Issue Section:
 Overview Articles
Most of the concern over the last decade about US vulnerability to bioterrorism has focused on terrorist use of pathogens to attack the civilian population. This concern increased in the wake of the September 11 terrorist attacks on the World Trade Towers and the Pentagon and the anthrax letter attacks on US Senate offices and the media. However, a number of analysts have pointed out that terrorist attacks on livestock or crops, although unlikely to cause terror, are also a concern because they could be executed much more easily and could have serious economic consequences (Frazier and Richardson 1999, Horn and Breeze 1999, Casagrande 2000, Wheelis 2000). It is worth considering the consequences for the US economy had there been a widespread and sudden outbreak of foot-and-mouth disease (FMD) shortly after September 11. The stock market probably would have plunged even further, and its recovery could have been significantly delayed. More substantial consequences are easy to imagine. This article will give an overview of US vulnerability to agricultural bioterrorism and biocrimes; the accompanying articles and a previous article in this journal (Madden and van den Bosch 2002) examine individual facets in more detail. See Whitby and Rogers (1997), Schaad et al. (1999), Wheelis (1999a), Wilson et al. (2000), and Whitby (2002) for a historical overview of attacks on agriculture with biological weapons, a subject that is beyond the scope of this article.

Diseases have a significant negative impact on agricultural productivity

The burden on agriculture of endemic and naturally imported epidemic disease is high, confirming the capacity of animal and plant diseases to cause economic harm. The United States is free of many globally significant livestock diseases because of effective surveillance of herds and imports and aggressive eradication campaigns. Even so, approximately $17.5 billion dollars are lost each year because of diseased livestock and poultry. In general, losses from animal disease account for 17% of the production costs of animal products in the developed world and nearly twice that amount in the developing world (ARS 2002).
The total cost of crop diseases to the US economy has been estimated to be in excess of $30 billion per year (Pimentel et al. 2000). The costs include reduction in the quantity (e.g., reduced bushels per acre) and quality (e.g., blemished fruit, toxins in grain) of yield; short-term costs of control (e.g., cost of purchasing and applying pesticides) and long-term costs (e.g., development of resistant varieties of crops through breeding and development of new pesticides); extra costs of harvesting and grading diseased agricultural products (e.g., separating diseased from disease-free fruit); costs of replanting blighted fields; costs of growing less desirable crops that are not susceptible to the dominant plant pathogens in an area; higher food prices; unavailable foods; trade disruptions; and public and animal health costs caused by the production of toxins by some plant pathogens (Zadoks and Schein 1979, James et al. 1991, Madden and Nutter 1995).
In contrast to the sweeping campaigns undertaken to eliminate the most virulent diseases of livestock, efforts generally have not been made to eradicate diseases of crops (Strange 1993). One goal of plant disease control has been to maintain most indigenous diseases at a low or very low incidence level through a range of management techniques (Fry 1982). The exception is when a disease has a very narrow geographic distribution (as would a newly introduced exotic disease), spores are not dispersed great distances, and disease incidence is low. In such a situation, eradication may be feasible.
Despite the high toll endemic disease and periodic incursions of epidemic disease exact on agriculture, many pathogens have not appeared in the United States at all, while others have made only very rare appearances, and still others were eradicated decades ago (especially with animals); many of these are considered to be serious threats to agriculture (Brown and Slenning 1996, Madden 2001). Thus, the exotic, highly contagious pathogens causing these diseases could be chosen as bioweapons for the large economic consequences that could result from their introduction. Pathogens that cause diseases such as FMD, rinderpest, African swine fever (ASF), soybean rust, Philippine downy mildew of maize, potato wart, and citrus greening could, if introduced into the continental United States, have serious consequences for the US economy.

Agricultural bioterrorist attack can have severe economic consequences

Even a massive outbreak of plant or animal disease in the United States would not cause famine; the agricultural sector is too diverse, too productive, and too closely regulated for that to be a realistic possibility. However, a successful attack could have severe economic consequences. The most substantial impact would be the loss of international markets for animal or plant materials. Member nations of the World Trade Organization are entitled to ban imports of plant or animal materials that may introduce exotic diseases into their territories (Wheelis 1999b, FAS 2001). Thus, importing countries that are themselves free of a particular highly contagious animal or plant disease will routinely impose sanitary or phytosanitary restrictions on trade with countries in which that disease breaks out. This can result in billions of dollars of lost trade.
For instance, as soon as the first case of FMD was reported in the United Kingdom last year, the European Union (and other countries) immediately blocked imports of British beef, sheep, and swine and products derived from them. The total sum of lost revenues from contracted international markets has not yet been determined, but it will certainly be billions of dollars. For the United States, with $41 billion of beef, $19 billion of diary, and $14 billion in pork sales annually (USDA 1997), the trade consequences of an outbreak of FMD could be much larger. A recent study of the impact that an outbreak of FMD would have on California agriculture concluded that losses, using conservative estimates, would be $6 billion to $13 billion even if the outbreak were contained within California and eradicated within 5 to 12 weeks (Ekboir 1999).
Karnal bunt of wheat, caused by the fungus Tilletia indica, provides another example of severe economic consequences caused by agricultural disease. About 80 countries ban wheat imports from regions with karnal bunt infections, even though the disease does not have a large direct effect on crop yield (Bandyopadhyay and Frederiksen 1999). When the disease was discovered in Arizona and surrounding areas in 1996 (probably from an accidental introduction from Mexico), there was an immediate threat to the overall $6 billion per year US wheat crop, since about 50% of produced wheat is exported. Because of this threat, the Animal and Plant Health Inspection Service (APHIS) of the US Department of Agriculture (USDA) immediately mobilized efforts to contain the outbreak within the original small area and to eradicate the disease. From 1996 to 1998, APHIS spent over $60 million on the effort, and growers in this small affected area lost well over $100 million from lost sales and increases in production costs (Bandyopadhyay and Frederiksen 1999; John Neesen [APHIS], personal communication, 2 April 2002). In this case, the localized nature of the outbreak allowed the United States to convince its trading partners that none of the contaminated wheat was entering the market, and wheat exports continued from the rest of the country. Unfortunately, karnal bunt was recently discovered again, this time in Texas (AgNet 2001), and a new round of containment and eradication efforts has been initiated.
In some cases, domestic demand can also be significantly affected. Even minor outbreaks of disease that can potentially infect people can have severe economic consequences. Since September 11, a mere three cases of mad cow disease have been found in Japan; yet as a consequence, Japanese beef sales dropped approximately 50% during this period (Watts 2001).
In addition to the costs that result from reduced international and domestic demand, the costs of containment can be quite substantial, as the examples discussed above make clear. Thus, even for commodities that are not exported in large amounts, an outbreak of disease that provokes vigorous eradication efforts may have a substantial economic effect. Taiwan, for instance, spent about $4 billion in an unsuccessful effort to eradicate FMD after it was introduced to the country in 1997 (Wilson et al. 2000).

Containment, eradication, or control?

As demonstrated in the examples above, introductions of exotic pathogens that cause highly contagious animal or plant diseases may elicit rapid and aggressive attempts to contain and eradicate them, and these measures commonly cause more economic damage in the short term than the disease itself. Despite the costs, such intervention is often justified, since if exotic highly infectious diseases become endemic, the long-term costs would be much greater than the costs of containment.
Containment and eradication of exotic animal diseases is commonly done by culling all potentially exposed animals to break the chain of transmission (Ferguson et al. 2001a, 2001b). Thus, small numbers of infected animals can lead to the slaughter of large numbers of healthy ones. Many of the animal diseases that are potential bioterrorist threats are caused by viruses, for which there is no practical therapy once the animal is infected. Therefore, transmission cannot be interrupted by treatment, but only by culling diseased and exposed animals or by vaccination (when that is an option—see below). In contrast, about 75% of plant diseases are caused by fungi, and these can be controlled, with varying degrees of effectiveness, by the application of fungicides (Strange 1993, Agrios 1997). For many high-value-per-acre crops (e.g., fruit and vegetables, ornamentals), fungicides are used in routine control of endemic diseases. Some fungicides actually move systemically within plants and can arrest the infection process during the early phases of infection. More commonly, however, fungicides are applied to the surface of plants and are used prophylactically to provide short-term protection from fungal infection (Fry 1982). When an introduced disease is discovered, infected and possibly exposed plants are culled (“rogued”), and fungicides can be used to treat plants in surrounding areas (even for low-value-per-acre plants) to prevent infection. This method is expensive, it fails to prevent all infections, and it can have negative environmental consequences.
Transmission of bacterial and viral crop diseases is difficult to control with chemical pesticides, unless such diseases are transmitted by insect vectors, in which case insecticides may be useful (Madden et al. 2000). Because of these difficulties, containment and eradication of bacterial pathogens depend heavily on quarantining infected areas and removing all infected and exposed plants.
The only chance of successfully containing and eradicating a crop pathogen is to start the process relatively soon after introduction, when the focus of infection is small, there are few infected individuals, and the dispersal distance of spores is short. For some diseases, such as rust of several crops (e.g., stem rust of wheat, caused by Puccinia graminis f. sp. tritici), spores can be dispersed very long distances (thousands of kilometers), so the spread of disease can be substantial before the pathogen is discovered (Campbell and Madden 1990). For these reasons, eradication is generally not attempted.

Agricultural bioterrorist attack requires relatively little expertise or technology

One of the reasons that a bioterrorist attack on human populations is difficult is that the development of an effective bioweapon is a technically daunting task (Falkenrath et al. 1998). Many of the antipersonnel agents that have been used as weapons (“weaponized”) are poorly transmitted among humans (e.g., anthrax), so a large amount has to be disseminated at once to cause large numbers of casualties. The only effective way to infect large numbers of people simultaneously is to generate a respirable aerosol. However, aerosol preparation to a particle size that is effective in causing inhalational disease is quite difficult, and once so prepared, it is highly hazardous to the perpetrators themselves (unless they are vaccinated and taking prophylactic antibiotics).
Other anti-human agents are contagious (e.g., Yersinia pestis, the causative agent of plague), but they too have to be disseminated in large quantities for widespread infections, because agent transmission can be interrupted by antibiotic treatment. Since organisms such as Y. pestis do not form the highly resistant spore form that Bacillus does, it is technically quite challenging (and dangerous) to prepare a large stockpile of the agent. Still other agents are viral rather than bacterial, and their preparation and weaponization is even more challenging because of the more demanding technologies needed for laboratory culture.
In some special situations, highly contagious viruses could be effectively introduced by voluntarily infected terrorists who would travel to the target area during the incubation period of the disease. This reportedly was done a number of times in the 1950s and 1960s in an effort to infect Native Americans in the Matto Grosso of Brazil, by land speculators who would be able to purchase tribal lands once the natives no longer inhabited them (Davis 1977). However, in the developed world, for any disease other than smallpox, it is unlikely that such a low-tech method would be effective. Thus, the technical expertise required to mount a mass-casualty biological attack on the human population is formidable and probably beyond the capabilities of most terrorist groups (and indeed of many nations). However, the recent anthrax infections have clearly shown that only a few cases are sufficient to produce a large psychological impact on the population.
Unfortunately, the same difficulties do not exist for many of the diseases that would make effective agricultural bioterrorist weapons. These diseases of animals and crops are highly contagious and spread effectively from a point source, as the recent FMD outbreak in the United Kingdom dramatically confirms. Moreover, humans can safely handle the causative organisms, with no risk of becoming infected. None of the plant pathogens of concern, nor most of the animal pathogens, cause disease in humans. Thus, there is no need for vaccination, prophylactic antibiotic use, or special precautions to prevent infection of the perpetrators.
Although a small outbreak may not produce a large psychological impact (relative to a single person dying of anthrax or smallpox), several of these pathogens owe much of their economic impact to trade sanctions that are imposed in response to a few cases; thus, even small outbreaks can have very large economic effects. A few cases of FMD scattered around the country could interrupt much of US animal product exports for several months, even if the outbreaks were promptly contained (importing countries would want to wait several weeks or months to verify that the outbreak was truly contained before resuming imports). Obviously it is technically easier to cause a few scattered cases of disease than to prepare a kilogram-sized stockpile of weaponized agent for aerosol distribution.
Material to initiate an outbreak of plant or animal disease therefore does not have to be prepared in large quantity—a few milligrams could be sufficient to initiate multiple outbreaks in widely separated locations—if the goal is to disrupt international trade, or if the terrorists are sufficiently patient to allow a crop disease to develop over several months by transmission from individual to individual. And the agent does not necessarily have to be grown in the laboratory or otherwise manipulated—a small amount of natural material taken from a diseased animal or plant can serve without any additional manipulation. For instance, a few hundred microliters of scrapings from the blistered mucosa of an FMD-infected animal, or blood from an animal hemorrhaging from ASF, or a handful of wheat tillers heavily infected by the stem rust pathogen can provide more than enough agent to initiate an epidemic. Such materials are readily available in many places in the world where the diseases of concern are endemic, and they can be obtained and transported without any particular expertise other than what is necessary to recognize the disease symptoms with confidence. Since only small amounts are needed, they can be easily smuggled into the country with essentially no chance of detection.
Dissemination of many introduced pathogens likewise requires relatively little expertise. Animal virus preparations could be diluted and disseminated with a simple atomizer in close proximity to target animals, or the preparation smeared directly on the nostrils or mouths of a small number of animals. This could be done from rural roads with essentially no chance of detection. Dissemination of animal diseases could also be done surreptitiously at an animal auction or near barns where animals are densely penned (as in chicken houses or piggeries). For plant diseases, simply exposing a mass of sporulating fungi to the air immediately upwind of a target field could be effective, if environmental conditions were favorable for infection. The biggest challenge of introducing a plant pathogen is probably timing the release with the appropriate weather conditions (Campbell and Madden 1990). If pathogens are released immediately before the start of a dry period, few, if any, infections are likely to result. However, if released at the start of a rainy period, these pathogens could cause a major epidemic.
The technical ease of introducing many agricultural pathogens makes it more likely that terrorists or criminals would release pathogens in several locations in an attempt to initiate multiple, simultaneous outbreaks. This would ensure that trade sanctions would be imposed, because it would undermine any argument that the outbreaks are localized and do not jeopardize importing countries. It would also be more likely to overwhelm the response capacity and lead to the uncontrollable spread of disease. This is the principal way in which a bioterrorist attack would differ from a natural disease introduction, and it raises the question whether a system designed to respond to natural introductions can deal effectively with sudden, multifocal outbreaks.
Under some circumstances, a pathogen could be effectively introduced without the perpetrators entering the country. This is, of course, true of crops planted on both sides of an international border, such as sorghum along the Mexican border or wheat and barley along the Canadian and Mexican borders. These crops have experienced disease outbreaks that spread from acreage outside the United States—sorghum ergot, karnal bunt of wheat, and barley stripe rust, for example (Chen et al. 1995, Bandyopadhyay and Frederiksen 1999). Thus, such pathogens could be deliberately introduced in an adjacent country, a potential advantage to a terrorist if disease surveillance and control programs were less effective there. Multiplication of the pathogen in the foreign acreage could lead to large numbers of spores blowing across the border and initiating an outbreak that could very quickly become very large. International movements of animals also provide opportunities for the introduction of an animal pathogen without the perpetrator having to enter the country. For instance, the very serious 1997–1998 outbreak of classical swine fever in the Netherlands was due to inadequate disinfection of a swine shipment from Germany (Greiser-Wilke et al. 2000). However, this is an unlikely route for bioterrorist attack, because its effectiveness requires a failure of quarantine or disinfection procedures.
International trade in nonliving agricultural materials can also introduce disease. An example is the 2000 outbreak of FMD in Japan, which occurred simultaneously in two widely separated locations. Investigation suggested that at least one of these locations was infected by straw imported from China (where the FMD virus is endemic) and used as bedding for cattle (Matsubara 2000). Although this particular event appears to have been natural, it shows that deliberate contamination of materials such as bedding or fodder could initiate simultaneous outbreaks at widely scattered locations, making containment extremely difficult. The invasive Asian longhorned beetle, a pest of trees, probably arrived in the United States from China via a similar route, from eggs laid in wooden packing material. If this pest becomes established in the United States, the damage to fruit, lumber, and tourism industries is expected to exceed $40 billion (APHIS 2002). For crops, seeds present similar vulnerability. Many plant pathogens either infect or reside on seeds (Agrios 1997), and a considerable (and increasing) amount of seed for US crops is produced outside the United States (Condon 1997).
Finally, there is a substantial moral difference between killing people and killing plants and animals, and a corresponding difference in the intensity of expected law enforcement response and legal consequence. Thus moral norms and legal consequences can be expected to constitute less of a disincentive to the agricultural bioterrorist than to the bioterrorist who targets people.

Agricultural bioterrorists could have a variety of motives

We normally think of a terrorist attack as being ideologically motivated, and this is certainly one motive for attacking the agricultural sector. The immense potential for economic damage could make this kind of attack attractive to enemies of the United States, particularly because it is relatively easy and safe for the perpetrators. Given the escalating scale of terrorist attacks on the United States, this is cause for serious concern.
In addition to the political and religious ideological motivations for terrorism, agriculture provides some new ones (Casagrande 2000). There is considerable opposition to the increasing use of genetically modified (GM) crop plants and domestic animals, which have been largely developed in the United States and are most widely used here. For instance, GM soybeans accounted for 63% of the crop grown in the United States in 2001 (Bohan 2001). Worldwide, over 130 million acres are planted with GM crops, because they possess increased resistance to herbicides or insects. The United States alone cultivates almost 70% of the total acreage of GM crops (James 2001). Opposition to the use of GM crops and animals has sometimes taken the form of vandalism and destruction (Greenstone 2001), and it is quite possible that some activists will at some point turn to diseases as weapons to attack GM organisms. Radical animal rights groups may wish to attack animal agriculture to prevent corporations from profiting from animal suffering. Ingrid Newkirk, president of People for the Ethical Treatment of Animals, stated recently that she openly hoped “that it [FMD] comes here [the United States]. It will bring economic harm only for those who profit from giving people heart attacks and giving animals a concentration camp-like existence. It would be good for animals, good for human health, and good for the environment” (Elsner 2001).
Attacks on the agricultural sector could also be motivated by greed, properly termed “biocriminality” rather than bioterrorism. The major shifts in agricultural markets and commodity prices that could result from a successful attack could provide such economic motivation. Profit could be made by the manipulation of futures markets, selling short the stock of major agrochemical companies, or intentionally sabotaging overseas competitors to capture lost import markets. For instance, outbreaks of FMD have changed global dominance in the export of pork. In the 1980s, Denmark supplied most of the pork imported by Japan. After a 1982 FMD outbreak halted pork exports from Denmark, Taiwan filled Japan's need for pork and continued to be its primary supplier even after Denmark was declared free of FMD. After the 1997 FMD outbreak in Taiwan, the United States captured the Japanese pork market and continues to supply Japan with most of its imported pork (FAS 1997). Corporations, individuals, organized crime groups, and even national governments might be attracted to the very large financial gain that is at least theoretically achievable from the judicious use of plant or animal diseases to manipulate markets or commodity prices.
Another possible motive is revenge. The United States and the United Nations Drug Control Program have supported research and development of the use of plant pathogens for killing or reducing yields of opium poppy, coca, and cannabis (Kleiner 1999, Jelsma 2001). The programs involved selection of virulent strains of fungi, consideration of large-scale production of fungal spores, and testing of the most efficient ways of delivering the spores. The work is exactly analogous to the anticrop biological weapons programs of the former Soviet Union and the United States during the cold war (Whitby and Rogers 1997). Because of various political and social pressures, these programs are on hold or moving very slowly. However, if the deliberate release of plant pathogens to destroy drug crops did ever go ahead, there could be a powerful incentive for those in the illicit drug business to retaliate by releasing plant pathogens into US crops (Stone 2000).

Prevention and response

In response to a 1998 Presidential Decision Directive (White House 1998), and especially to the attacks on September 11 and the subsequent anthrax infections, considerable effort is now being expended to reduce the vulnerability of the United States to bioterrorist attack. A small part of this effort is aimed at protecting crops and livestock. Among other things, a National Research Council committee is evaluating the vulnerability of US agriculture to biological attack and determining strategies for dealing with that vulnerability. USDA and other federal departments and agencies are in the process of making many changes, and it is premature to comment on their effectiveness. However, it is worth making a few general points and sketching the scientific agenda as we see it.
Aggressive counterterrorism measures and greater international intelligence sharing can be expected to reduce the likelihood of a bioterrorist attack on agriculture. Severe criminal penalties may also act as a deterrent. The 1989 Biological Weapons Antiterrorism Act prohibits the development, production, and stockpiling of biological agents for use as a weapon, and it explicitly applies to anti-plant and anti-animal agents. Violation incurs penalties up to life imprisonment. Nevertheless, no measures, singly or in combination, can eliminate the threat. And a bioterrorist attack on the agricultural sector, because of its relative ease, safety, and minimal technical requirements, is probably less likely to be deterred than an attack on human targets. Since agricultural bioterrorist attacks cannot be prevented altogether, an effective response plan to minimize the effects is essential. Knowledge that the United States can respond quickly and effectively to terrorist attacks, and can minimize their impact, would itself serve as an additional deterrent.
An effective response to an agricultural bioterrorist attack is in principle no different from effective response to a natural introduction of exotic diseases. The differences are largely quantitative: A bioterrorist attack is more likely to be multifocal, and it is more likely to begin explosively (because larger numbers of pathogens could be initially involved). These considerations suggest that at a minimum, existing response strategies should be evaluated and improved to deal with the more demanding outbreak situations that would most likely result from a bioterrorist attack.
However, there is good reason to question whether our existing response capabilities are adequate even to deal effectively with fully natural disease introductions. The experience of the United Kingdom in 2000 and Taiwan in 1997 with FMD, or the Netherlands with classical swine fever in 1997–1998, shows that even developed countries with advanced agricultural health services can be overwhelmed by some outbreaks. This suggests that more radical changes in the way we approach outbreak control may be necessary.
Effective preparation for a bioterrorist attack has several components. Probably most important is early detection. However, US farmers, veterinarians, plant pathologists, and agricultural extension agents are generally not well prepared to rapidly identify exotic animal and plant diseases. Thus a significant educational task confronts us. It is hard to overestimate the importance of that task; even a couple of days' delay in identifying an exotic animal disease can mean the difference between an easily controllable outbreak and one that escalates out of control because of rapid transmission. The UK FMD outbreak is thought to have been as serious as it was largely because of the failure to identify infected sheep for more than a week, during which time the sheep were transported and the disease spread throughout the country (DEFRA 2001, Ferguson et al. 2001a, 2001b).
For crop diseases, there is an additional problem: Crops are grown over millions of acres, and there is no way of carefully observing a very large proportion of individual plants. The first plants with symptoms typically are observed only after substantial spread has already occurred; 0.1% or more of the plants in an area may need to be infected before symptoms are first noticed (Campbell and Madden 1990). This may be too late for successful eradication, especially for highly contagious diseases such as rusts, which have spores that travel long distances. Long-term efforts are needed to develop strategies and technologies to reduce the time to discovery.
Confirmation of a diagnosis of most of the diseases of concern is done with accurate, sensitive molecular techniques, but samples may have to be shipped across the country, consequently delaying confirmation of a diagnosis for days. Expanded local capacity, at least at the state or regional level, to diagnose relevant exotic diseases is therefore important. Complicating the problem, diagnostic laboratories for plant pests (diseases and insects) are typically run by land grant universities and state departments of agriculture. These are often underfunded and understaffed, and they may not have the facilities or supplies to run molecular or biochemical assays or to provide rapid turnaround times. These labs need to be better supported to deal quickly with exotic pathogens.
Beyond the need to expand local capacity is the necessity that diagnostic technology be able to detect diseased animals or plants before they are symptomatic or contagious. This will be especially important after a disease outbreak is established and intensive containment and eradication efforts are being pursued. This would allow earlier culling of infected herds or fields, thereby greatly limiting pathogen transmission. It would also allow the culling of exposed herds or fields to be delayed until there was evidence of infection, since there would then still be time for culling before further disease transmission occurred. These measures could dramatically reduce unnecessary culling and thus reduce containment costs. Such technology has been developed for FMD, for example, but it is not yet widely available.
Diagnosis of presymptomatic plants is practical only for systemic diseases, those in which the pathogen is transported throughout the plant by the vascular tissue. Many threatening plant pathogens, however, are not systemic, and infection is localized; in these cases, tests on samples of plant material will not reliably detect infection. This limitation to presymptomatic diagnostic testing has hindered eradication efforts for citrus canker (Gottwald et al. 2001, Schubert et al. 2001), a bacterial disease discovered in Florida in 1995 that has global quarantine significance.
Rapid and accurate diagnosis is the cornerstone of effective control, but comprehensive planning for response is also required. Many of the decisions about control strategies can be (and should be) made in advance, so that action can be taken immediately upon notification of an outbreak. Response plans obviously will be specific for each disease of concern. For animals, these will be almost exclusively diseases on List A of the Office International des Epizooties's (OIE 2000). However, for crops, there is no worldwide consensus on the most threatening pathogens that could be used as biological weapons (MacKenzie 1999, Schaad et al. 1999, Madden 2001). Developing a consensus for a list of the major bioterrorist threats is thus the first priority in protecting crops. Such a list is necessary to guide the development of surveillance plans, diagnostic tests, and response plans for best containing and eradicating an introduced pathogen.
Probably the most important technical development for animal disease control would be to develop effective vaccines for all diseases of concern. FMD vaccines, for instance, are each protective against only one of the various strains of FMD virus, and they give only limited protection, requiring revaccination every six months or so (Ferguson 2001a, 2001b). A polyvalent, long-lasting vaccine could provide valuable control options. Vaccines also need to be designed such that a vaccinated animal can be reliably distinguished from a previously infected animal, because seriological evidence is used to document disease-free status for the purpose of international trade. These vaccines could be donated to international efforts for disease control, thereby keeping stockpiles renewed, production capacity busy, and the risk of importation of disease low.
Control strategies for crop diseases depend on the epidemiology of the particular disease (Vanderplank 1963, Fry 1982, Madden and van den Bosch 2002) and on the cropping system. For field crops such as wheat, for example, breeding for resistance is a fundamental approach for control of many diseases. However, breeding is done only for endemic diseases, because it is too expensive to develop resistant varieties for pathogens that are not present. Clearly, efforts must be made to at least identify sources of resistance for threatening pathogens so that the time it takes to develop new varieties is reduced. Genomics should speed this process considerably through the eventual development of plant cultivars that have general resistance to multiple plant pathogens.
Genomic technologies should also facilitate the development of a new generation of pesticides that combine high specificity, high effectiveness, and low environmental and health risks (for a discussion of genomics and drug discovery, see Wheelis 2002). Because plant disease control will very likely continue to rely heavily on pesticide use, substantial research and development efforts are warranted, including genome sequencing of important current and potential pests and their hosts.
The United States currently responds to large outbreaks of agricultural disease by deploying teams of specialists to assist in the diagnosis, containment, and eradication of the disease (APHIS 1998, USDA 2002). There are too few of these teams in the United States to effectively control a large or multifocal outbreak of highly contagious disease (as would probably be the case with an intentional disease introduction). Having more of these teams would increase our capability to respond quickly and effectively to large disease outbreaks, whether the outbreak is intentionally or unintentionally caused.
Because the United States normally has few outbreaks of disease that require such a response, these teams could be deployed internationally to combat disease outbreaks. This international deployment would have several benefits. The members of these teams would gain valuable experience in the diagnosis and containment of diseases that do not often occur in the United States, experience that they could not otherwise acquire. These internationally deployed professional teams could limit the extent of serious animal diseases in other countries, thereby diminishing the chances of an accidental introduction of that disease into the United States and minimizing the opportunities for a terrorist to obtain the pathogen from the environment. Plant diseases, for reasons discussed above, are less likely to be eradicated or contained globally, and thus the benefits of international deployment of specialists may be less obvious for them. Even so, the experience the teams gain would be valuable in itself, and the humanitarian benefit to developing countries would bring valuable goodwill.

Conclusion

Despite our best efforts, this country will continue to be vulnerable to deliberate introductions of exotic plant and animal diseases by terrorist groups with an ideological agenda or by governments, corporations, or individuals with a profit motive. The vulnerability to agricultural bioterrorist attack is a consequence of the intrinsically low security of agricultural targets, the technical ease of introducing consequential diseases, and the large economic repercussions of even small outbreaks. It is exacerbated by structural features of US agriculture that are unlikely to change without forceful government intervention: low genetic diversity of plants and animals, extensive monoculture, and highly concentrated animal husbandry.
While the vulnerability cannot be eliminated, effective response can minimize the damage from both intentionally and naturally introduced disease. We have suggested an aggressive scientific agenda: continuing education programs for farmers, veterinarians, and extension specialists; development of new diagnostics, vaccines, and pesticides; development of new sensing technologies for early identification of plant disease outbreaks; development of plant varieties resistant to diseases not yet endemic; and an increase in the number of outbreak control specialists assigned to international disease control efforts. This is an expensive agenda, but cost-effective in context—a single serious outbreak prevented or quickly controlled could pay for the program several times over. Given the ever-increasing international traffic in agricultural commodities combined with decreasing transit times, we can expect continued natural introductions of exotic plant and animal diseases. These will easily justify the cost of the programs that we recommend. Clearly, aggressive action is warranted to address the deficiencies of our current response system, for both naturally and deliberately introduced plant and animal diseases, given the billions of dollars at stake.

References cited

AgNet
.
2001
.
Karnal bunt in the news again
.
Canadian Phytopathological Society
 . (20 May 2002; www.cps-scp.ca/karnalbunt.htm).
Agrios

G. N.
.
1997
.
Plant Pathology
 .
4th ed.

New York
:
Academic Press
.
[APHIS] Animal and Plant Health Inspection Service, US Department of Agriculture
.
1998
.
APHIS' role in animal health and trade
 . USDA APHIS Factsheet, August. (20 May 2002; www.aphis.usda.gov/oa/pubs/fsprotect.html).
[APHIS] Animal and Plant Health Inspection Service, US Department of Agriculture
.
2002
.
Asian longhorned beetle (Anoplophora glabripennis)
. USDA APHIS Factsheet, January. (20 May 2002; www.aphis.usda.gov/oa/pubs/fsalb.html).
[ARS] Agricultural Research Service, US Department of Agriculture
.
2002
.
ARS National Programs, Animal Health
 . (20 May 2002; http://nps.ars.usda.gov/programs/programs.htm?npnumber=103).
Bandyopadhyay

R.

Frederiksen

R. A.
.
1999
.
Contemporary global movement of emerging plant diseases
. Pages
28
–
36
. in Frazier TW, Richardson DC, eds.
Food and Agricultural Security: Guarding against Natural Threats and Terrorist Attacks Affecting Health, National Food Supplies, and Agricultural Economics
 .
New York
:
New York Academy of Sciences
.
Bohan

P.
.
2001
.
GMO crops here to stay or gone with the wind?
. Reuters Online, 5 November. Document HA2001115130000011. (6 June 2002; http://special.northernlight.com/gmfoods/gone_with_the_wind.htm).
Brown

C. C.

Slenning

B. D.
.
1996
.
Impact and risk of foreign animal diseases
.
JAVMA
 
208
:
1038
–
1040
.
Google Scholar
PubMed
Campbell

C. L.

Madden

L. V.
.
1990
.
Introduction to Plant Disease Epidemiology
 .
New York
:
John Wiley and Sons
.
Casagrande

R.
.
2000
.
Biological terrorism targeted at agriculture: The threat to US national security
.
Nonproliferation Review
 
7
:
98
–
99
.
Google Scholar
CrossRef
Chen

X.

Line

R. F.

Leung

H.
.
1995
.
Virulence and polymorphic DNA relationships of Puccinia striiformisf. sp. hordei to other rusts
.
Phytopathology
 
85
:
1335
–
1342
.
Google Scholar
CrossRef
Condon

M.
.
1997
.
Implications of plant pathogens to international trading of seeds
. Pages
17
–
30
 in
McGee

DC
, ed.
Plant Pathogens and the Worldwide Movement of Seeds
 .
St. Paul (MN)
:
APS Press
.
Davis

S. H.
.
1977
.
Victims of the Miracle: Development and the Indians of Brazil
 .
New York
:
Cambridge University Press
.
[DEFRA] Department for Environment, Food, and Rural Affairs
.
2001
.
How the 2001 outbreak of foot and mouth began
 . DEFRA Factsheet, 1 May. (28 May 2002; www.defra.gov.uk/animalh/diseases/fmd/about/current/source.asp).
Ekboir

J. M.
.
1999
.
Potential Impact of Foot-and-Mouth-Disease in California: The Role and Contribution of Animal Health Surveillance and Monitoring Services
 .
Davis
:
University of California, Agricultural Issues Center
.
Elsner

A.
.
2001
.
PETA cheers disease spread
 . Boston Globe, 3 April, p.
A5
.
Falkenrath

R.

Newman

R.

Thayer

B.
.
1998
.
America's Achilles' Heel
 .
Cambridge (MA)
:
MIT Press
.
[FAS] Foreign Agricultural Service, US Department of Agriculture
.
1997
.
Foot-and-mouth disease spreads chaos in pork markets
 . FASonline, October. (20 May 2002; www.fas.usda.gov/dlp2/circular/1997/97-10LP/taiwanfmd.htm).
[FAS] Foreign Agricultural Service, US Department of Agriculture
.
2001
.
The World Trade Organization and U.S. agriculture
 . FASonline, Fact Sheet, March. (20 May 2002; www.fas.usda.gov/info/factsheets/wto.html).
Ferguson

N. M.

Donnelly

C. A.

Anderson

R. M.
.
2001a
.
Transmission intensity and impact of control policies on the foot-and-mouth epidemic in Great Britain
.
Nature
 
413
:
542
–
548
.
Google Scholar
CrossRef
Ferguson

N. M.

Donnelly

C. A.

Anderson

R. M.
.
2001b
.
The foot-and-mouth epidemic in Great Britain: Pattern of spread and impact of interventions
.
Science
 
292
:
1155
–
1160
.
Google Scholar
CrossRef
Frazier

T. W.

Richardson

D. C.
. eds.
1999
.
Food and Agricultural Security: Guarding against Natural Threats and Terrorist Attacks Affecting Health, National Food Supplies, and Agricultural Economics
 .
New York
:
New York Academy of Sciences
.
Fry

W. E.
.
1982
.
Principles of Plant Disease Management
 .
New York
:
Aca demic Press
.
Gottwald

T. R.

Hughes

G.

Graham

J. H.

Sun

X.

Riley

T.
.
2001
.
The citrus canker epidemic in Florida: The scientific basis of regulatory eradication policy for an invasive species
.
Phytopathology
 
91
:
30
–
34
.
Google Scholar
CrossRef
PubMed
Greiser-Wilke

I.

Fritzemeier

J.

Koenen

F.

Vanderhallen

H.

Rutili

D.

De Mia

G.

Romero

L.

Rosell

R.
Sanchez-Vizcaino

J.

San Gabriel

A.
.
2000
.
Molecular epidemiology of a large classical swine fever epidemic in the European Union in 1997–1998
.
Veterinary Microbiology
 
77
:
17
–
27
.
Google Scholar
CrossRef
PubMed
Horn

F. P.

Breeze

R. G.
.
1999
.
Agriculture and food security
. Pages
9
–
17
. in Frazier TW, Richardson DC, eds.
Food and Agricultural Security: Guarding against Natural Threats and Terrorist Attacks Affecting Health, National Food Supplies, and Agricultural Economics
 .
New York
:
New York Academy of Sciences
.
Greenstone

M. H.
.
2001
.
GMOs: Tree hackers, bathwater, and the free lunch
.
BioScience
 
51
:
899
.
Google Scholar
CrossRef
James

C.
.
2001
.
Global review of commercialized transgenic crops: 2001
.
ISAAA Briefs
 
24
. Preview. (17 May 2001; www.isaaa.org/publications/briefs/Brief_24.htm).
James

W. C.

Teng

P. S.

Nutter

F. W.
.
1991
.
Estimated losses of crops from plant pathogens
. Pages
15
–
51
. in
Pimentel

D
, ed.
CRC Handbook of Pest Management in Agriculture
 .
Boca Raton (FL)
:
CRC Press
.
Jelsma

M.
.
2001
.
Vicious Circle: The Chemical and Biological ‘War on Drugs.’
 . Amsterdam: Transnational Institute. (17 May 2001; www.tni.org/drugs/).
Kleiner

K.
.
1999
.
Operation eradicate
.
New Scientist
 
163
(2203)
:
20
.
MacKenzie

D.
.
1999
.
Run, radish, run
.
New Scientist
 
164
(2217)
:
36
–
39
.
Madden

L. V.
.
2001
.
What are the nonindigenous plant pathogens that threaten U.S. crops and forests?
 . APSnet feature. (17 May 2002; www.apsnet.org/online/feature/exotic/).
Madden

L. V.

Nutter

F. W.
.
1995
.
Modeling crop losses at the field scale
.
Canadian Journal of Plant Pathology
 
17
:
124
–
137
.
Google Scholar
CrossRef
Madden

L. V.

van den Bosch

F.
.
2002
.
A population-dynamic approach to assess the threat of plant pathogens as biological weapons against annual crops
.
BioScience
 
52
:
65
–
74
.
Google Scholar
CrossRef
Madden

L. V.

Jeger

M. J.

van den Bosch

F.
.
2000
.
A theoretical assessment of the effects of vector-virus transmission mechanism on plant virus disease epidemics
.
Phytopathology
 
90
:
576
–
594
.
Google Scholar
CrossRef
PubMed
Matsubara

K.
.
2000
.
Final Eradication of Foot and Mouth Disease in Japan
 .
Tokyo
:
Ministry of Agriculture
.
[OIE] Office International des Epizooties
.
2000
.
OIE classification of diseases
 . (20 May 2002; www.oie.int/eng/maladies/en_classification.htm).
Pimentel

D.

Lach

L.

Zuniga

R.

Morrison

D.
.
2000
.
Environmental and economic costs of nonindigenous species in the United States
.
BioScience
 
50
:
53
–
65
.
Google Scholar
CrossRef
Schaad

N. W.

Shaw

J. J.

Vidaver

A.

Leach

J.

Erlick

B. J.
.
1999
.
Crop Biosecurity
 . APSnet feature. (20 May 2002; www.apsnet.org/online/feature/Biosecurity/Top.html).
Schubert

T. S.

Rizvi

S. A.

Sun

X.

Gottwald

T. R.

Graham

J. H.

Dixon

W. N.
.
2001
.
Meeting the challenge of eradicating citrus canker in Florida—again
.
Plant Disease
 
85
:
340
–
356
.
Google Scholar
CrossRef
Stone

R.
.
2000
.
Experts call fungus threat poppycock
.
Science
 
290
:
246
.
Google Scholar
CrossRef
PubMed
Strange

R. N.
.
1993
.
Plant Disease Control
 .
London
:
Chapman and Hall
.
[USDA] US Department of Agriculture
.
1997
.
Profile of the nation's agriculture
 . (14 June 2002; www.nass.usda.gov/census/census97/volume1/us-51/us1figs.pdf).
[USDA] US Department of Agriculture
.
2002
.
Emergency Programs Manual
 . Washington (DC): USDA, APHIS, Plant Protection and Quarantine. (20 May 2002; www.aphis.usda.gov/ppq/emergencyprograms/manual.pdf).
Vanderplank

J. E.
.
1963
.
Plant Diseases: Epidemics and Control
 .
New York
:
Academic Press
.
Watts

J.
.
2001
.
Japan's government tries to allay BSE fears
.
Lancet
 
358
:
2057
.
Google Scholar
CrossRef
PubMed
Wheelis

M.
.
1999a
.
Biological sabotage in World War I
. Pages
35
–
62
. in Geissler E, Moon JEvC, eds.
Biological and Toxin Weapons: Research, Development and Use from the Middle Ages to 1945
 .
Oxford (UK)
:
Oxford University Press
.
Wheelis

M.
.
1999b
.
Outbreaks of Disease: Current Official Reporting
 . Bradford (UK): University of Bradford, Department of Peace Studies. Briefing Paper No. 21. (20 May 2002; www.brad.ac.uk/acad/sbtwc/briefing/bp21.htm).
Wheelis

M.
.
2000
.
Agricultural Biowarfare and Bioterrorism
 . Edmonds (WA): Edmonds Institute. (20 May 2002; www.fas.org/bwc/agr/main.htm).
Wheelis

M.
.
2002
.
Biotechnology and biochemical weapons
.
Nonproliferation Review
 
9
:
48
–
53
.
Google Scholar
CrossRef
Whitby

S. M.
.
2002
.
Biological Warfare against Crops
 .
Hampshire (UK)
:
Palgrave
.
Whitby

S.

Rogers

P.
.
1997
.
Anti-crop biological warfare—implications of the Iraqi and US programs
.
Defense Analysis
 
13
:
303
–
318
.
Google Scholar
CrossRef
White House
.
1998
.
Combating terrorism: Presidential Decision Directive 62
 . Critical Infrastructure Assurance Office, Fact Sheet 62. (20 May 2002; www.info-sec.com/ciao/62factsheet.html).
Wilson

T. M.

Logan-Henfrey

L.

Weller

R.

Kellman

B.
.
2000
.
A review of agroterrorism, biological crimes, and biological warfare targeting animal agriculture
. Pages
23
–
57
. in Brown C, Bolin CA.
Emerging Diseases of Animals
 .
Washington (DC)
:
ASM Press
.
Zadoks

J. C.

Schein

R. D.
.
1979
.
Epidemiology and Plant Disease Management
 .
New York
:
Oxford University Press
.

Author notes

Mark Wheelis (email: mlwheelis@ucdavis.edu) works in the Section of Microbiology, University of California, Davis, CA 95616. His research interests are in the history of biological warfare, especially in the First World War, and the scientific aspects of biological and chemical arms control.
Rocco Casagrande (email: rcasagrande@surfacelogix.com) works for Surface Logix, Inc., Brighton, MA 02135. His research interests are in the development and testing of devices to detect and analyze biological weapons and in biological defense policy.
Laurence V. Madden (email: MADDEN.1@osu.edu) is with the Department of Plant Pathology, Ohio State University, Wooster, OH 44691. His research interests are in epidemiology of plant diseases, modeling and analysis of epidemics, and the prediction of disease outbreaks and crop losses.
© 2002 American Institute of Biological Sciences