Invasive Species, Databases and Decision Making:
Critical Issues
Geoff Norton
Centre for Biological Information Technology
The University of Queensland
Brisbane, Australia
Introduction:
The use of computer based systems to aid pest management has a history going back to the 1960’s (e.g. Norton and Holling, 1979; Norton and Mumford, 1993). Software has been developed and applied to such topics as understanding pest population dynamics, forecasting pest outbreaks, simulating the effect of control measures on pest levels (including biological control) and providing training and decision support products. However, despite the considerable resources that have been allocated to such activities over the past 40 years or so, the impact of what we now call Information and Communication Technology (ICT) on pest management has been much less than we might expect. This reflects general disillusion with decision support systems in agriculture (McCown, 2002).
The main reasons for this lack of impact can all be viewed in the context of sustainability. In practice, many developments in ICT that have been aimed at improving pest management have not lasted very long. In some cases, particularly where the main purpose is one of pushing forward the technological boundaries by developing new approaches and techniques, we would not expect them all to be continued. A certain degree of failure is inevitable, as an integral part of the research process. However, it is a different situation where the aim of ICT development is primarily to provide practical decision support for pest management. Whether this support is intended to aid decisions made by farmers, advisors, quarantine officers, national park managers or policy makers, if the support is short-lived, it will clearly not have much impact and constitute a waste of resources.
What are the factors that affect the sustainability of decision support systems? There are likely to be many factors, including specific events such as the ICT developer moving on and not being replaced or the closure of the agency providing the decision support service. In many cases, however, lack of sustainability is likely to be related to the benefits and costs associated with the decision support system. If the benefits of a particular decision support tool are low (due to lack of use or other reasons) and/or the costs of maintaining it are high, then the chances of attracting resources by whatever means to update and maintain the system will not be good.
In this paper I take this concept of benefits and costs as the basis for addressing what I regard as critical issues associated with invasive species, databases and decision support. My purpose is to explore some of the ways in which we might produce a more favourable benefit:cost relationship and so increase the sustainability and therefore the longer term impact of ICT decision support for invasive species. I should emphasise that my purpose in this paper is primarily to raise issues that need to be considered in further development of decision support systems for invasive species, rather than prescribing a specific approach. Since all developments in this field are dynamic and subject to many variables that are frequently changing, flexibility of approach is another key factor affecting sustainability.
The paper consists of three sections. The first section provides a very brief review of the main options for dealing with invasive species. Next, information of value for making decisions on these options is considered, together with some examples of ways in which ICT can provide this information. The third and final section considers the role of invasive species databases in providing practical assistance to real-world decision makers, and ways of increasing the efficiency and sustainability of these systems.
Invasive species – management options
A simplified presentation of the options for dealing with invasive species is provided in Fig.1. The options for responding to the risk of invasive species are shown in bold. Let’s look at these options in turn, consider the information need and how ICT tools might help in making this information available.
Fig.1. Options for dealing with invasive species
1. Quarantine:
Invasive species can be introduced deliberately or by accident (Andow, 2003). In both cases, there are a range of quarantine strategies and measures that can be employed to reduce the risk of invasive species entering a region or country. These options may be general measures or targeted quarantine measures, aimed at specific species, associated with specific commodities or pathways.
2. Invasive species surveillance and management
Once an invasive species has entered a country, three options are available (Fig.1) – attempt eradication; attempt to contain its spread; or develop an integrated pest management strategy to keep the invasive species at levels that are acceptable.
The best option for any particular invasion will depend on a number of factors, particularly the time taken to detection and how difficult it is to control or kill the invasive species. Fig.2 shows the type of relationship one might expect between the cost of eradication and the time taken to detection. This time to detection will, in turn, depend on the effort devoted to surveillance and making resource managers aware of potential problems. Decisions on which strategy to adopt will be based implicitly, if not explicitly, on a comparison between the estimated cost and benefits of eradication and the costs and benefits associated with strategies not to eradicate.
Figure 2. Likely relationship between the cost of eradication and the time to detection
2.1 Eradication
The cost of eradicating invasive species is often substantial. For instance, the cost of eradicating the invasive Papaya fruit fly (Bactrocera papaya) from Queensland and Australia was $35 million. It is estimated that the fruit fly was present in Queensland orchards for at least 12 months before it was detected in 1995. Apparently a farmer alerted authorities when he noticed something unusual about fruit fly attack in his orchard (fruit flies were attacking at an earlier stage of fruit development than usual).
Currently in Queensland there is an eradication campaign for another invasive species – the fire ant (Solenopsis invicta): a budget of $175 million has been allocated to this eradication program.
There have been several eradication programs in Japan, including the oriental fruit fly (Bactrocera dorsalis), as reported by Shiga (1996).
2.2 Contain spread
Measures to limit the spread of invasive species will depend on the species concerned. In Australia - “Of the more than 28,000 plant species introduced into Australia, over 2,500 have become naturalised” and “Of those species 'gone bush' in recent decades, 65% came from urban gardens and parks (CRC for Australian Weed Management, 2004). Therefore, in this case, policy and regulation measures restricting the species that plant nurseries are allowed to sell can play an important role in restricting the spread of invasive plants. Domestic quarantine strategies and other measures can contain the spread of other types of invasive species.
2.3 Design IPM strategies
Where a decision has been made not to attempt to eradicate an invasive species, but to live with it, the efficacy of various control measures - chemical, cultural, biological – can be explored, individually and in combination, in searching for suitable and acceptable IPM strategies. Alternatively, more radical options may be considered, such as utilising the invading species, such as a woody weed for fuel or biomass production.
Invasive species – information needs
Having briefly reviewed the main options for dealing with invasive species, let us now look at the information that is needed to make informed decisions on these options and the role that ICT can play in generating and disseminating this information.
Pest risk analysis
Pest risk analysis provides the basis for making informed decisions on the best strategy to adopt for targeted species. Information for making such analyses can be provided from –
· Pest databases – that provide pest lists associated with specific commodities and countries. A number of systems provide pest lists – for example:
o New Zealand provides lists of all the pests that have been recorded on commodities being exported from New Zealand (See Box 1).
o Zhejiang University, China provides an on-line quarantine pest search by host and source (in Chinese) - see Zhejiang University web site
o The Australian Plant Pest Database - see Plant Health Australia (APPD) web site - provides data on the incidence of pests in Australia, providing a basis for risk analysis.
· Bio-climatic models - that predict the likely distribution of an invasive species, based on climatic data. Two examples of tools used for this purpose are –
o GARP Genetic Algorithm for Rule Set Production: see GARP web site
o Climex – see Climex web site
Box 1. Part of the New Zealand Ministry of Agriculture and Forestry web site that provides commodity pest lists – see New Zealand MAF web site
Commodity Pest Lists
Please note - the search function and PDF summaries on this page are password protected.
Request a password or Log in.
Purpose and application
The pest lists provided here are of organisms (mostly pests and diseases) that have been recorded on a selection of plant commodities being exported from New Zealand. For example, the apple pest list contains all organisms that have been recorded on apple fruit being exported. MAF maintains these lists on behalf of the industries involved. The lists are sent to selected importing countries for them to identify any organisms that are considered of quarantine concern and that are actionable if intercepted on arrival.
Invasive species surveillance and management
To monitor for new biological invasions, many countries have systematic surveillance systems, involving baiting, field surveys and other techniques. This data provides important information for deciding on and implementing responses to invasive species. Important functions that ICT can play in this context include –
· Identification or diagnostic support - The role that Lucid and other computer-based key systems can play in providing support for identification is discussed in another paper presented at this workshop (Norton, 2004). Another way of providing support for identification is the PestNET system (PestNET – see web site). Focussing initially on Asia and the South Pacific, plant health officers who detect an unknown species can email the PestNET list-server, attaching an image of the unidentified specimen. Others in the network then provide suggestions on what the identity of the specimen is likely to be.
· Taxonomic databases and species fact sheets - Many taxonomic databases now exist on-line, many being part of the Global Biodiversity Information Facility (GBIF) and many focussing on invasive species, such as the IUCN/SSC Invasive Species Specialist Group (ISSG) Global Invasive Species database. Other databases, such as the Australian Virtual Herbarium (see AVH web site), provide a means of searching for details of voucher specimens in various herbariums.
· Distribution maps and databases - Various GIS and other mapping systems can be used to record and disseminate real-time information about the spread of invasive species. For instance the Australian Virtual Herbarium species distribution mapper is linked to working taxonomic databases.
· Other management tools – include –
o facilities that allow users to notify the system of new invasive species detections, which “alert” users
o an Internet search engine that identifies those nurseries selling “noxious” weeds and warns the operators they are in breach of regulations (Ron Stinner, pers.com.)
Invasive species, databases and sustainability
In the introduction to this paper, I suggested we can increase the value and sustainability of invasive species databases by increasing the benefits of the database and reducing the costs of operating and maintaining it.
Let’s look first at the benefits. Who are the potential stakeholders who contribute to and utilise information contained in the database? The main players are indicated in Fig 2 (see also Andow, 2003).
Figure 2. Key potential stakeholders of an invasive species database
Ideally, the major stakeholders (both contributors and recipients) should be involved in discussions about the objectives and overall design of the system, to increase ownership and the relevance and ease-of-use of the system. To improve the value of an invasive species database, we need to consider such questions as - Who are likely to be the main users? What information would be of most use to them? Is this information easy for them to find and access? Does there need to be summary information in other languages? To what extent can users provide valuable information to the database?
Some of the problems with existing databases include an interface that is not easy for non-specialists to use; for instance, where one has to type in the genus and species name to retrieve information, rather than being able to browse a list or search by common name. Another problem is incomplete entries to the database – if users frequently cannot get the information they need, they are unlikely to continue to use the system.
Let’s turn now to ways of reducing the costs of providing a database service for invasive species. While various technical ways of reducing costs could be explored, the most obvious means is to utilise information that already exists. One of the main objectives of the Global Biodiversity Information Facility (GBIF) is to encourage a federated approach to database development and access. Individual countries and agencies develop taxonomic databases for their own specific use but this database also becomes a portal in a global system. When a user wants to find out information about a specific species, a search is carried out using a retrieval engine like DiGIR (Distributed Generic Information Retrieval – see DiGIR web site) to search existing databases. The Australian Plant Pest Database (Plant Health Australia web site) provides another example of this type of distributed system, involving all the states of Australia.
This same model is now being applied to invasive species. GISIN (Global Invasive Species Information Network) is being launched under the Convention on Biological Diversity (CBD) and the Global Invasive Species Programme (GISP), possibly as a thematic network linked to GBIF. The purpose is to –
• Provide a platform for sharing information on invasive species via the Internet
• Allow the reporting and tracking of new invasions
• Develop and share ICT tools for identification, mapping, and predicting the spread of invasive species.
• Build the capacity of network members in the development of invasive species information tools.