysis of Environmental Risks and Living Modified Organisms. IPPC-FAO, Rome. International Plant Protection Convention, 2005. ISPM No. 22. Requirements for the Establishment of Areas of Low Pest Prevalence. IPPC-FAO, Rome, p. 7. International Plant Protection Convention, 2010. ISPM No. 5: Glossary of Phytosanitary Terms. IPPC-FAO, Rome. International Plant Protection Convention, 2011. International Plant Protection Convention. Available at: www.ippc.int (Accessed 1 August 2011). Irey, M., Gottwald, T.R., Graham, J.H., Riley, T., Carlton, G., 2006. Post-hurricane Analysis of Citrus Canker Spread and Progress towards the Development of a Predictive Model to Estimate Disease Spread Due to Catastrophic Weather Events. Online. Plant Health Progress. http://dx.doi.org/10.1094/PHP-20060822-01-RS. Jarrad, F., Whittle, P., Barrett, S., Mengersen, K., 2010. Barrow Island's biosecurity: catching the unknown invader. Significance 7 (2), 53e57. Jessup, A.J., Dominiak, B.C., Woods, B., De Lima, C.P.F., Tomkins, A., Smallridge, C., 2007. Area-wide management of fruit flies in Australia. In: Vreysen, M.J.B., Robinson, A.S., Hendrich, J. (Eds.), Area-wide Control of Insect Pests: from Research to Field Implementation. Springer, Dordrecht, The Netherlands, pp. 685e697. Kelly, P.L., Reeder, R., Kokoa, P., Arocha, Y., Nixon, T., Fox, A., 2011. First report of a phytoplasmas identified in coconut palms (Cocos nucifera) with lethal yellowing-like symptoms in Papua New Guinea. New Dis. Rep. 23, 9. Knox, O.G.G., Anderson, C.M.T., Allen, S.T., Nehl, D.B., 2006a. Helicotylenchus dihystera in Australian cotton roots. Australas. Plant Pathol. 35, 287e288. Knox, O.G.G., Anderson, C.M.T., Nehl, D.B., Vadakattu, V.S.R.G., 2006b. Observation of Tylenchorhynchus ewingi in association with cotton soils in Australia. Australas. Plant Dis. Notes 1, 47e48. Kruger, H., Thompson, L., Clarke, R., Stenekes, N., Carr, A., 2009. Engaging in Biosecurity: Gap Analysis. Bureau of Rural Sciences, Canberra, Australia. Lefoe, G., Dominiak, B.C., Worsley, P., Davies, J., 2014. Elm leaf beetles Xanthogaleruca luteola (Muller) dispersal across south eastern Australia (1989-2011). Plant Prot. Q. 29 (2), 61e65. Leite, R.P., Mohan, S.K., 1990. Integrated management of the citrus bacterial canker disease caused by Xanthomonas campestris pv. citri in the State of Parana, Brazil. Crop Prot. 9, 3e7. Low-Choy, S., Whittle, P., Anderson, C.M.T., 2011. Quantitative approaches to designing plant biosecurity surveillance systems. In: McKirdy, S. (Ed.), Biosecurity in Agriculture and the Environment. Canberra, Australia. Lloyd, A., Leach, P., Kopittke, R., 1998. Effects of exposure on chemical content and efficacy of male annihilation blocks used in the eradication of Bactrocera papayae in North Queensland. General Appl. Entomology 28, 1e8. Magarey, R.D., Colunga-Garcia, M., Fieselmann, D.A., 2009. Plant biosecurity in the United States: roles, responsibilities and information needs. BioScience 59, 875e884. Malathi, V.G., Radhakrishnan, G., Varma, A., 2003. In: Lobenstein, G., Thottappilly, G. (Eds.), Cotton. In 'Virus and Virus-like Diseases of Major Crops in Developing Countries'. Kluwer Academic Publishers. McMaugh, T., 2005. Guidelines for Surveillance for Plant Pests in Asia and the Pacific. ACIAR Monograph No. 119, Canberra, Australia. Meats, A.W., Clift, A.D., Robson, M.K., 2003. Incipient founder populations of mediterranean and Queensland fruit flies in Australia: the relation of trap catch to infestation radius and models for quarantine radius. Aust. J. Exp. Agric. 43, 397e406. Nairn, M.E., Allen, P.G., Inglis, A.R., Tanner, C., 1996. Australian Quarantine: a Shared Responsibility. Department of Primary Industries and Energy, Canberra.
National Plant Biosecurity Status Report, 2012. Plant Health Australia, Canberra ACT, Australia. New South Wales Department of Primary Industries, 2016. Russian wheat aphid. Primefact 1489, 3. Pimentel, D., McNair, S., Janecka, J., Wightman, J., Simmonds, C., O'Connell, C., Wong, E., Russel, L., Zern, J., Aquino, T., Tsomondo, T., 2001. Economic and environmental threats of alien plant, animal, and microbe invasions. Agric. Ecosyst. Environ. 84, 1e20. Plant Health Australia, 2010. PLANTPLAN: Australian Emergency Plant Pest Response Plan. Version 2. Plant Health Australia, Canberra, ACT. Plant Health Australia, 2011a. National Plant Surveillance Reporting Tool. Available at. : www.planthealthaustralia.com.au/our_projects/psdb/index.cfm (Accessed 1 August 2011). Plant Health Australia, 2011b. Australian Plant Pest Database. Available at: www. planthealthaustralia.com.au/go/phau/capacity-and-capability/informationsupport-systems/appd (Accessed 1 August 2011). Quarantine Domestic, 2011. Quarantine Domestic. Available at: www. quarantinedomestic.gov.au (Accessed 1 August 2011). Rajagopalan, P., Naik, A., Katturi, P., Kurulekar, M., Kankanallu, R., Anandalakshmi, R., 2012. Dominance of resistance-breaking Cotton leaf curl Burewala virus (CLCuBuV) in northwestern India. Archives Virology 157, 855e868. Simmonds, N.W., 1962. The Evolution of the Bananas. Longman, London. Shivas, R.G., Smith, M.W., Marney, T.S., Newman, T.K., Hammelswang, D.L., Cooke, A.W., Pegg, K.G., Pascoe, I.A., 2005. First record of Nematospora coryli in Australia and its association with dry rot of Citrus. Australas. Plant Pathol. 34, 99e101. Stanaway, M.A., Zalucki, M.P., Gillespie, P.S., Rodriguez, C.M., Maynard, G.V., 2001. Pest risk assessment of insects in sea cargo containers. Aust. J. Entomology 40, 180e192. Stufkens, M.A., Teulon, D.A.J., 2003. Distribution, host range and flight pattern of the Lettuce aphid in New Zealand. N. Z. Plant Prot. 56, 27e32. Suckling, D.M., Kean, J.M., Stringer, L.D., Caceres-Barrios, C., Hendrichs, J., ReyesFlores, J., Dominiak, B.C., 2016. Eradication of Tephritid fruit fly populations: outcomes and prospects. Pest Manag. Sci. 72, 456e465. Swingle, W.T., 1967. The botany of citrus and its wild relatives. In: Reuther, W., Webber, N.J., Batchelor, L.D. (Eds.), Citrus Industry, pp. 190e430. Tarr, S.A.J., 1957. The host range of cotton leaf curl virus in Sudan. Emp. Cotton Grow. Rev. 34, 258e262. Tarr, S.A.J., 1964. Virus Disease of Cotton. Commonwealth Mycological Institute, Kew, Surrey, England. Tobin, P.C., Kean, J.M., Suckling, D.M., McCullough, D.G., Herms, D.A., Stringer, L.D., 2014. Determinants of successful arthropod eradication programs. Biol. Invasions 16, 401e414. Thompson, L., Stenekes, N., Kruger, H., Carr, A., 2009. Engaging in Biosecurity: Literature Review of Community Engagement Approaches. Bureau of Rural Sciences, Canberra, Australia. van Halteren, P., 2000. Diagnostics and national plant protection organisations. OePP/EPPO Bull. 30, 357e359. Weinert, M.P., Jacobson, S.C., Grimshaw, J.F., Bellis, G.A., Stephens, P.M., Gunua, T.G., Kame, M.F., Davis, R.I., 2004. Detection of huanglongbing (citrus greening disease) in Timor-Leste (east timor) and in Papua New Guinea. Australas. Plant Pathol. 33, 135e136. Wilson, J., Stephenson, B.P., Gill, G.S.C., Randall, J.L., Vieglas, C.M.C., 2004. Principles of response to detections of new plant pest species and the effectiveness of surveillance. N. Z. Plant Prot. 57, 156e164. Wyle, F.R., Griffiths, M., King, J., 2008. Development of hazard site surveillance programs for forest invasive species: a case study from Brisbane, Australia. Aust. For. 71, 229e235.
Contents lists available at ScienceDirect
Crop Protection journal homepage: www.elsevier.com/locate/cropro
Australian plant biosecurity surveillance systems Chris Anderson a, Samantha Low-Choy b, c, Peter Whittle c, Sharyn Taylor b, € cker a, Richard Davis e, Cherie Gambley d, Linda Smith d, Peter Gillespie a, Holger Lo Bernie Dominiak a, * a
Biosecurity and Food Safety, NSW Department of Primary Industries, Locked Bag 21, Orange, New South Wales, 2800 Australia Plant Biosecurity CRC, University of Canberra, Bruce, Australian Capital Territory, 2617 Australia c Mathematical Sciences, Queensland University of Technology, GPO Box 2434, Brisbane, Queensland, 4001 Australia d Queensland Department of Agriculture and Fisheries, Applethorpe Research Station, GPO Box 501, Stanthorpe Queensland 4378 Australia e Northern Australia Quarantine Strategy, Department of Agriculture and Water Resources, PO Box 96, Cairns International Airport, Queensland 4870 Australia b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 December 2016 Received in revised form 24 May 2017 Accepted 24 May 2017
Australia is an island nation and a primary producer of agricultural and horticultural products. There is a large diversity of plant biosecurity threats which could adversely impact on Australia's production and exports. Surveillance has traditionally been used to monitor pests and optimise production. Increasingly surveillance is being used for early detection of exotic incursions, demonstration of eradication of incursions and pest freedom from exotic or endemic pests. These newer uses of surveillance utilise general and specific surveillance: surveillance data is maintained in electronic databases. Specific surveillance is a targeted surveillance search used by industry or state regulators for a specific pest to support pest freedom or other trade standards. The plant biosecurity surveillance cycle shows the flow of surveillance operations. In this paper, this cycle is demonstrated by case studies including pre border and the northern Australian at-border surveillance for the Australian-Asian interface. Within Australia, the multiple plant pest surveillance program was established in most capital cities where there are high flows of population and produce. As an industry example, the cotton industry surveillance program, particularly for cotton leaf curl, demonstrates how plant biosecurity surveillance operates within an industry. Asiatic citrus canker is another example of industry pertinent surveillance. Finally, surveillance for the purpose of declaring pest freedom areas is reviewed using fruit flies and currant lettuce aphid as examples. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Market access Freedom Detection Monitoring
Contents 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
Introduction - a diversity of threats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Why do we do surveillance? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pest records to support pest status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . How do we do surveillance? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . General surveillance? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Specific surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The plant biosecurity surveillance cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case studies: successful surveillance systems in action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . Case study. Pre border - surveillance North of Australia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case study. Border e Northern Australia quarantine strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case study. Post border e the multiple plant pest surveillance program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case study. Post border e Australian cotton disease surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
* Corresponding author. E-mail address: [email protected] (B. Dominiak). http://dx.doi.org/10.1016/j.cropro.2017.05.023 0261-2194/© 2017 Elsevier Ltd. All rights reserved.
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. 9 . 9 . 9 11 11 12 12 13 14 14 15 15
C. Anderson et al. / Crop Protection 100 (2017) 8e20
13. 14. 15. 16.
Case study. Post border e early detection of cotton leaf curl disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case study. Post border e emergency surveillance for Asiatic citrus canker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case study. Post border e maintaining pest free areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Concluding comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction - a diversity of threats Invasive species cost an estimated US$314 billion annually, with over 120,000 plant, animal and pest species believed to have invaded Australia, Brazil, India, South Africa, the United States of America and the United Kingdom alone (Pimentel et al., 2001). Recognising the need for international standards to facilitate trade of plants and plant produce around the world, while minimising the movement of plant pests, the International Plant Protection Convention (IPPC) develops International Standards for Phytosanitary Measures (ISPMs). These standards include surveillance, defined as ‘an official process that collects and records data on pest occurrence or absence by survey, monitoring or other procedures’ (IPPC, 2010). As such, plant biosecurity surveillance provides a scientific basis for the assessment and management of risks posed to plant production and the environment by plant pests. The term ‘pest’ refers to “… any species, strain or biotype of plant, animal or pathogenic agent, injurious to plants or plant products” (IPPC, 2010). The modern age of globalisation means that the threat from unwanted pests is greater than ever before (Hulme, 2009; Stanaway et al., 2001). Global pathways of entry through tourism, immigration and trade are open and largely unregulated in many areas (Hulme, 2009). Data published on the IPPC website indicates that Australia has reported on the status of 49 new plant pests comprised of 17 fungi, 12 invertebrates, eight viruses and virus like organisms, five bacteria and seven weeds (Table 1 covers the period from 13 November 2006 to 21 January 2014). These likely conservative numbers demonstrate the diversity of biosecurity threats confronting managers, scientists and persons designing and implementing surveillance systems. Of these pests, five have been eradicated, and a further 10 are under eradication or official control. The remaining detections were considered not feasible or cost-beneficial to eradicate, reflecting the fact that they were of minor concern or, more significantly, that detection did not occur until post-border establishment and spread. Moreover, many of these plant pests are cryptic, being vectored within or on other organisms and having symptoms that are hard to distinguish from endemic and established pests or abiotic constraints. Because of this they are difficult to detect through traditional surveillance activities. 2. Why do we do surveillance? Plant biosecurity is a suite of risk management activities deployed across a continuum of jurisdictions, from pre-border, at border, to post-border activities that minimise the impact of plant pests on natural and managed ecosystems (Magarey et al., 2009; Nairn et al., 1996). Surveillance is a fundamental component of any plant biosecurity system, as knowledge of pest status is the basis for managing risk. Plant biosecurity surveillance systems are designed and implemented to assess pest status for many different reasons (McMaugh, 2005). These fall broadly into three categories:
9
16 17 17 18 19 19
1. surveillance to facilitate trade (e.g. area freedom, areas of low pest prevalence, pest free places of production); 2. surveillance to support pest management and control (e.g. commodity specific pest surveillance and monitoring of endemic pests); and 3. surveillance to facilitate early detection, and support and validate responses to pest incursions. Surveillance systems often serve more than one of these purposes, which are intrinsically linked. For example in south eastern Australia, routine surveillance in the Fruit Fly Exclusion Zone (FFEZ) (Anon, 1996) was carried out systematically to validate area freedom (Category 1), and support pest management and control (Category 2) by ensuring early detection (Category 3) so that measures could be rapidly implemented to prevent establishment of Queensland fruit fly (QFly) (Bactrocera tryoni (Froggatt)) and provide early detection of Mediterranean fruit fly (Medfly) (Ceratitis capitata (Weidemann)) (Dominiak and Daniels, 2012). Such measures thereby facilitated market access for fruit grown within the FFEZ on the basis of area freedom. 3. Pest records to support pest status Surveillance is a tool used to generate pest records, which in turn support determinations of pest status, through sufficient evidence in the form of accurate and reliable pest records (IPPC, 1998; van Halteren, 2000). The accuracy and reliability of a pest record is directly impacted by the design and implementation of the surveillance system through which it was generated. A pest record ‘provides information concerning the presence or absence of a pest, the time and location of the observations, host(s) where appropriate, the damage observed, as well as references or other relevant information pertaining to a single observation’ (IPPC, 1998). Substantial effort is directed towards ensuring individual reliability of pest records. The reliability of pest records ‘is based on consideration of the data in regard to the collector/identifier, the means of technical identification, the location and date of the record, and the recording/publication of the record’ (IPPC, 1998). Further, the reliability of pest records will be improved where nationwide or international database and metadata standards are in place to provide consistency and coherency in pest record management. Hill (2013) noted the importance of verifying old records with contemporary surveillance to ensure the validity of historic records in databases. The incorporation of fully documented voucher pest specimens in institutional repositories is imperative to support surveillance activities. Pests may impact several categories including economic, social and environmental categories. Economic pests impact on individual producers, groups of producers and regional/international market access. This category of pest usually drives a funded response or eradication. Australia conducts surveillance for fruit flies to demonstrate freedom from particular species or to detect incursions (Dominiak and Daniels, 2012; Dominiak and Worsley, 2016). Pests such as Asian gypsy moths (Lymantria spp.) cause
10
C. Anderson et al. / Crop Protection 100 (2017) 8e20
Table 1 Plant pest reports in Australia and their status. Information published on IPPC website - correct as of February 2014 (IPPC, 2011). Dates of incursions range from 13 November 2006 to 21 January 2014. Note that Vegetable leaf minor was originally detected in this period and was detected again in 2017 and the table includes this more recent detection for completeness. Common name
Scientific name
Host Commodity
Taxon
Status
Date reported
Alternaria
Alternaria sp.
Fungus
Present: in all parts of the area
16 November 2006
Angelonia flower break virus Canna yellow mottle badnavirus Siam weed Mango Fruit Borer
Angelonia flower break virus
Parsley (Petroselinum crispum) Verbena, Angelonia and Phlox spp. Canna spp.
Virus
Absent: intercepted only
30 April 2009
Virus
Present: at low prevalence
17 May 2007
Weed Insect
Present: but managed Present: only in some areas
11 November 2013 3 December 2009
Weed Mite Insect Fungus Fungus Virus
Present: under eradication Present: only in some areas Absent: pest eradicated Present: under eradication Present: only in some areas Present at low prevalence
25 23 21 21 10 30
Fungus
Present: only in some areas
28 October 2011
Koster's curse New Hebrides coconut mite Cocoa pod borer Chestnut blight Citrus leaf spot Dracaena mottle badnavirus Black scab of jojoba
Canna yellow mottle badnavirus Eupatorium odoratum Citripestis eutraphera
Clidemia hirta Colomerus novahebridensis Conomorpha crameralla Cryphonectria parasitica Cryptosporiopsis citri Dracaena mottle badnavirus
Mango (Mangifera indica), cashew (Anacardium occidentale)
Mango Malformation Disorder Garlic allexivirus Potato Cyst Nematode
Fusarium mangiferae
Coconut (Cocos nucifera) Cocoa (Theobroma cacao) Chestnut (Castanea sativa) Citrus spp. Happy Plant (Dracaena spp.) Jojoba (Simmondsia California) Mistflower (Ageratina riparia) Mango (Mangifera indica)
Garlic allexiviruses Globodera rostochiensis
Garlic (Allium sativum) Potato (Solanum tuberosum)
Virus Nematode
Willow black canker Tobacco Slug Beetle
Physalospora miyabeana Hemileia wrightiae Lema bilineata
Willow (Salix spp.) Wrightia pubescens Prairie Ground Cherry, evaluated as a biocontrol agent for jimsonweed
Fungus Fungus Insect
Yellow Burrhead Vegetable Leaf Miner
Limnocharis flava Liriomyza sativae
Rice blast
Magnaporthe grisea
Miconia vine
Mikania vine Narcissus Late Season Yellows Virus Teak leaf rust
Miconia calvescens, M. nervosa, M. racemosa Mikania micranthia Narcissus Late Season Yellows Virus Olivea tectonae
Panicum mosaic virus
Panicum mosaic virus
Grapevine leaf rust Branched broomrape
Solenopsis Mealybug
Phakopsora euvitis Phelipanche ramose/P. mutelii (formerly Orabanche ramosa/O. mutelii) Phenacoccus solenopsis
Banana freckle
Phyllosticta cavendishii
Phytophthora elongata Impatiens downy mildew
Phytophthora elongata Plasmopara obducens
Mango leaf gall midge Asian strain of bacterial canker of kiwi fruit Spider mite Red imported fire ant
Procontarinia pustulata Pseudomonas syringae pv. actinidiae Schizotetranychus baltazari Solenopsis invicta Striga asiatica
White smut fungus
Elsinoe australis (new pathotype) Entyloma ageratinae
Aphid Tomato Torrado Virus
Takecallis arundinariae Tomato torrado virus
Myrtle rust Sugarcane smut
Uredo rangelii Ustilago scitaminea Valsa malicola
February 2010 April 2009 January 2014 September 2010 October 2012 April 2009
Fungus
Present: only in some areas
25 November 2010
Fungus
Present: under eradication Present: only in some areas Absent from WA: Eradicated; Present Victoria: Under official control Present: only in some areas Present: only in some areas Present: only in some areas
22 February 2008; 14 January 2010 17 January 2012 11 November 2008; 14 January 2010
Weed
Present: Present: control Present: control Present:
Daffodil (Narcissus sp.)
Weed Virus
Present: only in some areas Present: only in some areas
5 March 2010 11 January 2009
Teak (Tectona grandis)
Fungus
31 October 2006
Buffalo grass (Stenotaphrum secundatum cv. Palmetto) Vitis spp.
Virus Fungus Weed
Present: in all parts of the area where host crop is grown Present: only in some areas; Present: at low prevalence Absent: Eradicated Present: under official control
11 March 2008 25 February 2010
Insect
Present: only in some areas
18 March 2010
Fungus
4 October 2013
Oomycete Oomycete
Present: only in some areas; under eradication Present: only in some areas Present: only in some areas
25 November 2010 17 May 2007
Insect Bacteria
Present: only in some areas Present: only in some areas
11 November 2008 12 July 2011
Insect Insect Weed
Present: Present: Present: control Present: Present:
only in some areas under eradication subject to official
26 July 2012 24 February 2010 22 July 2013
only in some areas only in some areas
23 April 2009 10 October 2008
Solanaceae, Cucurbitaceae, Fabaceae, Asteraceae Rice (Oryza sativa)
Cotton (Gossypium hirsutum) Banana (Musa sp.) Cavendish sub group Eucalyptus spp. Impatiens spp. Mango (Mangifera indica) Kiwi fruit (Actinida chinensis) Citrus spp. Sugarcane (Saccharum officinarum) Bambusa spp. Tomato (Solanum lycopersicum) Species in Myrtaceae Sugarcane (Saccharum officinarum) Apple (Malus domestica)
Aquatic weed Insect
16 November 2006 25 November 2010 23 April 2009
Fungus
Insect Virus Fungus Fungus Fungus
under eradication subject to official
5 March 2010 24 April 2017
subject to official
22 August 2011
only in some areas
5 March 2010
Present: only in some areas Present: established in all parts where host is grown Unknown
6 July 2010
5 May 2010 13 Nov 2006 1 February 2011
C. Anderson et al. / Crop Protection 100 (2017) 8e20
11
Table 1 (continued ) Common name
Scientific name
Invasive ant High Plains Virus Citrus canker
Wasmannia auropunctata Wheat high plains virus Xanthomonas citri subsp. citri Xanthomonas fragariae
Angular leaf spot Mahogany Angular Leaf Spot Yucca bacilliform badnavirus
Xanthomonas sp. Yucca bacilliform badnavirus
Host Commodity
Taxon
Status
Date reported
Wheat (Triticum aestivum) Citrus spp.
Insect Fungus Bacteria
Present: under eradication Present: only in some areas Absent: eradicated
11 January 2012 21 December 2012 23 January 2009
Bacteria
Absent: eradicated
30 November 2012
Bacteria
Present: only in some areas
5 February 2009
Virus
Present: at low prevalence
17 May 2007
Strawberry (Fragaria x ananassa) African mahogany tree (Khaya senegalensis) Yucca elephantipes
health concerns to residents (Dominiak et al., 2013). Red imported fire ants (Solenopsis invicta Burren) would adversely impact Australia's outdoor lifestyle and surveillance is conducted for early detection (Dominiak et al., 2010). Red imported fire ants and yellow crazy ants (Anoplolepis gracilipes (Smith)) impact the environment removing a range of biota: the latter was detected and eradicated in northern New South Wales (Dominiak et al., 2010, 2011a). Recent incursions of amenity tree pests could reduce tree canopies and this is undesirable in Australia where skin cancer and sun exposure are health concerns (Dominiak et al., 2008; Lefoe et al., 2014). 4. How do we do surveillance? Data to support claims of pest status e.g. that a pest is “known not to occur” can be gathered through general surveillance (also called passive surveillance) and specific surveys (also called active or targeted surveillance) (IPPC, 1998; van Halteren, 2000). General surveillance is defined as a process whereby information on particular pests that are of concern for an area is gathered from many sources (IPPC, 1997). Specific surveys are defined as procedures by which National Plant Protection Organisations (NPPOs) obtain information on pests of concern on specific sites in an area over a defined period of time (IPPC, 1997). Specific surveys can be expensive due to the costs of sending specially trained teams into the field, but can potentially provide more reliable and comprehensive pest records and therefore more precise quantitative estimates of confidence in pest status compared to general surveillance. However, general surveillance can be a very powerful tool for the reporting of new pest incursions, especially where budgetary limitations restrict the level of specific surveys conducted. For example, the majority of new pest reports in New Zealand occur through general surveillance rather than through the deployment of specific surveys (Wilson et al., 2004). In most cases, the capacity of a surveillance system to detect pests will improve when both strategies are deployed at all stages of the plant biosecurity surveillance cycle. Surveillance is often done to a standard. Fruit flies are a major economic pest in Australia and the world. Fruit flies are amongst the easiest of insects to monitor due to their attraction to lures (Suckling et al., 2016). Queensland fruit fly is not detected in Western Australia but that state maintains trapping to demonstrate continuing absence (Dominiak and Daniels, 2012). Similarly decades of fruit fly monitoring have demonstrated that Bactrocera neohumeralis (Hardy) is not detected south of Sydney (Dominiak and Worsley, 2016). On the margins of insect range, populations decline until permanent populations are difficult to maintain. Standards such as areas of low pest prevalence can be clearly declined for species such as Queensland fruit fly (Dominiak et al., 2015). These standards may be extrapolated to other species such as B. neohumeralis (Dominiak and Worsley, 2016). In areas where populations are eradicated, the pest status can be demonstrated by
regular surveillance (Dominiak and Daniels, 2012) or by statistical analysis of the likelihood of eradication (Dominiak et al., 2011a). 5. General surveillance? General surveillance includes: (1) data mining and analysis of information about specific pests from multiple sources; and (2) engagement with stakeholders to increase awareness of target pests and thereby encourage widespread reporting through official channels. Sources of information for data mining include NPPOs, research institutions, scientific publications, consultants, and industry bodies. Data collected may include the results of specific surveys, as well as the records of diagnostic laboratories, scientific collections, and the results of pest research. Examples of Australian general surveillance tools that use these sources of information together include: the National Plant Surveillance Reporting Tool (NPSRT) which collates records of government and industry funded surveys across a range of commodities (Plant Health Australia, 2011a); the Australian Plant Pest Database (APPD) which provides online access to specimen records in 14 scientific collections (Plant Health Australia, 2011b); Australia's Virtual Herbarium which provides a central point for access to plant records in each Australian state (AVH, 2011); and the Biosecurity Surveillance Incident Response and Tracing (BioSIRT) tool which is a national system for collating and analysing results of specific surveys completed in response to plant pest incursions. It has been used by some states but not all. Some states already maintain their own electronic databases that can quickly be mined for trade purposes, particularly to demonstrate pest freedom (Dominiak et al., 2007). Other states keep paper records or multiple spreadsheets and these are difficult to quickly data mine. Another form of data mining and analysis that is not commonly used in the plant world is syndromic surveillance through which real-time data sets like web searches and sales figures of chemicals may be interrogated. These data could reveal patterns to provide early warning of pest and disease outbreaks. Such a technique may be useful in the plant world to predict for example patterns of fruit fly distribution in fruit fly free regions or states (Dominiak and Daniels, 2012). An important aspect of general surveillance is the fostering of “stakeholder” or “grass-roots” surveillance, through education and awareness campaigns encouraging the public to report potential pest detections. A good example of this is the Northern Australian
12
C. Anderson et al. / Crop Protection 100 (2017) 8e20
Quarantine Strategy (NAQS) that specifically targets local communities to foster awareness and reporting of biosecurity threats. The use of “engaged” stakeholders was captured in a recent surveillance design targeting exemplar species entering Barrow Island in north western Western Australia (Jarrad et al., 2010). The design (Barrett et al., 2009) accommodated both general surveillance corresponding to potential detection by workers, and targeted surveillance via a variety of methods including trapping and observations in situ. Workers on the island were divided into two categories: engaged workers that were considered “active” since they were sufficiently motivated to attend training sessions, contrasting with other workers considered “passive”. Another recent case where “engaged” stakeholders were utilised to support surveillance activities occurred during the incursion of myrtle rust (Puccinia psidii Winter) in NSW in 2010 (Carnegie and Cooper, 2011). Stakeholders including the nursery and garden industry, state and local government, the Australian tea tree industry and bush regenerators were engaged through a series of formal and informal meetings and training sessions, media releases, weekly email updates, the provision of extension material including leaflets and posters and a website that was regularly updated with information on the operation. Emails, media releases, extension material and the website were all designed to increase awareness of the disease and how to report it. Reporting mechanisms included a biosecurity email address and the national Exotic Plant Pest (EPP) hotline. These channels resulted in several reports of myrtle rust being received through the email address and hotline on a weekly basis and assisted in the targeting of delimiting surveys: the results of which were logged in BioSIRT. In response to the continued public interest, an online reporting tool is under development by the Atlas of Living Australia (ALA) which will enable the public to log reports of any pest on an interactive map (ALA, 2011). This system will also record the general level of awareness of the reporter so as to attribute confidence levels to each pest report. The data generated through the ALA mapping project will be available to the public, and will be utilised by a range of stakeholders, from government to private individuals, to manage any incursion into the future. In addition to specialised surveillance scenarios like the myrtle rust response and the Multiple Plant Pest Surveillance Program, several national plant biosecurity awareness programs in Australia promote public reporting through the EPP hotline. This is a national free-call telephone number that connects the public directly to relevant state plant protection organisations. Both Plant Health Australia's “Spotted anything unusual” campaign, and the Quarantine Domestic program (Quarantine Domestic, 2011) promote the EPP Hotline as the primary point of contact. This simplifies the reporting process for the public. The hotline creates a single national tool through which general surveillance data on plant pests is captured. A similar national hotline operates in New Zealand and also takes reports of environmental and animal pests and diseases (Froud and Bullians, 2010). Hotlines may reduce the face-to-face contact between government and stakeholders traditionally done by face-to-face contacts or farm visits (Kruger et al., 2009). Hotlines should always form part of an integrated engagement program which includes face to face options for increasing stakeholder awareness since the accuracy and reliability of reports to hotlines increases with increased awareness among callers. 6. Specific surveys Specific surveys, sometimes referred to as active or targeted surveillance, are official procedures by which NPPOs obtain information on pests of concern on specific sites in an area over a defined period of time (IPPC, 1998). Specific surveys include
surveillance for target pests as well as commodity or host surveillance (IPPC, 1998). There are three types of specific surveys: detection, delimiting and monitoring (IPPC, 1998). These form the backbone of the plant biosecurity surveillance cycle, are complimentary, and usually interdependent. One form of surveillance may lead to another. When deployed in conjunction with general surveillance, they improve the precision of the surveillance system. A detection survey is a survey conducted in an area to determine if pests are present (IPPC, 1998) and is implemented: (1) when pest status needs to be determined or verified for trade purposes e.g. establishing or maintaining a pest free area, area of low pest prevalence, pest free place of production or pest free production site; (2) for early detection of emergency plant pest incursions; and (3) for compilation of pest and host lists (Evans and McMaugh, 2005). Detection surveys and general surveillance are often deployed simultaneously to improve the probability of pest detection and both may lead to delimiting surveys. Delimiting surveys are conducted in the initial stages of an emergency plant pest incursion to establish the boundaries of the infested and free areas (IPPC, 1998) and to inform decisions about the feasibility of eradication. Monitoring surveys are ongoing surveys that go beyond simply recording presence or absence. Surveys may verify the characteristics of a pest population and to provide ongoing validation of pest status in an area, for example during and following completion of an eradication program, or to establish an area of low pest prevalence for trade purposes (IPPC, 1998; Evans and McMaugh, 2005). There are many methodologies and designs employed in the development of specific surveillance strategies that will not be dealt with here (Low-Choy et al., 2011). More generally, specific surveys rely on careful stratification of an area according to the associated risk of pest entry, establishment and spread, and the consequences of incursion; a process collectively referred to as pest risk analysis (PRA) (Anon, 2007). PRA considers analysis of trade patterns along with pest biology and epidemiology (IPPC, 2004). For example, the Multiple Plant Pest Surveillance Program targets early detection surveys for exotic pests in areas that have been assessed through this process. Following the introduction of the Emergency Plant Pest Response Deed (EPPRD) in Australia in 2005, Plant Health Australia has worked with industries to develop high priority plant pest threat lists through the PRA process, which assist targeting of surveillance and preparedness activities (Plant Health Australia, 2010). Case studies provide specific examples of surveillance and preparedness activities currently in place in Australia as a result of the industry biosecurity planning process. PRA is also the basis for commodity import risk analysis (IRA) that is used to set policy and regulate international and domestic trade (Australian Government Department of Agriculture, Fisheries and Forestry, 2009). 7. The plant biosecurity surveillance cycle The IPPC (1998) describes pest status in an area as unknown, present, absent or transient. Each of these states has several variations dependent upon the level of pest management in place and the availability of reliable records (IPPC, 1998). Pest status in an area could be considered unknown when: (1) there is a lack of accurate and reliable records for the pest e.g. existing records do not have sufficient coverage or are otherwise not sufficiently representative across the area of interest; and (2) where there are reasons to consider that pest status may have changed (IPPC, 1998). Changes in pest status may occur for various reasons including climate
C. Anderson et al. / Crop Protection 100 (2017) 8e20
change, changes in crop management and diversity, new trade patterns that open new pathways for pest entry, or an undetected incursion. When the pest status of an area is unknown, it can be resolved through a cycle of complementary and often interdependent surveillance activities consisting of both general surveillance and specific surveys. We call this process the Plant Biosecurity Surveillance Cycle (Fig. 1). The extent to which phases of the Plant Biosecurity Surveillance Cycle are engaged is determined by the extent to which the target pest has or is likely to have penetrated the biosecurity continuum and the magnitude of risk associated with the target pest. For example, where there is no reason to suspect a pest has ever been detected post-border, a determination to that effect will be based on the results of “detection” activities which may include data mining and the results of specific surveys deployed at ports of entry (Gillespie, 2003). In such a case there would be no need for “delimiting” or “monitoring” (the lower two-thirds of the cycle, Fig. 1). However when a high-risk pest such as Asiatic citrus canker is detected post-border, then all phases of the cycle will apply (Fig. 1). Here we consider the post-border detection of a high risk pest (Fig. 1), which may be detected first through general surveillance, including reports by the public to plant health organisations, or through specific surveys designed to detect the target around high risk points of entry (IPPC, 1998). Often, activities associated with “detecting” may lead directly to a determination of pest absence through repetition of these activities e.g. annual early detection surveys. Alternatively, following detection, the surveillance cycle progresses to the “delimiting” phase (Fig. 1), with specific surveys designed to determine pest prevalence in an area and structured, in part, according to reports gathered through general surveillance. Based on data gathered through delimiting surveys, along with expert advice, managers may determine that it is feasible to proceed with eradication of the pest, or alternatively that eradication is not feasible. Surveillance during eradication often involves a combination of further delimiting and monitoring surveillance to ascertain further
Pest Status: Unknown
DetecƟng
DelimiƟng
N
Feasible to Eradicate?
Y Monitoring
Pest Status: Present Transient Absent Fig. 1. Plant biosecurity surveillance cycle.
13
spread of the pest and monitor eradication/control efforts respectively (Fig. 1). Following an eradication campaign, or if eradication is not feasible, delimiting surveillance will cease and activities will shift wholly to monitoring. The pest status of an area posteradication is likely to be accepted by trading partners only after monitoring for an agreed period. This period of ongoing monitoring is of particular importance when attempting to demonstrate area freedom following an incursion and eradication program (Collins and Vanstone, 2010). If a pest status of ‘absent’ is confirmed by monitoring, the cycle will revert to detection in order to ensure preparedness for future incursions. Where pest status is determined to be ‘present’ or ‘transient’, monitoring may go on indefinitely. Thus, detection, delimiting, and monitoring are complimentary and interdependent components of the plant biosecurity surveillance cycle. When considered together, the components provide robust cumulative data to support determinations of pest status across the biosecurity continuum. 8. Case studies: successful surveillance systems in action The success of a plant biosecurity surveillance system is measured by its capacity to detect pests to a required standard. This standard is often set to reflect a nation's appropriate level of protection (ALOP), but may also be defined by the circumstances at hand, for example a higher standard may be adopted during an eradication program. The efficacy of a surveillance system is directly proportional to its sensitivity, i.e. the probability of detecting the pest, given it is present. System sensitivity is largely influenced by the underlying conceptual and quantitative models (Low-Choy et al., 2011). However, sensitivity is also influenced by three key non-statistical considerations: (1) availability of resources; (2) accurate diagnostics; and (3) levels of stakeholder engagement. Resource availability, including availability of equipment and trained personnel, directly impacts on sensitivity. For example, where a trapping grid is incomplete or inappropriately spaced, or where staff are not competent to identify disease symptoms, the sensitivity of the system will be reduced. This will also be the case where diagnostic labs are ill-equipped, and where diagnosticians do not have access to current techniques or to comprehensive reference collections. Incorrect or inconclusive pest identification can lead to unnecessary expenditure when indigenous species are identified as exotic and vice versa (Holmes et al., 2008). This should be avoided through adequate resourcing of diagnostic laboratories and scientific reference collections, and investment in ongoing training for diagnosticians and taxonomists. The goal of stakeholder engagement should be to foster pest awareness and preparedness so as to enhance general surveillance at all stages of the plant biosecurity surveillance cycle. A two-way flow of information that encourages collaboration, ownership and feedback is always preferable to top-down flow of information without opportunity for interaction (Kruger et al., 2009; Thompson et al., 2009). Engaged stakeholders are active participants in general surveillance and thereby improve overall sensitivity of the system. Feedback mechanisms are also important when it comes to timely reporting of results. Delays in reporting can lead to delays in the implementation of pest management responses and can put at risk the pest status of trading partners by allowing potentially infested produce to move across borders without adequate treatment. The longer a pest remains undetected, the more the chance of successful eradication decreases and the cost to government, industry and the environment increases (Magarey et al., 2009). Thus the design and implementation of successful surveillance systems is of utmost importance in managing the risk of plant pests to the environment and the economy worldwide.
14
C. Anderson et al. / Crop Protection 100 (2017) 8e20
Plant biosecurity surveillance systems in the Australian context are unique as the target areas are vast, the population widely distributed (similar to North America), and the temperatures can range from sub tropical to sub Arctic. Regardless, the underlying reasons for doing surveillance and the principles of successful surveillance design are universal. The following are examples of successful surveillance systems deployed for different purposes at different points along the biosecurity continuum in the Australian context. 9. Case study. Pre border - surveillance North of Australia For over 20 years, plant health scientists from Australia and Papua New Guinea (PNG) have conducted regular joint surveys in parts of PNG close to the Indonesian border. This surveillance aims to provide early warning of new incursions of plant pests of significance to both nations. In recent years, these have been collaborations between PNG's National Agriculture Quarantine and Inspection Authority (NAQIA) and the Australian federal government's Department of Agriculture and Water Resources. Early in this period, a number of similar collaborative surveys were conducted by the same Australian scientists and counterparts from Indonesia's quarantine service in the Indonesian province of Papua, which occupies the eastern half of the island of New Guinea. The surveys involve large teams, comprising counterparts from each country in each of the disciplines of botany, entomology, and plant pathology, and local extension staff. For periods of 2e4 weeks per survey, these teams travel great distances through sometimes highly inhospitable and remote terrain by road, light aircraft and boat. Scientists visit and examine as many different locations and crop plants of economic importance as possible, and occasionally also other plants. Specimens are collected as necessary. Most recently, ‘smart phone’ technology has been employed using GPS photo-tagging software to record field data. Samples and specimens are processed in the field for long term storage, treated to minimise biosecurity risk and returned to Australia under suitable import permit for laboratory analysis. Recent regional movements of certain pathogens of banana and citrus are particularly significant. These commodities are major Australian production industries, each valued at around $500 million in 2010-11 (National Plant Biosecurity Status Report, 2012). They are also crops vital to subsistence production and local commerce in PNG (Gibson, 2000). Moreover, the island of New Guinea is part of the centre of origin of both banana and citrus implying significant threat to germplasm resources (Simmonds, 1962; Swingle, 1967). The surveys of Papua Province in the late 1990s led to the first records on the New Guinea land mass of two key wilt pathogens of banana known to be spreading to the west through Indonesia. These were ‘tropical’ race 4 of Fusarium oxysporum Schlecht. f.sp. cubense (E.F. Smith) Snyder and Hansen (Davis et al., 2000a) and the blood disease bacterium (Ralstonia haywardii subspecies celebensis) (Davis et al., 2000b). These discoveries threaten the banana industries of both PNG and adjacent far north Queensland (Davis et al., 2001). Investigation of any wiltelike problems of banana in PNG therefore became a high priority activity. Surveys were done by destructive internal examination of banana plants showing yellowing and leaf death. Over several years, a widespread and devastating lethal wilt of mostly cooking banana plants was investigated and no symptoms of fusarium or bacterial wilt disease were seen. Instead, a set of characteristic internal symptoms were consistently found that were not those of any known pathogen or disorder. The cause remained unknown for about a decade. Subsequently, a research interest in phytoplasmas associated with monocotyledonous plants developed following an outbreak of a
new lethal wilt disease of coconuts in northern PNG (Kelly et al., 2011). Davis et al. (2012) built up a body of evidence that implicated the phytoplasmas closely related to the new coconut phytoplasma as the causal agent. This would be a unique and completely new wilt disease of banana worldwide if pathogenicity can be confirmed in inoculation tests. Early surveys in Papua Province also focused on one of the worst diseases of citrus trees, huanglongbing disease (HLB), caused by the phloem-limited bacterium, ‘Candidatus Liberibacter asiaticus’. This was because the vector, Asian citrus psyllid Diaphorina citri (Kuwayama, 1908), had been found on the island in 1990 (Aubert, 1990) with later dispersal monitored by the joint surveys. Elsewhere in the world, vector establishment in new locations has preceded disease outbreaks by a few years and HLB was indeed confirmed in Papua, near the PNG border in 1999 (Davis et al., 2000c). Subsequent surveys in PNG led to records of both insect and disease in the border town of Vanimo in 2002 (Weinert et al., 2004). A campaign of quarantine containment followed and delimiting surveys have documented little movement of D. citri and HLB (unpublished data, 2015). 10. Case study. Border e Northern Australia quarantine strategy Border surveillance in Australia is delivered by an array of different government systems, particularly at ports of entry. This case study features a surveillance program that is particularly comprehensive and unusual. Established in 1989, the Australian government's Northern Australia Quarantine Strategy (NAQS) is designed to address some of the unique quarantine and biosecurity challenges facing Australia's northern coastline, from Broome in Western Australia to Cairns in Queensland. The key driver behind NAQS is Australia's close proximity to Indonesia, Timor Leste and especially Papua New Guinea (PNG), the coast of which at its closest point is less than 5 km from Australia's Saibai Island. These northern neighbours have many pests, diseases and weeds not present in Australia and there is a very real risk of movement south, by natural or human-mediated means. Early detection of any such incursions would facilitate eradication before their establishment and resulting negative impact on Australia's natural environment or plant-based industries. However, the extensive coastline, low population density and relative scarcity of commercial plantings pose challenges to effective surveillance outcomes. As a result, NAQS has established a strategy that combines surveillance conducted by experienced plant scientists. The program has a strong public awareness program, encouraging local people and communities to recognise and report unusual problems. A cornerstone of this strategy is the ongoing cooperation and goodwill of the northern indigenous people. This is achieved through their active involvement in NAQS survey activities at several levels, from facilitating access permissions to local ranger groups assisting on the ground. Target lists of plant diseases, insect pests and weeds are developed through assessment of the risks, pathways and hosts. These form a focus for the survey of domestic plantings of likely host plants, community farms and nurseries. NAQS scientists collect and analyse specimens after processing them to ensure no biosecurity risk. This ongoing specimen-based record of plant health status can be used to confirm presence or absence of targeted threats. A typical example of such surveillance in action is outlined below from the Torres Strait: 44,000 km2 of ocean that separates Australia from PNG, with numerous islands, 17 major settlements, and an indigenous population of around 7000 people. The risk of pest and disease movement is actively increased by monsoon weather patterns and traditional movements of people and trading
C. Anderson et al. / Crop Protection 100 (2017) 8e20
of goods permitted under international treaty. These islands have a clear history of acting as natural stepping stones for entry of pests from the north. Examples include spiralling white fly (Aleurodicus disperses Russell 1965) (Grimshaw et al., 2006) and black Sigatoka disease (Mycosphaerella fijiensis Morelet 1963) of banana (Hayden et al., 2003). In August 2008, a NAQS survey team comprising a botanist, an entomologist and a plant pathologist spent a day on the small Torres Strait island of Warraber. The population is approximately 230 and most families on the island are accustomed to annual NAQS surveys. The team was examining backyards, asking permission at each household to examine food and other plants. Upon entering one property, the entomologist's eyes were drawn to a tomato plant showing dramatic signs of leaf mining damage: something not normally found on NAQS surveys. Closer observation revealed Agromyzid flies associated with the affected foliage. Samples of the flies were collected in vials of alcohol and mined leaves were left with Warraber's resident quarantine officer for carefully controlled rearing to facilitate emergence of adult confirmatory specimens. Later expert examination identified the insect as the vegetable leaf miner, Liriomyza sativae (Blanchard), previously only known to occur in distant parts of south-east Asia. Because of the potential risk this insect poses to a broad range of Australian vegetable industries, a chain of notifications and response actions were initiated by both the federal and Queensland governments. This example clearly demonstrates the value of trained and vigilant specialists working with local communities and other government agencies to detect an exotic pest. 11. Case study. Post border e the multiple plant pest surveillance program Australia has strong border quarantine systems and the Australian Department of Agriculture intercepts upwards of 20,000 confirmed exotic pests each year. There will always be a risk of pests entering Australia by both natural and human-mediated means. Thus ongoing post-quarantine border surveillance in urban areas is vital for the rapid detection of pest incursions (Stanaway et al., 2001). To assist with the early detection of plant pests, the Multiple Plant Pest Surveillance Program (MPPS) was initiated by the Australian Office of the Chief Plant Protection Officer and is implemented by state and territories (Wyle et al., 2008). The primary objective of the MPPS Program is to provide early detection of a number of agreed exotic and established relevant pests. In New South Wales the program assesses some pest targets beyond the greater Sydney region, but with the majority within the Sydney basin as it is the major entry point for most passengers and freight entering Australia. The MPPS Program was designed by examining potential incursion pathways for exotic pests to identify the most likely pathways. For most pests, risk pathways include the direct environment in or adjacent to possible entry points like Quarantine Approved Premises (QAPs). Following the pre-border inspections and clearance, QAPs receive recently imported plants or plant products to allow post entry quarantine treatments and inspection. Therefore QAPs may carry some level of risk given the cryptic nature of many plant pests. The existing processes to manage possible risks associated with QAPs are comprehensive; however they provide a documented basis for targeting surveys that go beyond simply looking in the general vicinity of a port of entry. Other possible entry points include areas of high tourist throughput like zoos, botanical gardens and freight hubs like shipping ports and airports. In other cases, potential for unregulated movements of host material into urban environments was considered when selecting surveillance sites. Target exotic pests include tramp ants
15
(e.g. red imported fire ant, yellow crazy ant), aphids (eg. Russian wheat aphid, Diuraphis noxia (Mordvilko)), parasitic bee mites (eg. Varroa jacobsoni (Oudemans)), Asian gypsy moth (Lymantria dispar L.) and Asiatic citrus psyllid (Diaphorina citri (Kuwayama)). Detection methods employed in the MPPS Program are varied and based on the biology and epidemiology of the targeted pest. For example, tramp ants are surveyed by manual collection of samples on hosts or on the ground, particularly near ports. Food lures using fishbased cat food or peanut butter may be used to attract ants to monitoring sites but pesticides are rarely used to kill ants: fruit fly traps use an attractant and a pesticide to capture flies visiting traps (Dominiak et al., 2015). Pheromone traps also are used to target Asian gypsy moth (Dominiak et al., 2013). Diuraphis noxia was detected in southern New South Wales in production areas but not by the MPPS program (New South Wales Department of Primary Industries, 2016). A network of exotic fruit fly pheromone traps across the Sydney basin provide feedback for exotic fruit fly incursions (Gillespie, 2003). This includes a small number of special traps for the Solanum fruit fly (B. latifrons (Hendel)) that are targeted in suburbs where the cultural practices of residents may increase the risk of pest incursion. Apart from traps, visual inspection of suitable host plants accounts for the primary method for many target exotics like Asian citrus psyllid (D. citri) or Asian citrus longicorn beetle (Anoplophora chinensis (Forster)). Importantly the setting of explicit targets for active surveillance targets should be tempered with the inclusion of the inspection of any other targets that ‘look damaged’ to the professional eye. The detection in NSW of two non-target exotic species (Corythuca ciliata (Say) and Chaitophorus sp), albeit of minor or no economic significance (Dominiak et al., 2008), attest to field surveillance skills and taxonomic expertise validating and supporting targeted surveillance activities. The MPPS Program also plays an important role in monitoring the current distribution and host range of a small number of established pests. Under a global warming regime it is predicted that temperature and rainfall changes may potentially alter a pest range, which may impact on pest status between states with implications for trade. Surveys are targeted to areas where pest distributions would likely extend as a result of climate change. For example, the MPPS Program monitors established pests where current distribution is not well understood (e.g. Bactrocera cucumis (French)) and conducts delimiting surveys where there is a need to establish pest status in an area e.g. melon thrips (Thrips palmi Karny), spiralling whitefly (A. dispersus) and silverleaf whitefly biotype B and Q (Bemisia tabaci (Gennadius)). A range of collection methods are employed for established fruit flies, in addition to the trapping methods mentioned previously. For example, visual inspection and collection of adults on hosts, and collection of host fruits for subsequent larval rearing is an important sampling strategy especially for B. cucumis and other rainforest species. Host leaf collection for pest thrips and whitefly targets is standard with identification confirmed by a diagnostic lab. Methodologies follow the principles outlined in McMaugh (2005). The MPPS Program also undertakes general surveillance through the provision of information to the public in a bid to increase quarantine awareness and promote early detection. In the Sydney basin, information brochures about quarantine surveillance are distributed in high risk suburbs including areas adjacent to QAPs. Individuals are encouraged to report suspect detections through the Exotic Plant Pest Hotline. 12. Case study. Post border e Australian cotton disease surveys The methodology described for cotton could be extrapolated to
16
C. Anderson et al. / Crop Protection 100 (2017) 8e20
the full diversity of Australian plant production, both large and small, with modifications as required. Things to take into account are (1) pests and diseases appear in different phases or stages of crop development so monitoring should preferably be conducted during the emergence of main or dominant disease and pest species and (2) a survey of 10% of the crop acreage in a farm or region provides accurate information on the prevalence and development of diseases and pests. In Australia, a number of surveillance programs are supported and conducted by the cotton industry and federal and state governments as part of the surveillance strategies for exotic and endemic pests of cotton. This case study will focus on state surveillance systems that employ general multi-pest surveys to provide an early warning of the incidence and spread of endemic pests, and the incursion of new and exotic pests of cotton. The cotton industry in Australia runs biannual disease surveys conducted by plant pathologists of the New South Wales and Queensland state governments. The objective of these surveys is to monitor the distribution and importance of key endemic pests and record the presence or absence of new or exotic diseases. Seven endemic diseases are routinely assessed. Absence data for several high priority exotic pathogens as identified through the industry biosecurity planning process is also collected and provided to Plant Health Australia for entry into the NPSRT database. Surveys have been conducted in New South Wales since 1984/5 and in Queensland since 2002/3, with surveillance undertaken by experienced plant pathologists. The survey was designed to assess disease among a random sample of representative plants in each field, avoiding edge and row effects that may confound results. Consequently early detection of exotic viral diseases is provided by a separate surveillance system. Fields are selected from a representative set of survey farms chosen from each commercial cotton growing region. Plants are assessed along at least two 200 m diagonal transects in each field, where 10 groups of 10 plants spaced 20 m apart are destructively sampled. Four rows of plants are visually inspected while walking, which when taken over two transects, equates to approximately 4.5 km of crop, or 45,600 plants visually assessed in each field. GPS data is recorded at each survey point to ensure that the same entry point can be revisited. Disease scoring methods vary between early and late season surveys depending upon the disease being assessed. For example, during early surveys, plant stand is assessed in one metre of row, following which 10 plants are dug and assessed for black root rot (Thielaviopsis basicola (Berk & Broome) Ferraris 1912). During late season surveys, plants are first assessed for the presence and severity of Alternaria leaf spot, cotton bunchy top, and the incidence of boll rot caused by various fungi and bacteria. The stem of each plant is then cut and assessed for vascular discolouration, which may be caused by Fusarium, Verticillium or sudden wilt. Discoloured stems are sent to a diagnostic laboratory for confirmation of the presence of endemic, unknown or exotic strains of Fusarium oxysporum f.sp. vasinfectum, (G.F. Atkinson) Snyder & Hansen or Verticillium dahliae Kleb. Above ground symptoms are relatively easy to visually detect and sample, however soil borne disorders are frequently less obvious visually and require more resources to sample (Knox et al., 2006a, 2006b). Similar issues prevail for crops such as potatoes and grapes. Two cotton examples highlighting how routine diagnostics provides early detection of potential disease problems are: 1) the recent identification of Nematospora coryli (Kurtzman 1995), a pathogenic yeast associated with seed and internal boll rot of cotton. In the USA this pathogen has caused up to 60% fibre loss and reduced seed quality. This insect-vectored pathogen is suspected of being present undetected in Australia for at least ninety years prior to its isolation from citrus in 2004 (Shivas
et al., 2005). It was identified for the first time from cotton bolls in 2010 collected during late season surveys. 2) in November 2012 during annual disease surveys, stunted plants were observed in one field. The roots of these plants had numerous swellings. Root samples were collected and reniform nematode (Rotylenchulus reniformis Linford and Oliveira 1940) was confirmed in the roots by nematologist Jenny Cobon (DAF Queensland). This finding established the need for an intensive survey of the region which determined that reniform nematode was prevalent across the district at populations causing up to 30% yield loss. An industry funded research project has commenced to investigate strategies to manage this new issue. This case study demonstrates the usefulness of general multipest surveillance systems in broadacre agriculture in providing (1) an ongoing evaluation of current integrated disease management practices and (2) early detection for a suite of exotic pests and previously unknown pests. 13. Case study. Post border e early detection of cotton leaf curl disease Cotton leaf curl disease (CLCuD) (Begomovirus spp.) causes serious economic losses to cotton production in many regions where it occurs, in particular the Indian subcontinent. For example, CLCuD cost the Pakistan industry an estimated US$5 billion between 1992 and 1997. During the 2009-1010 season losses of up to 100% were observed in many growing regions in India (Rajagopalan et al., 2012). CLCuD is not known to be present in Australia and represents a serious biosecurity risk. The disease was first recorded in Africa in the 1920s, in Pakistan and India during the 1990s and most recently in 2010 in China (Cai et al., 2010). The disease is present in Pakistan, India, Sudan, Egypt, China, Tanzania, Malawi and Nigeria. In Sudan during the 1950s the disease caused yield losses of up to 40% (Tarr, 1957, 1964; Malathi et al., 2003). In preparedness for potential incursions of CLCuD the Australian cotton industry has developed a surveillance system to facilitate early detection of the disease and/or expedite implementation of delimiting surveillance if an incursion occurs. The surveillance system for detecting CLCuD was developed using the endemic Cotton bunchy top disease and Tomato yellow leaf curl virus as model systems. Training staff in the recognition of symptoms of CLCuD and when and how to inspect plants has also been an important component of this research. CLCuD can be detected in cotton crops by visual inspection of plants for disease symptoms. The pattern of disease spread within a crop for diseases caused by insect-transmitted viruses often exhibit an ‘edge effect’ where disease incidence is higher in plants closer to the inoculum source. This is usually a result of the proximity of crop plants to an area of perennial volunteer plants, weeds or other alternative hosts for the pathogens and/or insect vectors, particularly if the crop is located downwind from the inoculum source. This edge effect is normally evident in early infections where disease incidences are low. The recommended method to detect early outbreaks of CLCuD includes concentrated inspections in areas adjacent to potential sources of pathogen inoculum. However, in some instances an inoculum source may not be obvious and survey of all edges of the cotton block should be considered. The surveillance system for detecting CLCuD was developed to facilitate early detection to minimise the potential economic costs to the Australian cotton industry. It was validated in India where CLCuD is endemic and inspection of Australian cotton crops using this system has commenced. The same system will be useful for delimiting surveys during an eradication campaign and to establish pest free status of an area post-eradication or to provide an ongoing
C. Anderson et al. / Crop Protection 100 (2017) 8e20
monitoring system to support disease management if eradication is not feasible. The recommended system for detection of CLCuD is based on using confidence intervals for detecting disease at given incidences as described by Cannon and Roe (1982) and used in the eradication of citrus canker from Queensland from 2004 to 2009 (Gambley et al., 2009). This model suggests if disease is present at an incidence of at least 1%, inspection of 300 individuals will provide a 95% confidence in detecting that disease. This assumes a 100% success in detection of the disease-affected individuals and that individuals are selected randomly for inspection. In practice, a 100% success in disease detection is unlikely, thus the number of individuals inspected is increased to accommodate this. For example, assuming detection success is only 50%, inspection of 600 individuals is required to maintain the same 95% confidence interval. It is also impractical to inspect randomly selected plants. Therefore, the strategy was modified to provide a practical approach to survey cotton crops for CLCuD incorporating both the expected surveillance sensitivity and the expected higher incidence of disease on crop edges. 14. Case study. Post border e emergency surveillance for Asiatic citrus canker Citrus canker is a serious disease of citrus caused by the bacterium Xanthomonas citri subsp. citri. This disease causes major economic losses in southeast Asia where it originated and in many citrus growing countries including the USA (Irey et al., 2006; Gottwald and Irey, 2007), Argentina (Canteros, 2004) and Brazil (Leite and Mohan, 1990). It is absent from Australia but remains a major threat to the Australian citrus industry both from direct crop losses and the potential loss of export markets. In July 2004, citrus canker was detected in the Queensland growing region surrounding the township of Emerald. A Pest Quarantine Area (PQA) and the National Citrus Canker Eradication Program were established to manage containment and eradication (Gambley et al., 2009). The PQA had a radius of about 50 km encompassing the township and all cultivated citrus in the Emerald district. Eradication was accomplished and finally declared in early 2009, after expenditure of $17.6 million and over 200 000 staff hours (Gambley et al., 2009). The cost was well below the net benefit of $70 million to Queensland alone, estimated by the Australian Bureau of Agricultural and Resource Economics. Surveillance was a key component of the containment and eradication program. Surveillance data was used for a range of different decision making processes, several different strategies were used, consistent with International Standards for Phytosanitary Measures. These included delimiting the host destruction area, detection surveys to establish area freedom for growing districts outside the PQA, epidemiology studies, monitoring disease decline and eradication in residual host material, and monitoring for disease freedom on newly planted citrus trees after the eradication period. Delimiting surveys were required at the outset to determine the extent of the incursion in Queensland. A notable learning from the delimiting surveys was that time should be allowed for the disease to develop and spread to a detectable level. Citrus canker was detected on three properties in the Emerald area but it took ten months following the first report of the disease and three rounds of surveillance to confirm its presence on the third property. The disease was probably present on all three properties during the first round of surveillance but at that time the incidence and severity on the second and third properties was below surveillance sensitivity. It wasn't until a period of environmental conditions favourable for rapid disease development and spread occurred that the disease
17
was detected on the subsequent two properties. Delimiting surveillance strategies evolved as the emergency response and eradication program unfolded, covering citrus and putative host species on all properties, public and private, in the entire PQA. Delimiting surveillance was undertaken in growing areas without established links in Queensland and all other states, to underpin area freedom status for these areas. In Queensland, this was designed statistically to give 95% confidence that disease was not present above a 1% incidence and included a total of 457 547 trees. Inspections were done on 329 commercial production properties and 13 citrus nurseries. Epidemiology studies (Gambley et al., 2009) aimed at quantifying the potential inoculum present in the PQA included inspecting all trees surrounding a confirmed positive infection, documenting the disease severity and the location of the disease on those trees. The data from this information provided insight on the amount of citrus canker inoculum present and gave valuable clues to how long the disease was present prior to detection, the extent and direction of its within-block spread, and the factors affecting its spread. The low amount of inoculum detected and the limited spread of the disease on the second and third properties assisted in making decisions pertinent to cost-effective eradication of the disease. For example, the data were used to determine the zone of destruction of native citrus surrounding the infested properties. A second strategy for evaluating inoculum was to survey for potential reservoirs of the bacterium in the absence of living host material. Following tree destruction, a non-host period of 18 months was imposed and infested properties were subjected to multiple rounds of cultivation to enhance the microbial degradation of any remaining citrus material. Towards the end of this period a survey was done of 300 sites, each 0.8 m2 in size, spread across the three infested properties, to inspect for presence of residual citrus material, which if detected, was tested for the bacterium. The survey included inspecting the soil surface at each site and the subsurface to a depth of up to 30 cm. The results from this survey provided confidence that the host-free period was sufficient and there was a negligible risk of emergence of the disease on newly planted hosts. Finally, newly planted citrus had to be monitored to demonstrate success in eradication and to justify a claim for pest area freedom. This involved inspecting young trees in the propagation nursery prior to arrival at the properties, then multiple inspections over time once they were planted. The earliest plantings were monitored for 18 months. The data from all the various surveillance systems used allowed successful initial containment of the disease, its subsequent eradication and provided the statistical basis for establishing pest free area status for the Emerald citrus growing district. Pest Freedom was acknowledged in early 2009. 15. Case study. Post border e maintaining pest free areas As a major producer and importer of plants and plant products, Australia places a high priority on maintaining its ability to export products to pest-sensitive markets. Pest free areas (PFA) are a mechanism for facilitating trade into pest-sensitive markets (IPPC, 1998). They provide assurance to the importing country that a particular pest does not exist in a defined area and that no disinfestation treatment is required prior to market access. PFAs can be small regulated areas, or a state, or an entire country, or a region comprised of several states or countries. Areas of Low Pest Prevalence (ALPP) are an intermediate step between pest free area and endemic areas (IPPC, 2005). ALPP allow low numbers of pests to be present and permit control measures to prevent pest populations reaching populations equivalent to breeding populations
18
C. Anderson et al. / Crop Protection 100 (2017) 8e20
(Dominiak et al., 2015). Endemic areas are deemed to have breeding populations present at all times and all produce must be disinfested before being sent to pest sensitive countries. Fruit flies are commonly considered to be the world's most significant horticultural pest group. Different species are adapted to particular environments and countries. In Australia, the native Qfly is the most economically important fruit fly, particularly for eastern Australia. Qfly is not present in the states of Western Australia and South Australia (Dominiak and Daniels, 2012). By conducting targeted surveillance, these states can demonstrate long periods of absence in which Qfly has not been recorded and therefore easily satisfy PFA requirements. Both states maintain a series of trapping grids to monitor Qfly as a requirement to maintain pest free status. In addition, parts of Victoria, New South Wales and South Australia had formed a PFA named the Fruit Fly Exclusion Zone (FFEZ) (Anon, 1996) which straddled a large area of horticultural production. Within this PFA, surveillance was conducted to demonstrate continuing absence of both Qfly and Medfly (Jessup et al., 2007). Medfly has not been detected for more than 60 years in the FFEZ that is consequently recognized as a Medfly PFA (Dominiak and Daniels, 2012). Similarly Qfly was not established in the FFEZ although on occasions flies entered the area, were detected by the surveillance system and eradicated. Acceptance of the FFEZ as a Qfly PFA required surveillance for one year of demonstrated absence through specific surveys (Anon, 1996). An early detection surveillance program comprised of specific surveys because most trading partners are sensitive to Qfly and Medfly. Specific surveillance was responsible for providing confidence that produce harvested from the FFEZ was not infested with Qfly. Permanently deployed traps were inspected weekly in the summer and fortnightly in the winter with 41 inspections per year. The design of the traps and wick replacement was controlled by the trade agreement (Anon, 1996). Any detections of pest fruit fly in the monitoring grid must be reported to federal and state authorities within 48 h of detection: federal agencies immediately advised importing countries. There were two staged surveillance responses within the trade agreement. The standard early detection surveillance system uses Lynfield traps deployed on a 400 m network in towns and 1000 m array in orchards. When two male flies were detected, a second stage of delimiting surveillance was engaged, whereby the trapping array was intensified. An additional (or supplementary) 15 Lynfield traps and 16 MacPhail traps using a wet food lure were deployed to delimit the incursion and to determine in particular if there was a genuine breeding population present or if the trapped flies were an isolated incursion not associated with a resident population (Meats et al., 2003). A breeding population was a breach of the PFA rules while isolated incursions or transient populations provide no threat to PFA status. An outbreak was declared if five male flies were trapped within 1 km within 14 days and the area within 15 km of the epicenter was declared a quarantine area (Dominiak et al., 2011b; Dominiak, 2012) and the pest free status was suspended until the incursion was eradicated. Similarly if larvae were detected in locally produced fruit or if a gravid female was trapped, an outbreak was declared. All fruit within this quarantine area must then be treated if it was to be sold into fruit fly sensitive markets. Quarantine and eradication measures were deployed and the trapping network utilised to monitor the effectiveness of the eradication (Anon, 1996). Following successful eradication, the intensive trapping array was returned to the standard network. Florec et al. (2013) reported that it was more cost effective for pest free areas to exclude pests compared with the costs of eradication and losses associated with disinfestations and loss of market access based on their pest free status. The engagement of the general public would help support reporting of unusual events via
general surveillance. Various state and national websites advertise the restrictions for movement of produce and through road blocks where vehicles are inspected for potentially infested produce (Dominiak and Coombes, 2010). Community awareness of Qfly is promoted through extension material including pamphlets. Residents of the FFEZ are encouraged to report Qfly detections to state plant protection organisations. Ongoing maintenance of the FFEZ as a pest free area provides access for produce into fruit fly sensitive markets and reduces the cost on industry by negating the need for costly disinfestations. Engaged local communities and a surveillance network is likely to result in early detection. Early detection usually means that pests are contained to a small area and small areas are pivotal to successful eradication (Tobin et al., 2014). Australia was regarded as being a pest free area for B. papayae (Drew and Hancock) (Papaya fruit fly (PFF)). Bactrocera papayae was detected in Cairns (Queensland) in 1995 and all other Australian states regulated susceptible Queensland produce entering their states. These other states were obligated to establish a monitoring grid to demonstrate that the pest was not present in their state (Jessup et al., 2007; Gillespie, 2003) and that their trade restrictions were reasonable. All Australian states successfully demonstrated their pest free status by their targeted surveillance. Queensland could again declare that it was pest free from PFF after a range of control and management techniques were successful (Lloyd et al., 1998) and supported by zero detection in targeted and general surveillance. Until 2000, both Australia and New Zealand were considered a PFA for currant-lettuce aphid (CLA) Nasonovia ribisnigri (Mosley). In 2002, CLA was detected in New Zealand (Stufkens and Teulon, 2003). Australia regulated the movement of New Zealand lettuce products and intensified early detection surveillance in all Australian states to demonstrate its pest free status as a country. CLA was detected in Tasmania (an Australian island state) in 2004 (Hill, 2012). Mainland Australia remained pest free and regulated Tasmanian host produce in an effort to maintain their pest status and maintain their pest free market access. In 2005, CLA was detected in Victoria, a mainland state immediately north of Tasmania. The other mainland states then regulated the movement of susceptible produce from Victoria (and Tasmania) and maintained their CLA free status. There was therefore no limitation to trade between any mainland state except Victoria. By February 2006, CLA was detected in New South Wales and by December 2006, CLA had spread across all other states (Dominiak et al., 2009) and all Australian states lost their pest free CLA status. By 2006, however, improvements in integrated pest management including the introduction of resistant lettuce cultivars had minimised the adverse impacts of the pest. This case study demonstrates how surveillance was implemented to maintain pest freedom status. It also can provide a valuable forewarning of likely incursions and thereby prompt development of management strategies to minimise the impact of a pest before it arrives to maintain pest freedom status and to justify trade restrictions (Dominiak et al., 2009). The demonstration of pest freedom by surveillance also has significant trade advantages. Pest Free Areas remain a cornerstone of trade providing a competitive and financial advantage for those areas or states that can fulfill the requirements of a PFA. Surveillance, both general and targeted, remains pivotal for the maintenance of PFAs. 16. Concluding comments Plant biosecurity surveillance systems collect information about pest status which is used to inform decisions that impact on market access, management of established pests and responses to exotic plant pest incursions. These systems may deploy general surveillance and specific surveys at all stages of the plant biosecurity
C. Anderson et al. / Crop Protection 100 (2017) 8e20
surveillance cycle. Successful surveillance systems are usually based on sound quantitative designs, are adequately resourced, utilise accurate diagnostic services, and involve an element of community and stakeholder engagement to increase the value of general surveillance data. The case studies described in this paper highlight several successful plant biosecurity surveillance systems in Australia that demonstrate these principles in action. Historically, many exotic pests have been detected post-entry and establishment, rather than pre- or at-border, and eradication may not be technically feasible or economically justified. For example, Myrtle rust was probably present in Australia for many months if not some years prior to its detection in cut flower production (Carnegie and Cooper, 2011). Improving the ability of plant biosecurity systems to successfully detect pests before they establish or soon after should therefore be a priority for biosecurity managers across the globe. Detecting and stopping a pest before it is established is far more cost-effective than managing an established pest and reduces the overall impact on production, trade and the environment (Florec et al., 2013). Early detection will optimise the likelihood of eradication as affected areas remain small (Tobin et al., 2014). Acknowledgements Dr Paul Phelong provided mentoring in the early stages of the project. Dr Louise Rossiter reviewed and improved earlier versions of the manuscript. This review did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References ALA, 2011. Atlas of Living Australia. Available at: http://www.ala.org.au/ (Accessed 1 August 2011). Anon, 1996. Code of Practice for Management of Queensland Fruit Fly. Standing Committee on Agriculture and Resource Management. Department of Primary Industries, Canberra. Anon, 2007. Pest Risk Analysis (PRA) Training: Participant Manual. FAO, Rome. Aubert, B., 1990. Integrated activities for the control of huanglongbin-greening and its vector Diaphorina citri Kuwayama in Asia. In: Aubert, B., Tontyaporn, S., Buangsuwoon, D. (Eds.), Proceedings of the 4th International Asia Pacific Conference on Citrus Rehabilitation, pp. 133e144. Chiang Mai, Thailand, 4-10th Feb. 1990. Australian Government Department of Agriculture Fisheries and Forestry, 2009. Import Risk Analysis Handbook 2007. Canberra. AVH, 2011. Australia's Virtual Herbarium. Available at: http://chah.gov.au/avh/ (Accessed 1 August 2011). Barrett, S., Whittle, P., Mengersen, K., Stoklosa, R., 2009. Biosecurity threats: the design of surveillance systems based on power and risk. J. Environ. Ecol. Statistics 17 (4), 503e519. Cai, J.H., Xie, K., Lin, L., Qin, B.X., Chen, B.S., Meng, J.R., Liu, Y.L., 2010. Cotton leaf curl Multan virus newly reported to be associated with cotton leaf curl disease in China. New Dis. Rep. 20, 29. Cannon, R.M., Roe, R.T., 1982. Livestock Disease Surveys: a Field Manual for Veterinarians. Australian Government Publishing Service, Canberra. Canteros, B.I., 2004. Management of citrus canker in Argentina. A review. In: AitOubahou, A., El-Otmani, M. (Eds.), Proceedings of International Society of Citriculture, pp. 448e451. Agadir, Morocco. Carnegie, A.J., Cooper, K., 2011. Emergency response to the incursion of an exotic myrtaceous rust in Australia. Australas. Plant Pathol. 40, 346e359. Collins, S., Vanstone, V., 2010. WA PCN Freedom Claim. Final report for project PT10701. Horticulture Australia, Sydney, Australia. Davis, R.I., Moore, N.Y., Bentley, S., Gunua, T.G., Rahamma, S., 2000a. Further records of Fusarium oxysporum f.sp. cubense from New Guinea. Australas. Plant Pathol. 29, 224. Davis, R.I., Fegan, M., Tjahjono, B., Rahamma, S., 2000b. An outbreak of blood disease of banana in Irian Jaya, Indonesia. Australas. Plant Pathol. 29, 152. Davis, R.I., Jacobson, S.C., Rahamma, S., Gunua, T.G., 2000c. Surveillance for citrus huanglongbing (greening) disease in New Guinea and north Queensland. Australas. Plant Pathol. 29, 226. Davis, R.I., Moore, N.Y., Fegan, M., 2001. Blood disease and Panama disease: two newly introduced and grave threats to banana production on the island of New Guinea. In: Bourke, R.M., Allen, M.G., Salisbury, J.G. (Eds.), Food Security for Papua New Guinea. Proceedings of the Papua New Guinea Food and Nutrition 2000 Conference. PNG University of Technology, pp. 816e821. Lae, 26-30 June
19
2000 Australian Centre for International Agricultural research (ACIAR) Proceedings No. 99, 892p, Canberra. Davis, R.I., Kokoa, P., Jones, L.M., Mackie, J., Constable, F.E., Rodoni, B.C., Gunua, T.G., Rossel, J.B., 2012. A new banana wilt disease associated with phytoplasmas in Papua New Guinea. Australas. Plant Dis. Notes 7, 91e97. Dominiak, B.C., Gillespie, P.S., Tom, G., 2007. Changes in data management and the impediments to change e a case study for fruit fly surveillance. Plant Prot. Q. 22 (3), 95e98. Dominiak, B.C., Gillespie, P.S., Worsley, P., Locker, H., 2008. Survey for sycamore lace bug, Corythucha ciliata (Say) (Hemiptera: Tingidae) in New South Wales during 2007. General Appl. Entomology 37, 27e30. Dominiak, B.C., Links, I.J., Fletcher, M.J., Worsley, P., McDougall, S., 2009. Detection and spread of currant-lettuce aphid Nasonovia ribisnigri (Mosley) (Hemiptera: Aphididae) in New South Wales. General Appl. Entomology 38, 27e30. Dominiak, B.C., Coombes, N., 2010. Review of the impact of the Tristate community awareness program on road travellers e 1999/2000. Plant Prot. Q. 25 (1), 2e8. Dominiak, B.C., Worsley, P., Gillespie, P.S., Loecker, H., Kerr, M., Shattuck, S.O., 2010. Search for red imported fire ants Solenopsis invicta Burren in New South Wales from 2004-2006. Plant Prot. Q. 25 (1), 15e18. Dominiak, B.C., Gott, K., McIver, D., Grant, T., Gillespie, P.S., Worsley, P., Clift, A., Sergeant, E.S.G., 2011a. Scenario tree analysis of zero detections and the eradication of yellow crazy ant (Anoplepis gracilipes (Smith)), in New South Wales, Australia. Plant Prot. Q. 26 (4), 124e129. Dominiak, B.C., Daniels, D., Mapson, R., 2011b. Review of the outbreak threshold of Queensland fruit fly (Bactrocera tryoni Froggatt). Plant Prot. Q. 26 (4), 141e147. Dominiak, B.C., 2012. Review of dispersal, survival, and establishment of Bactrocera tryoni (Diptera: Tephritidae) for quarantine purposes. Ann. Entomological Soc. Am. 105, 434e446. Dominiak, B.C., Daniels, D., 2012. Review of the past and present distribution of mediterranean fruit fly (Ceratitis capitata Weidemann) and Queensland fruit fly Bactrocera tryoni Froggatt) in Australia. Aust. J. Entomology 51, 104e115. Dominiak, B.C., Gillespie, P.S., Subasinghe, R., 2013. Surveillance for Asian gypsy moth (Lymantria dispar asiatica L.) between 2005 and 2012 in New South Wales, Australia. Plant Prot. Q. 28 (1), 12e14. Dominiak, B.C., Wiseman, B., Anderson, C., Walsh, B., McMahon, M., Duthie, R., 2015. Definition of and management for areas of low pest prevalence for Queensland fruit fly Bactrocera tryoni Froggatt. Crop Prot. 72, 41e46. Dominiak, B.C., Worsley, P., 2016. Lesser Queensland fruit fly Bactrocera neohumeralis (Hardy) (Diptera: Tephritidae: Dacinae) not detected in inland New South Wales or South of Sydney. General Appl. Entomology 44, 9e15. Evans, G., McMaugh, T., 2005. Surveying for Plant Pests: ISPM Terminology. Workshop on Sanitary and Phytosanitary (SPS) Awareness. ASEANET, Bogor, Indonesia. Florec, V., Sadler, R.J., White, D., Dominiak, B.C., 2013. Choosing the battles: the economics of area wide pest management for Queensland fruit fly. Food Policy 38, 203e213. Froud, K.J., Bullians, M.S., 2010. Investigation of biosecurity risk organisms for the plant and environment domains in New Zealand for 2008 and 2009. N. Z. Plant Prot. 63, 262e269. Gambley, C.F., Miles, A.K., Ramsden, M., Doogan, V., Thomas, J.E., Parmenter, K., Whittle, P.J.L., 2009. The distribution and spread of citrus canker in Emerald, Australia. Australas. Plant Pathol. 38, 547e557. Gibson, J., 2000. The economic and nutritional importance of household food production in PNG. In: Bourke, R.M., Allen, M.G., Salisbury, J.G. (Eds.), Food Security for Papua New Guinea. Proceedings of the Papua New Guinea Food and Nutrition 2000 Conference. PNG University of Technology, pp. 37e44. Lae, 2630 June 2000 Australian Centre for International Agricultural research (ACIAR) Proceedings No. 99. Gillespie, P., 2003. Observations on fruit flies (Diptera: Tephritidae) in New South Wales. General Appl. Entomology 32, 41e47. Gottwald, T.R., Irey, M., 2007. Post-hurricane analysis of citrus canker II: predictive estimation of disease spread and area potentially impacted by various eradication protocols following catastrophic weather events. Online. Plant Health Prog. http://dx.doi.org/10.1094/PHP-2007-0405-01-RS. Grimshaw, J.F., Waterhouse, B.M., Weinert, M.P., 2006. The value of early detection and internal quarantine boundaries in the management of incursions: some examples in plant protection from northern Australia and Papua New Guinea. In: Pest and Disease Incursions: Risks, Threats and Management in Papua New Guinea. Australian Centre for International Agricultural research (ACIAR), Canberra, p. 199. TV Price p 154-160 Technical Reports No. 62. Hayden, H.L., Carlier, J., Aitken, E.A.B., 2003. Genetic structure of Mycosphaerella fijiensis populations from Australia, Papua New Guinea and the Pacific islands. Plant Pathol. 52, 703e712. Hill, L., 2012. The currant lettuce aphid, Nasonovia ribisnigri arrives in Tasmania: part 1. Vic. Entomol. 42, 29e31. Hill, L., 2013. The common armyworm, Mythima converta (Walker) (Noctuidae: Lepidoptera), a seasonal resident in Tasmania. Plant Prot. Q. 28, 114e119. Holmes, T.P., Bell, K.P., Byrne, B., Wilson, J.S., 2008. Economics aspects of invasive forest pest management. In: Holmes, T.P., Prestemon, J.P., Abt, K.L. (Eds.), The Economics of Forest Disturbancesdwildfires, Storms, and Invasive Species. Springer, Berlin, pp. 381e406. Hulme, P.E., 2009. Trade, transport and trouble: managing invasive species pathways in an era of globalization. J. Appl. Ecol. 46, 10e18. International Plant Protection Convention, 1997. ISPM No. 6: Guidelines for Surveillance. IPPC-FAO, Rome.
20
C. Anderson et al. / Crop Protection 100 (2017) 8e20
International Plant Protection Convention, 1998. ISPM No. 8: Determination of Pest Status in an Area. IPPC-FAO, Rome. International Plant Protection Convention, 2004. ISPM No. 11: Pest Risk Analysis for Quarantine Pests Including Analysis of Environmental Risks and Living Modified Organisms. IPPC-FAO, Rome. International Plant Protection Convention, 2005. ISPM No. 22. Requirements for the Establishment of Areas of Low Pest Prevalence. IPPC-FAO, Rome, p. 7. International Plant Protection Convention, 2010. ISPM No. 5: Glossary of Phytosanitary Terms. IPPC-FAO, Rome. International Plant Protection Convention, 2011. International Plant Protection Convention. Available at: www.ippc.int (Accessed 1 August 2011). Irey, M., Gottwald, T.R., Graham, J.H., Riley, T., Carlton, G., 2006. Post-hurricane Analysis of Citrus Canker Spread and Progress towards the Development of a Predictive Model to Estimate Disease Spread Due to Catastrophic Weather Events. Online. Plant Health Progress. http://dx.doi.org/10.1094/PHP-20060822-01-RS. Jarrad, F., Whittle, P., Barrett, S., Mengersen, K., 2010. Barrow Island's biosecurity: catching the unknown invader. Significance 7 (2), 53e57. Jessup, A.J., Dominiak, B.C., Woods, B., De Lima, C.P.F., Tomkins, A., Smallridge, C., 2007. Area-wide management of fruit flies in Australia. In: Vreysen, M.J.B., Robinson, A.S., Hendrich, J. (Eds.), Area-wide Control of Insect Pests: from Research to Field Implementation. Springer, Dordrecht, The Netherlands, pp. 685e697. Kelly, P.L., Reeder, R., Kokoa, P., Arocha, Y., Nixon, T., Fox, A., 2011. First report of a phytoplasmas identified in coconut palms (Cocos nucifera) with lethal yellowing-like symptoms in Papua New Guinea. New Dis. Rep. 23, 9. Knox, O.G.G., Anderson, C.M.T., Allen, S.T., Nehl, D.B., 2006a. Helicotylenchus dihystera in Australian cotton roots. Australas. Plant Pathol. 35, 287e288. Knox, O.G.G., Anderson, C.M.T., Nehl, D.B., Vadakattu, V.S.R.G., 2006b. Observation of Tylenchorhynchus ewingi in association with cotton soils in Australia. Australas. Plant Dis. Notes 1, 47e48. Kruger, H., Thompson, L., Clarke, R., Stenekes, N., Carr, A., 2009. Engaging in Biosecurity: Gap Analysis. Bureau of Rural Sciences, Canberra, Australia. Lefoe, G., Dominiak, B.C., Worsley, P., Davies, J., 2014. Elm leaf beetles Xanthogaleruca luteola (Muller) dispersal across south eastern Australia (1989-2011). Plant Prot. Q. 29 (2), 61e65. Leite, R.P., Mohan, S.K., 1990. Integrated management of the citrus bacterial canker disease caused by Xanthomonas campestris pv. citri in the State of Parana, Brazil. Crop Prot. 9, 3e7. Low-Choy, S., Whittle, P., Anderson, C.M.T., 2011. Quantitative approaches to designing plant biosecurity surveillance systems. In: McKirdy, S. (Ed.), Biosecurity in Agriculture and the Environment. Canberra, Australia. Lloyd, A., Leach, P., Kopittke, R., 1998. Effects of exposure on chemical content and efficacy of male annihilation blocks used in the eradication of Bactrocera papayae in North Queensland. General Appl. Entomology 28, 1e8. Magarey, R.D., Colunga-Garcia, M., Fieselmann, D.A., 2009. Plant biosecurity in the United States: roles, responsibilities and information needs. BioScience 59, 875e884. Malathi, V.G., Radhakrishnan, G., Varma, A., 2003. In: Lobenstein, G., Thottappilly, G. (Eds.), Cotton. In 'Virus and Virus-like Diseases of Major Crops in Developing Countries'. Kluwer Academic Publishers. McMaugh, T., 2005. Guidelines for Surveillance for Plant Pests in Asia and the Pacific. ACIAR Monograph No. 119, Canberra, Australia. Meats, A.W., Clift, A.D., Robson, M.K., 2003. Incipient founder populations of mediterranean and Queensland fruit flies in Australia: the relation of trap catch to infestation radius and models for quarantine radius. Aust. J. Exp. Agric. 43, 397e406. Nairn, M.E., Allen, P.G., Inglis, A.R., Tanner, C., 1996. Australian Quarantine: a Shared Responsibility. Department of Primary Industries and Energy, Canberra.
National Plant Biosecurity Status Report, 2012. Plant Health Australia, Canberra ACT, Australia. New South Wales Department of Primary Industries, 2016. Russian wheat aphid. Primefact 1489, 3. Pimentel, D., McNair, S., Janecka, J., Wightman, J., Simmonds, C., O'Connell, C., Wong, E., Russel, L., Zern, J., Aquino, T., Tsomondo, T., 2001. Economic and environmental threats of alien plant, animal, and microbe invasions. Agric. Ecosyst. Environ. 84, 1e20. Plant Health Australia, 2010. PLANTPLAN: Australian Emergency Plant Pest Response Plan. Version 2. Plant Health Australia, Canberra, ACT. Plant Health Australia, 2011a. National Plant Surveillance Reporting Tool. Available at. : www.planthealthaustralia.com.au/our_projects/psdb/index.cfm (Accessed 1 August 2011). Plant Health Australia, 2011b. Australian Plant Pest Database. Available at: www. planthealthaustralia.com.au/go/phau/capacity-and-capability/informationsupport-systems/appd (Accessed 1 August 2011). Quarantine Domestic, 2011. Quarantine Domestic. Available at: www. quarantinedomestic.gov.au (Accessed 1 August 2011). Rajagopalan, P., Naik, A., Katturi, P., Kurulekar, M., Kankanallu, R., Anandalakshmi, R., 2012. Dominance of resistance-breaking Cotton leaf curl Burewala virus (CLCuBuV) in northwestern India. Archives Virology 157, 855e868. Simmonds, N.W., 1962. The Evolution of the Bananas. Longman, London. Shivas, R.G., Smith, M.W., Marney, T.S., Newman, T.K., Hammelswang, D.L., Cooke, A.W., Pegg, K.G., Pascoe, I.A., 2005. First record of Nematospora coryli in Australia and its association with dry rot of Citrus. Australas. Plant Pathol. 34, 99e101. Stanaway, M.A., Zalucki, M.P., Gillespie, P.S., Rodriguez, C.M., Maynard, G.V., 2001. Pest risk assessment of insects in sea cargo containers. Aust. J. Entomology 40, 180e192. Stufkens, M.A., Teulon, D.A.J., 2003. Distribution, host range and flight pattern of the Lettuce aphid in New Zealand. N. Z. Plant Prot. 56, 27e32. Suckling, D.M., Kean, J.M., Stringer, L.D., Caceres-Barrios, C., Hendrichs, J., ReyesFlores, J., Dominiak, B.C., 2016. Eradication of Tephritid fruit fly populations: outcomes and prospects. Pest Manag. Sci. 72, 456e465. Swingle, W.T., 1967. The botany of citrus and its wild relatives. In: Reuther, W., Webber, N.J., Batchelor, L.D. (Eds.), Citrus Industry, pp. 190e430. Tarr, S.A.J., 1957. The host range of cotton leaf curl virus in Sudan. Emp. Cotton Grow. Rev. 34, 258e262. Tarr, S.A.J., 1964. Virus Disease of Cotton. Commonwealth Mycological Institute, Kew, Surrey, England. Tobin, P.C., Kean, J.M., Suckling, D.M., McCullough, D.G., Herms, D.A., Stringer, L.D., 2014. Determinants of successful arthropod eradication programs. Biol. Invasions 16, 401e414. Thompson, L., Stenekes, N., Kruger, H., Carr, A., 2009. Engaging in Biosecurity: Literature Review of Community Engagement Approaches. Bureau of Rural Sciences, Canberra, Australia. van Halteren, P., 2000. Diagnostics and national plant protection organisations. OePP/EPPO Bull. 30, 357e359. Weinert, M.P., Jacobson, S.C., Grimshaw, J.F., Bellis, G.A., Stephens, P.M., Gunua, T.G., Kame, M.F., Davis, R.I., 2004. Detection of huanglongbing (citrus greening disease) in Timor-Leste (east timor) and in Papua New Guinea. Australas. Plant Pathol. 33, 135e136. Wilson, J., Stephenson, B.P., Gill, G.S.C., Randall, J.L., Vieglas, C.M.C., 2004. Principles of response to detections of new plant pest species and the effectiveness of surveillance. N. Z. Plant Prot. 57, 156e164. Wyle, F.R., Griffiths, M., King, J., 2008. Development of hazard site surveillance programs for forest invasive species: a case study from Brisbane, Australia. Aust. For. 71, 229e235.