Enhancing the Science in a Science-Based System
Predicting the threats posed by nonindigenous plants, arthropods, and pathogens to plants in the United States is and will continue to be a substantial challenge. Having considered what is known about the development of invasions and about attempts to use that knowledge in a predictive manner, the committee reached the following four conclusions:
Conclusion 1. The record of a plant’s invasiveness in other geographic areas is currently the most reliable predictor of its ability to establish and invade in the United States. The same is true for arthropods and pathogens if plants that they can use elsewhere occur in the United States.
Reliance on natural history, that is, empirical studies and careful observation of species outside the United States, whether in their native ranges or in other new ranges—will remain a cornerstone of the prediction of the behavior of nonindigenous species before their entry into the United States. This practice is engrained in all current national and international systems for evaluation of the threats posed by nonindigenous species, such as the International Phytosanitary Protection Convention (FAO 1997, Pheloung et al. 1999).
Conclusion 2. There are currently no known broad scientific principles or reliable procedures for identifying the invasive potential of plants, plant pests, or biological control agents in new geographic ranges, but a concep-
tual basis exists for understanding invasions that could be developed into predictive principles.
The committee was not convinced that broadly successful methods for predicting invasiveness have been established. No reliable, all-inclusive, and tested list of potential invaders has been developed for plants, arthropods, or plant pathogens, nor has sufficient requisite information (species traits, circumstances of species’ entrance, persistence, spread, rankings of new ranges by degree of vulnerability) been assembled to permit the identification of invaders. And there is not adequate information to quantify the uncertainties in the available predictive schemes.
Expert judgment—the synthetic ability of investigators who have long experience in monitoring the fate of species in new ranges—is now used to bridge the gap between a body of largely qualitative data and a determination of the probability of an organism’s arrival and ability to establish. The collective experience of experts suggests that there is a conceptual basis for understanding invasions; the challenge is to transform judgment into transparent, repeatable, quantitative, and comprehensive predictive capability.
Conclusion 3. The inability to predict accurately which nonindigenous species will become invasive stems from a lack of comprehensive knowledge of the events that dictate species’ immigration (arrival), persistence (survival), and invasion (proliferation and spread) in new environments. The requisite knowledge would be based on critical observation of the natural history of nonindigenous species and experiments designed specifically to evaluate nonindigenous species in the stochastic environments they encounter in new ranges.
It is not now possible for experts or predictive systems to be more unequivocal or definitive about invasiveness, because only part of the needed information is available. The committee found that specific, relevant information on the performance of species in new ranges is lacking, especially the results from experiments designed to evaluate the ability of nonindigenous species to persist and proliferate in a broad range of environmental conditions. Heretofore there has been insufficient attention to the fate of immigrant populations in demographic terms, except for scattered accounts of organisms released as agents of biological control and some plant pathogens.
Conclusion 4. Some data on the natural history of plant pests exist, but they often reside in grey literature and in datasets that are not easily accessible. Data on events that potentially lead to invasions are frequently collected by federal or state agencies in the course of pest surveys and inspections and after releases of biocontrol agents. Such collections need to be more comprehensive and need to be implemented on a quantitative, statistically sound
basis. Datasets need to be organized in a way that allows them to be analyzed and evaluated from the perspective of understanding invasiveness.
Much information has been gathered on the modes, pathways, and circumstances of organisms’ dispersal on all geographic scales (Ridley 1930, van der Pijl 1969); indeed, these topics have been active arenas of scholarship among biologists for over a century. However, full value of this information is seldom obtained. Instead, we find basic references providing much the same litany of pathways, in the same detail, as was assembled decades earlier (Baker 1972, 1978, 1989).
The huge literature on the natural history of species is simultaneously a strength and a handicap in contributing to our understanding of the invasion process; much of the relevant information is mixed with ancillary information. A more systematized manner is needed to report and to access what is known about the natural history of potential immigrants, including species with a record of invasion, and results of localized pest surveys or evaluations of biocontrol releases. Such a system could provide a relatively rapid process for determining what is known about a species’ natural history, native and current new ranges, habitat, hosts, prey, and impact. It could also minimize duplication of efforts, and identify priorities for research or survey activities.
In the following recommendations, the committee points to ways of strengthening the scientific basis for predicting the invasive potential of plant pests. The first three recommendations are directed toward USDA-APHIS and its regulatory activities because they are related to our understanding of the scientific basis of prediction. The other recommendations require action by USDA, other federal and state agencies, and the scientific community. Recommendations 4-7 are related to the documentation and standardization that are needed to understand invasions better. Recommendations 8-10 focus on needed research, and recommendations 11 and 12 point to the organizational infrastructure and scientific expertise that are needed to make headway in predicting invasions.
Recommendation 1. The Port Information Network (PIN) database maintained by APHIS is a potentially valuable source of information for understanding the pathways by which potential invaders arrive at U.S. borders, but the utility and availability of the data could be substantially improved. Sampling methodology should be statistically designed and implemented consistently. Sampling protocols at ports and borders should be re-evaluated and revised as necessary to ensure that pest interception data are accurate
and meaningful. Data collection should be expanded to include vascular plants (in addition to those on the federal noxious weed and seed lists). Increased efforts are also needed to detect and identify pathogens consistently. Improved technology to detect hitchhiking insects and plant pathogens arriving with cargo, baggage, and related commodities could improve the utility of the PIN database (as well as reduce opportunities for new, potentially invasive pests to immigrate). The value of the database would be increased by including additional variables, such as a record of inspections that result in the detection of zero pests, some measure of the abundance of detected pests, and interceptions of nonquarantined pests. The data should be monitored consistently and regularly to identify and correct problems in data entry or maintenance. The PIN database should be accessible for analysis by investigators in universities and other agencies in collaboration with APHIS personnel who are familiar with the database.
There is great potential to learn more from the PIN data that vary in taxonomic status (such as individual species, families, guilds, orders, and class), spatial scale (such as selected ports, regions, or the entire United States) and temporal pattern. The complexity of the data and the background information needed to interpret it accurately require assistance from APHIS staff. APHIS should continue to collaborate in the analysis of the PIN data with scientists outside APHIS who have the relevant expertise and interests while working to make the database more independently accessible.
Recommendation 2. APHIS risk assessments combine a system of predicting an organism’s arrival and establishment with an estimation of the possible consequences. The assessments are based on scientific concepts but contain uncertainties because of gaps in available information. To strengthen the overall prediction of invasive potential, the basis of APHIS risk assessments should be better documented, and assumptions made in each step should be listed and explained, so that independent experts can rationally compare conclusions about the likelihood of arrival, establishment, and impact. The assessment procedure should be transparent, repeatable, peer-reviewed, and updated to capture new information and enhance expert judgment.
Although risk assessments are used by APHIS to manage imports, they constitute the largest body of predictive systems so far attempted, insofar as they and related systems formed by other nations incorporate a set of assumptions about an organism’s ability to arrive, establish, and cause damage. Those assumptions are tested after a decision to import (or not to import) a commodity or a new plant species. The arrival of a pest may or may not indicate a failure of the assessment, but examining the assumptions in light of new information or events requires that the assessments be explicitly stated.
Recommendation 3. The framework used by USDA to evaluate imported plants for potential release as forage, crops, soil reclamation, and ornamental landscaping should be expanded to include rapid multitiered evaluation of the hazards that these species might pose. Controlled experimental field screening for potentially invasive species should be pursued for species whose features are associated with establishment and rapid spread without cultivation and whose immigration history is unknown. Similar efforts to acquire life-history and population data in situations that approximate field settings would be beneficial in the case of nonindigenous insects and pathogens of concern, including species proposed for deliberate introduction.
A species’ performance in environments it has not encountered before can be quantified only through experimentation (Hairston 1989, Mack 1996a). The paucity of our knowledge of the early stages of an invasion demonstrates the need for experimentation based on carefully constructed hypotheses about a species in specific ranges under measured environmental conditions.
A geographically broad network of experimental gardens should be established to identify species that could readily spread and persist. In addition to averting the release of a potentially invasive species, results of field trials would substantially enhance the ability to detect common patterns of plant performance in new environments. Extending those ideas to insects and pathogens, including those under consideration for deliberate introduction, would be valuable, although practical considerations associated with experiments in field settings may be more difficult to overcome.
DOCUMENTATION AND STANDARDIZATION
Recommendation 4. Information on invasions by plants and plant pests around the world should be assembled and updated regularly. Explicit information on new invasions in the United States—such as description of new locales, the partitioning of the species’ genetic variation, and epidemiology of its spread—should be gathered and communicated more effectively to the scientific community; this information is essential in continually revising expert judgment. Careful recording of the circumstances of arrival, persistence, and invasion of nonindigenous species in the United States would substantially improve prediction and risk assessment.
First detection of the escape of a potentially harmful nonindigenous species in the United States and the rapid communication of this discovery among investigators, APHIS, land managers, and others interested in nonindigenous species are neither certain nor routine. Reporting of first detection is often restricted to informal alerts to subscribers on e-mail lists, and the reported information that is reported varies widely in completeness. If a high-profile invasive species, such as the Asian long-horned beetle or the zebra mussel, is detected, there may be an
article in a major local newspaper. But finding a new nonindigenous plant rarely rates such a high-profile announcement. A more coherent system of reporting would centralize the responsibility for gathering and disseminating alerts about new species over an established network.
The need to make such information readily and instantly available via the World Wide Web has been identified repeatedly as a goal for invasion biology in general and international and national biosecurity in particular (Ewel et al. 1999, Ricciardi et al. 2000). Some databases are being assembled with those general goals in mind. The current U.S. Geological Survey database of invasive species in the United States, although modest in scope, is a productive step. Other government agencies and some professional organizations have begun compiling similar information (Ricciardi et al. 2000). Such purposeful data assembly and rapid communication will facilitate the continuing revision of our knowledge of the attributes, circumstances, and environments in which immigrant species arrive, are extirpated or persist, and occasionally become invasive. The information contributes directly toward construction of a predictive framework for invasions.
Recommendation 5. A literature synthesis on the natural history of potential immigrant species, similar to the “Biology of Weeds” series published by the Canadian Journal of Plant Science, should be established, standardized, and made accessible via the Internet.
The natural history of plants proposed for introduction into the United States should be much better documented. The long-running “Biology of Weeds” series of publications in the Canadian Journal of Plant Science and the “Biological Flora of the British Isles” series in the Journal of Ecology each deal with one species (or at most a few closely related congeners) and illustrate an approach that the committee recommends. All the natural history and ecology known for the species is organized in these publications under standardized headings, such as “phenology”, “habitat”, “geographic ranges” (with maps), and “predators and parasites”. The articles reflect exhaustive searches of the literature to assemble a comprehensive record of knowledge for species (see Ricciardi et al. 2000).
Such literature syntheses for potential immigrant species should be broadened. We cannot and do not need to know the detailed life history of every species on the planet. But deliberately augmenting existing systems, as illustrated by those two journal series, would substantially improve expert judgment. In that regard, cooperation with biologists outside the United States, especially in nations that have species likely to arrive in the United States (China is a prime example), in the assembly of the relevant information about their biota should be encouraged and supported.
Recommendation 6. Information on the structure and composition of natural ecosystems in North America (and the disturbance regimes within
them) should be reinterpreted by the scientific community to analyze these ecosystems’ vulnerability to biotic invasion. Attention should be paid to identifying groups of native species that could be vulnerable or could facilitate the establishment of nonindigenous species.
Ecologists have actively described and quantified the natural communities of the United States for a century (Curtis 1959, Barbour and Major 1988, Franklin and Dyrness 1988). Although varying considerably in taxonomic and ecological thoroughness, this body of information is nonetheless impressive for its geographic and environmental comprehensiveness. Furthermore, attempts to categorize the different vulnerabilities of communities to biotic invasion have begun (Lonsdale 1999). To answer the hypothetical question of which species will invade which communities, we need to view the communities from the standpoint of the fundamental processes, forces, and factors that dictate community composition (Tilman 1997, Naeem et al. 2000). What influences the species composition, including nonplant and microbial species of each community, and provides any restriction or vulnerability to future species additions? Answers to such questions require not so much new community analyses as reinterpretation of existing information.
To predict future invaders, we need a better understanding of the functional groups—taxonomically unrelated species that perform similar community roles— in communities (for example, Metzger 2000). We need to identify communities whose lack of particular functional groups makes them vulnerable to the entry and establishment of nonindigenous species that perform those roles. For example, a community with little representation by plants with nitrogen-fixing nodules may be vulnerable to the arrival and persistence of nonindigenous species that are nitrogen fixers (Vitousek et al. 1987, Vitousek and Walker 1989).
Recommendation 7. A consensus on procedures to measure the impact of invasive species should be forged in the scientific and regulatory communities, and there should be more reporting of impacts of invasive species with standardized measures.
Without consistent measures of impact, conclusions about the potential consequences of an invasive plant pest will remain vague; this will hamper the incorporation of impact into predictive systems, including risk assessments. Establishing a common language for measures that characterize impact at different levels of biological organization, from effects on individuals to ecosystems, is essential. Measures of economic and social impact are also needed.
Recommendation 8. Research on host specificity among pathogens and the correlation of some life-history traits (such as dispersal mode, reproductive
system, and host range) with their documented ability to invade should be undertaken in relation to potential environments in new regions or areas. Question-oriented experimentation to elucidate relationships among species’ traits and their new environments and hosts should be supported.
Long-term investigation in plant pathology has provided voluminous information on the intimate host-pathogen associations that form reliable guides for future attack on the same hosts, even if pathogen and host meet in new geographic ranges (Parry 1990, Lucas 1998). Similar intimate associations occur among many nonindigenous phytophagous insects and their host plants. These interactions form the basis for strong predictions of whether an immigrant pathogen or arthropod will present a threat. Collection of this information needs to be supported. In contrast, few plants display such detrimental interaction with a restricted group of other plant species. And the link between plant species’ traits and their invasiveness has often proved difficult to identify. Nevertheless, life-history traits ultimately do influence the ability of a nonindigenous species to persist in a new range, and this link should be explored more fully, especially for insects and pathogens (Rejmanek and Richardson 1996, Reichard and Hamilton 1997).
In addition, of species that have the potential to immigrate and become invasive, little is known about many until they inflict harm. For instance, we know much about the biology of Melaleuca quinquenervia (melaleuca), Lantana camara (lantana), Salvina molesta (karibaweed), and Euphorbia esula (leafy spurge), but little or nothing about their apparently noninvasive congeners. This gap in our information denies us a powerful comparative tool with which we could potentially glean much insight into the performance of invasive species by understanding how specifically and quantitatively they differ from their closest relatives. Barrett (1979, 1980) explained much of the invasive ability of Eichornia crassipes (water hyacinth) by comparing the mating system and vegetative propagation of this horrific pest in comparison with its seemingly harmless congeners. Investigating the ecology of the congeners of invasive species could become a powerful tool for identifying characters or attributes of invaders that directly contribute to their invasive ability (Barrett 1979,1980, Schierenbeck et al. 1995).
Studies of associations and interactions between invasive species, particularly vectors and pathogens, can also benefit from an understanding of congeners. For example, beech bark disease is a lethal disease complex in which a scale insect, Cryptococcus fagisuga, predisposes beech bark to infection by several taxa of the fungal genus Nectria. Although a native Nectria species complexes with C. fagisuga to cause beech bark disease, an introduced species of Nectria (N. coccinea var. coccinea ectria) now predominates in most areas of the Northeast. In areas newly invaded by the scale insect, it appears to replace the native Nectria species rapidly, raising questions about the association of vectors with closely related pathogens and about the significance of the disease’s impact if only the insect and not the fungus had been introduced (Mahoney et al. 1999).
Recommendation 9. The release of nonindigenous organisms for biological control presents an opportunity to collect detailed demographic data on immigrant populations from the moment of their introduction. A substantial effort should be made to document the fate of these organisms, including the efficacy of the introduced organism on the target pest and on nontarget species, as a guide to the performance of unintended releases and as a mechanism to improve risk assessment in deliberate introductions of nonindigenous species.
Understanding and quantifying the forces that cause populations to become extinct, especially by natural forces in a new range, will prove central to building our understanding of the invasion process (Harper 1982, Mack et al. 2000). The comparatively little that is known in quantitative terms about invasions has been learned by following the rare event of a nonindigenous species that actually develops an invasion. We also need comprehensive knowledge of the circumstances under which almost all immigrants go extinct—whether in transit, at the point of entry, or as adventive species in a new range (Mack 2000)—or those that decline numerically after having seemingly attained persistence (Simpson 1984). Releases of organisms for biological control are, in effect, experiments that assess the performance of species in their native range and simultaneously in potential new ranges. Following the fate of these populations could be invaluable in that the genetic composition, precise native range, size of the founder population, and time of release are all probably known (Grevstad 1999a,b).
Recommendation 10. Plants native to the United States that are growing in other countries, such as in botanical gardens and arboretums, should be monitored to determine the species to which they are susceptible and to evaluate the potential for these species to arrive in the United States. The severity of the damage to native U.S. plants by pathogens, arthropods, and other taxa, and the abiotic and biotic forces that contribute actively to the damage, should be documented.
Expert judgment could be improved substantially through a deliberate and expanded effort to assess species that attack U.S. crops and native plants in ranges outside the United States. Cereal crops, cotton, soybeans, some pines, and numerous ornamental trees, shrubs, and herbaceous plants are grown worldwide and are exposed to pests not currently found in the United States. Reports of the organisms that attack these plants are invaluable when coupled with some estimate of the potential for the harmful species to arrive in the United States (Hewitt and Chiarappa 1977). For instance, the potential hazard of introduction of the Russian wheat aphid and plum pox was understood, on the basis of foreign experience, before these species arrived in the United States. But too many species of plant pathogens and phytophagous arthropods occur outside the United States to evaluate the potential damage from all of them. The list of species that deserve careful attention would be substantially reduced and focused by monitor-
ing of pests of U.S. crops and native U.S. species growing in foreign ranges. Such monitoring would provide early warning of nonindigenous species that, if introduced, would become threats in the United States. Australia has already begun a similar form of assessment for its invaluable native eucalypts growing in Southeast Asia, and its experience could form a valuable precedent (Wylie and Floyd 1999).
INFRASTRUCTURE AND EXPERTISE
Recommendation 11. A central repository of information relevant to immigrant species would accelerate efforts to strengthen the scientific basis of predicting invasion. Information collected by federal, state, and international agencies, academic researchers, and others should be brought together in a single information facility or service so that it can be evaluated collectively, to permit the construction of needed datasets and the design of appropriate experiments, and to document the circumstances surrounding invasions.
The considerable information that is required to build comprehensive predictions of invasion would be best assembled by a central authority with strong ties to and guidance from user groups. The information needed is so diverse that only a facility with a specific mission to gather it and make it available will be effective in achieving an understanding of invasion and an ability to predict it. Requests for a centralized authority on biological invasion have been made by many in the scientific community (Ewel et al. 1999, Schmitz and Simberloff 2001). Information that could be collected by such a facility includes much of the types suggested in this report, such as the documentation of newly detected nonindigenous organisms in the United States, information on biotic invasions around the world, and the results of organisms released for biological control. The creation of a central facility of this type would help with efforts to predict invasive species and is consistent with the goals of the National Invasive Species Council (2001).
Recommendation 12. Focused scientific investigation is inadequate on nonindigenous species and prospective new environments from the overarching perspective of invasive potential. Multidisciplinary collaboration should be encouraged and supported among scientists with taxonomic expertise and those who specialize in population biology, community ecology, epidemiology, and simulation modeling. Multidisciplinary training of established and new investigators is needed to provide the expertise needed to make the study of invasion biology predictive.
In spite of a long history of interest in biological invasion, scientific inquiry in invasion is still nascent. Progress in understanding and predicting invasion will depend on how well the insights of investigators with diverse training can be coalesced and directed to decipher the myriad combinations of immigrant species,
new environments, and novel circumstances that can produce biological invasion. The last 10 years has seen the emergence of a broad consensus that the prediction of biological invasion is of pressing national need (U.S. Congress 1993, Clinton 1999). It will take some time, however, to generate the predictive principles on which policy-makers, regulators, the scientific community and the public can have confidence.
On the basis on the current state of prediction of the fate and effect of nonindigenous species, the committee concludes that action needs to be initiated now to replace the largely retrospective or anecdote-based prediction process with a system derived from experimentation. (In a larger sense, science has always progressed in this manner—first a long period of natural history and observation, and then continuing, deliberate, question-based experimentation.) This sense of urgency in moving in a deliberate manner from expert judgment to prediction on the basis of experimentation is driven by the pressing societal need to deal fairly, adequately, and swiftly with an unparalleled volume and scope of trade involving nonindigenous species. These are policy issues, based in science, that cannot be left lingering and cannot be resolved through happenstance investigation.