Richard Baker and Alan MacLeod
Pest Risk Analysis Sub-Team, Plant Health Group, Central Science Laboratory, York, YO41 1LZ, United Kingdom; e-mail: (1) email@example.com (2) firstname.lastname@example.org
The international standard for phytosanitary measures ISPM 11 : Pest risk analysis for quarantine pests (and its revisions) provides a clear description of the procedures to be followed in conducting pest risk analyses but, apart from statements such as climatic modelling systems may be used, does not give further guidance on the tools and resources that can support the pest risk analyst. This paper summarizes the principal tools and resources available for the pest risk assessment component of pest risk analysis, giving examples of how they can be used. It also highlights the principal challenges for the future.
Of the international standards for phytosanitary measures, ISPM 11 : Pest risk analysis for quarantine pests (and its subsequent revisions to its current form, ISPM 11 : Pest risk analysis for quarantine pests, including analysis of environmental risks and living modified organisms) describes the procedures to be followed when conducting pest risk analyses, highlighting the key factors that need to be considered. Regional and national PRA schemes based on ISPM 11 have been developed to provide a logical structure for the analyst to follow. Examples are those by the European and Mediterranean Plant Protection Organization (EPPO, 1997 and in prep.) and the United States Department of Agriculture (APHIS, 2000). However, there is little guidance for risk analysts when searching for the data required and the tools and other resources that can ease their task. As a result, the production of pest risk analyses may appear to be daunting.
This paper shows, with examples, how various data sets and tools can be used to assist in the different stages of the pest risk assessment component of pest risk analysis. Although these have been developed primarily for invertebrate and pathogen pests of cultivated plants, they are equally valid for pests of uncultivated plants and plants as pests (Schrader and Unger, 2003; Baker et al., in press).
Much of the data required must come from within national boundaries: from the PRA area itself and, to assess entry potential, from the country of origin. Although general guidance on the sources of national data cannot be given, there are several international data sets, which can be consulted even when national data sets are difficult to obtain or are incomplete. For example, CAB Internationals Crop Protection Compendium (CABI, 2004) provides detailed pest and crop data sheets together with a structured commodity pest risk assessment scheme, which can be used to select species that might be present in a commodity based on the pest distribution and host range in the exporting country. Climatological information is also available (e.g. New, Hulme and Jones, 1999).
Although searching abstracts of the worlds scientific literature will also yield critical information on pest species, books, which cannot be effectively abstracted, should not be neglected. It must be borne in mind that little abstracting occurred prior to 1972. Pest alerts from regional plant protection organizations (NAPPO, 2003 and refer chapter 37; EPPO, 2003) may yield important information for initiating and compiling pest risk analyses. Even Web search engines, such as Google (www.google.com), may be helpful.
Although a variety of information is needed to assess the extent to which a pest will be able to pass through all the stages of the pathway from the origin to the PRA area, data on trade pathways and interceptions (detections in consignments) are most important. Unfortunately, no readily available compilations of these data sets exist and, for most purposes, specific enquiries for unpublished data have to be made. MacLeod and Baker (1998) analysed the European interception data for Thrips palmi, highlighted the role played by orchids from Thailand and showed how detections decreased once action had been taken to prevent pests travelling along this pathway.
Baker (2002) has reviewed the data required and the techniques that can be used both for assessing establishment potential and for predicting the limits to the distribution of quarantine pests once established in a country. Essentially, ecological factors (such as the suitability of the ecological environment, presence of hosts and natural enemies) and factors intrinsic to the pest itself (such as its reproductive strategy and genetic adaptability) should be considered. Even when a pests responses to the abiotic and biotic environment are poorly understood and when little can be obtained from the literature concerning the intrinsic factors, some judgements can still be made.
Fig. 1: Diabrotica virgifera virgifera, Western corn rootworm (adult shown above), is a serious pest of maize in northern United States and Canada. Climate modelling programs helped determine its potential distribution in the United Kingdom.
Reproduced from Crop Protection Compendium, © CAB International
If the area of origin and the host plants are known, climates in the area of origin can be compared with climates in the area under threat and the distribution of host plants determined. The computer program CLIMEX (Sutherst, Maywald and Skarratt, 1999) is particularly useful in predicting potential distribution based on current distribution even when an organisms responses to climate are unknown. Baker, Cannon and MacLeod (2003) showed how CLIMEX models predicting development based on temperature, maps of maize distribution, current climate data and climate change data can be used to predict the potential distribution of Diabrotica virgifera virgifera (figure 1) in the United Kingdom. Computer mapping software, known as a geographic information system (GIS), provides an extremely powerful method for analysing the different data sets and displaying the results.
Assessing the magnitude of the consequences for plants in the PRA area after establishment of a pest ideally requires knowledge of the pests impacts in its current range, sufficient biological data to predict its spread and population dynamics in the PRA area coupled with financial, economic, environmental and social data for the enterprises, ecosystems and people likely to be affected. As for establishment potential, assessments can still be made even if data are lacking, for example by using expert opinion. Morgan and MacLeod (1996) showed how a population model for Bemisia tabaci, a transmission model for Tomato yellow leaf curl virus and a gross margin budget for glasshouse tomatoes in the United Kingdom could be combined to estimate the financial consequence of the introduction of the virus and its vector.
Estimating the environmental consequences of pest introductions is more challenging. Assessing the potential environmental impacts in Europe of Phytophthora ramorum, the pathogen responsible for sudden oak death in California and Oregon, is particularly difficult. Key factors, such as those influencing infection and host damage (e.g. climate), are poorly known (Jones, Sansford and Brasier, 2003). While research is being carried out, key areas for conservation where susceptible hosts grow in climatic conditions most closely matching those in affected areas of the United States can be identified, and maps made and used to help target surveys and generate contingency plans.
There are many useful data sets and tools that can be employed in pest risk analysis. Although some of the examples given may seem complex, it is important to stress that pest risk analyses do not need to be long and detailed. ISPM 11 does not state that a risk analysis must achieve a specific level of detail or quantification for it to be acceptable. A risk analysis can be very short. For example, if a key component for establishment in the PRA area (e.g. a suitable host plant) is absent no further work may be necessary.
Further work needs to be carried out on two fronts. First, good pest risk analysis practice needs to be captured and distilled in some form of manual enabling those new to pest risk analysis to learn the discipline as quickly and easily as possible. Increased publication of pest risk analyses would also help to demonstrate how pest risk analyses should be constructed. However, a reliable mechanism for the publication of pest risk analyses is needed, because many journal editors find the PRA format incompatible with their requirements. Some mechanism to foster links between PRA practitioners worldwide would also be invaluable.
Second, additional resources need to be put into enhancing the science of pest risk analysis. There are many challenging components of pest risk analysis that would benefit from additional scientific input, such as modelling pest spread, increasing the temporal and spatial resolution of risk maps, scaling up from one outbreak to many, modelling how impacts change over time, quantifying environmental impacts and managing uncertainty. The increasing cooperation between plant health pest risk analysts and those analysing the risks posed by invasive alien species to biodiversity is already proving to be highly beneficial.
References and further information
APHIS. 2000. Guidelines for pathway-initiated pest risk assessments. USDA Animal and Plant Health Inspection Service (available at www.aphis.usda.gov).
Baker, R.H.A. 2002. Predicting the limits to the potential distribution of alien crop pests. In G.J. Hallman & C.P. Schwalbe, eds. Invasive arthropods in agriculture. Problems and solutions, pp. 207-241. Enfield, USA, Science Publishers Inc. 450 pp.
Baker, R.H.A., Cannon, R.J.C. & MacLeod, A. 2003. Predicting the potential distribution of alien pests in the UK under global climate change: Diabrotica virgifera virgifera. In Proceedings of the BCPC International Congress - Crop science and technology 2003, pp. 1201-1208. Alton, UK, BCPC Publications. 1236 pp.
Baker, R., Cannon, R., Bartlett, P. & Barker, I. (in press). Novel strategies for assessing and managing the risks posed by invasive alien species to global crop production and biodiversity. Annals of Applied Biology, 146 (in press).
CABI. 2004. Crop Protection Compendium. Wallingford, UK, CAB International (available at www.cabi.org).
EPPO. 1997. Pest risk analysis: pest risk assessment scheme. EPPO standard PM 5/3(1) (available at ftp://server.oepp.eppo.fr).
EPPO. 2003. EPPO alert list. European and Mediterranean Plant Protection Organization (available at www.eppo.org).
Jones, D.R., Sansford, C.E. & Brasier, C.M. 2003. Pest risk analysis. Phytophthora ramorum (available at www.defra.gov.uk).
MacLeod, A. & Baker, R.H.A., 1998. Pest risk analysis to support and strengthen legislative control of a quarantine thrips: the case of Thrips palmi. In Proceedings of the 1998 Brighton Crop Protection Conference - Pests and diseases, I, pp. 199-204. Alton, UK, BCPC Publications. 316 pp.
Morgan, D. & MacLeod, A. 1996. Assessing the economic threat of Bemisia tabaci (Gennadius) and tomato yellow leaf curl virus to the tomato industry in England and Wales. In Proceedings of the 1996 Brighton Crop Protection Conference - Pests and diseases, II, pp. 1077-1082. Alton, UK, BCPC Publications. 1242 pp.
NAPPO. 2003. Phytosanitary alert system. North American Plant Protection Organization (available at www.pestalert.org).
New, M., Hulme, M. & Jones, P.D. 1999. Representing twentieth-century space-time climate variability. Part I: development of a 1961-90 mean monthly terrestrial climatology. Journal of Climate, 12(3): 829-856.
Schrader, G. & Unger, J-G. 2003. Plant quarantine as a measure against invasive alien species: the framework of the International Plant Protection Convention and the plant health regulations of the European Union. Biological Invasions, 5: 357-364.
Sutherst, R.W., Maywald, G. F. & Skarratt, D. B. 1999. CLIMEX for Windows 1.1. Users guide. Australia, CSIRO Publications (information available at www.ento.csiro.au).
Note: All adopted international standards for phytosanitary measures (ISPMs), listed in appendix 1, are available at www.ippc.int.