Mark Wheelis

Section of Microbiology, University of California, Davis



What goals might an attack on the agricultural sector serve?


Attack the food supply of an enemy belligerent


This is the classical rationale for the inclusion of anti-plant programs in national biological weapons programs. Every major state BW program we know of has included an anti-agricultural component, from the WWI German use of anthrax and glanders against animals to the Iraqi program on wheat cover smut. For most agents, effective use would require large stockpiles and extensive delivery efforts; however, there is potential for delivery by secret agent to initiate point-source epidemics of highly contagious agents.


The relevance of this to terrorist use is slight; however, BW proliferator states are likely to include anti-animal and anti-plant agents in their developing arsenal, and if they also support terrorist groups, there is a slim chance that such groups might be allowed access to anti-agricultural agents for the purpose of bioterrorism.


Destabilize a government by initiating food shortages or unemployment


Disruption of the agricultural sector can cause profound dislocation of societies. Direct losses of plants or animals could cause food shortages, rises in food prices, and unemployment. All of these could, if severe, have serious destabilizing effects on social and political structures. Many developed countries are quite vulnerable to disruption of the agricultural sector, although their social and political institutions are fairly robust and the resulting discontent is probably unlikely to cause institutional collapse. Nevertheless, the potential for immense economic damage is high in a well-planned attack, and the consequences for the food supply, export trade, and financial markets could be very serious. Many developing countries are potentially quite vulnerable to such destabilization, particularly if they depend heavily on a single food crop or animal.


Alter supply and demand patterns for a commodity


A widespread-epidemic, or any outbreak that triggered the imposition or relaxation of trade restrictions, could result in significant changes of supply of the affected plant or animal materials on domestic and international markets. This in turn would open up or close markets for others (a possible motivating factor).  Biological attack could also be used to manipulate futures, and for other manipulations of the financial markets.


Control an undesirable plant or animal (biocontrol)


The use of legitimate, peaceful biocontrol is expanding steadily, and provides an unfortunate body of knowledge and range of ready-made delivery technologies for the interested agricultural bioterrorist or biowarfare program.


There have been two recent programs to develop pathogens of drug crops as biocontrol agents. These have been conducted under UNDCP auspices, funded and performed by the US (fungal pathogens of coca), and funded by the US and UK and performed by Uzbekistan (fungal pathogens of poppy). Both of these programs involve the development of biological agents and delivery devices, and both are presumably intended for use principally or entirely in other countries. However, none of the potential target states has agreed to allow the use of these agents for biocontrol, and several have now stated explicitly that they will not allow it. As a result of this refusal, UNDCP has now withdrawn its sponsorship of the anti-coca project. Although there is no evidence that the agents are being developed for hostile use, the absence of target country approval makes it equally difficult to demonstrate that they are being developed for peaceful purposes. This ambiguity raises legitimate concerns about compliance with Article I of the BTWC. Furthermore, once effective agents have been developed, the intense concern over the drug trade in drug-consuming states may lead to pressure to use them covertly, regardless of target country approval.


Terrorists and individuals might also be interested in biocontrol agents. The deliberate and illegal 1997 importation of Rabbit Hemorrhagic Disease Virus (RHDV) into New Zealand constitutes a past instance. And biological attack on genetically engineered plants might be considered by the more extreme activist groups opposed to such technology, a use that could in their view be a form of biocontol.

Who might be tempted to initiate an attack on the agricultural sector?




Countries might consider agricultural attack for military, political, ideological, or economic reasons. Since there could be quite severe consequences of being recognized as responsible for a biological attack, such efforts would likely be covert. This would entail an effort to make the outbreak appear natural—most probably a point-source outbreak, or multiple outbreaks with an apprently natural common source (see below).


Countries that are actively pursuing a secret military BW capability (thought to number about a dozen) are probably developing anti-agricultural agents for strategic use in the event of war. Iraq, for instance, was developing wheat cover smut as a weapon, presumably intended against Iran.


In the 1980’s, Iraq used chemical weapons extensively against Iran, and internally against civilian minorities, with virtually no political consequences. This has undoubtedly lowered the political threshold for use of BW in a regional or civil conflict.




Agricultural corporations, including producers, processors, and shippers, could benefit immensely from the economic impacts, market share changes, and financial market effects of a successful biological attack. Many also employ expert plant pathologists or veterinarians and have large collections of pathogens. The combination of motivation, expertise, and materials within a single, closed organization is worrisome. Of course, corporations, like countries, would run enormous legal risks if they perpetrated a biological attack, so if they were to choose to do this, it would be expertly designed to mimic a natural outbreak.


For both corporations and governments, decision to use bioweapons would be expected to require approval at the very highest level, thus reducing its likelihood. However, in both the possibility of mid- or lower-level zealots initiating unauthorized action has to be considered.


Organized crime


Because of the huge financial stakes in the agricultural sector, and because the foundation of the drug industry involves crop cultivation, organized crime may take an interest in biocriminal activities with agricultural targets.


Terrorist groups


Terrorist groups might be interested in agricultural bioweapons for a variety of reasons: international terrorist organizations for the harm they could do to enemy states or peoples, millennial groups for their potential contribution to societal collapse, anti-GMO groups for their potential value in deterring farmers from the use of genetically engineered crops or animals. In many cases of ideologically-motivated terrorist attack there would be willing assumption of responsibility by the perpetrator; in other cases there could be an attempt to disguise the outbreak as natural.




Individual perpetrators (biocriminals) could include disgruntled employees or ex-employees in the agricultural sector, ideologically motivated individuals, speculators on the commodities market, or individuals with a profit motive (such as the New Zealand farmer(s) assumed to have covertly imported RHDV).

What would be the consequences of attack on the agricultural sector?


Direct losses due to disease


Direct financial loss due to mortality or morbidity of domestic animals or crop plants can vary from insignificant to catastrophic. In many cases the direct losses would be modest and would fall on a small number of farms. One of the major determinants of the magnitude of the direct losses will be the rapidity with which the disease is noticed and diagnosed. In developed countries most of the foreign diseases of greatest concern would likely be identified fairly early, allowing the direct disease losses to be kept modest.


Losses due to efforts to contain outbreaks


The control of an outbreak of an imported, highly contagious animal or plant disease is routinely controlled by destruction of all potentially exposed healthy host organisms. With animal diseases, this normally means the slaughter of all host animals in the immediate vicinity. With plants, thousands of acres of crop plants may have to be destroyed to contain an outbreak. Thus the losses attendant on outbreak control can exceed, often by several orders of magnitude, the direct losses due to the disease itself.


Destruction of exposed hosts is often the only option when the agent is bacterial or viral. However, for fungal agents destruction of exposed crops may be reduced by the use of fungicides. However, this is an expensive process itself, so it adds significantly to the cost of the outbreak, and it may cause environmental damage.


A number of important threats to crop plants are from insect pests, rather than microbial pathogens. These outbreaks are usually controlled by use of pesticides rather than destruction of exposed plants, which, as with control of fungal disease, can cost large amounts. Widespread broadcast of insecticide may also cause environmental or human health damage as well. The BTWC certainly could cover pests as well as pathogens, as Article I refers to “microbial or other biological agents,” and the consultative process of the BTWC has been used to address concerns about a pest infestation in Cuba. However, this coverage has never been made explicit, and it would be useful to do so since there are so many insect pests of great potential for agricultural biowarfare or bioterrorism.


Losses due to sanitary or phytosanitary restrictions on international trade


Under the World Trade Organization, member states are allowed to impose import restrictions on agricultural products to prevent the importation of pests or disease agents. Thus importing countries free of a particular disease are usually quick to block imports from countries in which that disease breaks out. This happens frequently, as these diseases periodically resurface in areas that they have been absent from; trade restrictions typically last a month or two when control of the outbreak is rapid, or they may endure much longer if disease control is slow and difficult (eg the EU restriction on the import of UK beef due to the BSE outbreak).


Thus major agricultural exporters are particularly vulnerable. For instance, the Taiwan FMD outbreak in swine in 1997 probably only cost tens of thousands of dollars (US) in direct losses, but it cost 4 billion in eradication and disinfection costs, and a cumulative 15 billion in lost export revenues. An FMD outbreak in Italy in 1993 again had trivial direct costs, but nearly 12 million dollars in eradication and disinfection costs, and 120 million dollars in lost trade revenues.


Alternatively, the introduction of a disease into a country previously free of it would undermine the legitimacy of that country’s import restrictions under the WTO, forcing the lifting of the restrictions and opening up the market. This could bring significant additional losses to domestic producers.


Losses due to indirect effects (market destabilization, etc)


The substantial market effects of a widespread outbreak, or one that has major impacts on international trade, could have secondary effects, such as share-holder losses, revenue losses to processors and shippers, etc. In the extreme, if losses are very large and if future losses appear likely, significant levels of investor panic could lead to market destabilization.

What are the special features of attack on the agricultural sector?


Agents are not hazardous to perpetrators


With the exception of a few agents of zoonotic disease, most of the diseases that are likely to be considered for an attack on the agricultural sector are completely harmless to humans. They are thus much less challenging to produce, stockpile, and disseminate than lethal human pathogens.


Few technical obstacles to weaponization


A military style attack by airplane on large acreage of crops would require crop dusters and large stockpiles of agent. Nevertheless, nothing would be difficult to obtain on the open market. Less ambitious attacks would require much less in the way of equipment or agent stockpiles. If the goal is to cause only a few cases in order to disrupt trade, then no special equipment and only a few microliters of agent are needed. And, as discussed below, it is possible to introduce biological agents without even entering the target country.


Low security of vulnerable targets


Many potential sites for release of an animal agent, such as auction houses, have very low security. Access to large numbers of animals with destinations all over a country or region is simple and easy. Seeds, fertilizers, and pesticides provide routes for infection of crop plants, although of somewhat higher (but still not robust) security. And of course pastures and fields themselves have essentially no security at all.


Lower moral barrier to cross


It is often argued that there is an innate human revulsion to the use of disease as a weapon; if so, this could constitute an important disincentive to bioterrorism and biowarfare. However, it is unlikely that this sentiment extends to biological attack on plants or animals. Furthermore, the response after a biological attack on plants or animals would be less substantial than if the attack involved human victims, and the penalties of being identified as the perpetrator would be lighter.


Maximum effect may not require many cases


If the goal is to disrupt trade by introducing a highly contagious disease into territory from which it is absent, then the attack does not have to be constructed to cause a large number of cases—a handful of cases may be sufficient. Obviously it is much easier to cause a small outbreak than a large one


Point source to mimic natural introduction can be effective


Because of the high background of naturally-occurring disease, it is possible that a deliberately instigated outbreak could be mistaken for a natural one. If avoiding detection is important, an attack would be constructed to take advantage of this confusion. Especially if the goal is disruption of international trade, where few cases are necessary, it is feasible to construct an attack to appear to be a natural point-source outbreak.


Multiple point source outbreaks can be initiated by contaminating imported feed or fertilizer, without even entering the country


Many countries import materials such as straw, animal feed, or fertilizer. This provides an opportunity for introducing serious pathogens, without having to even enter the target country. It also allows the possibility of initiating multiple outbreaks over a large geographic area, in a way that mimics a natural event (such as the recent outbreak of foot and mouth disease in Japanese cattle in two widely separated prefectures, thought to have been introduced on straw imported from China).

How will Genomics and genEtic technology change the threat?


Genomics and proteomics make genotype-specific weapons possible.


The emerging sciences of genomics and proteomics are already beginning to transform biology and medicine. This will continue, at an accelerating pace. One outcome will be the possibility of constructing genotype-specific biological control or weapons agents. Although there has been much concern about the possibility of “ethnic weapons” targeting specific groups of humans, the likelihood is that this will prove impossible (due to the high and increasing amount of intragroup genetic heterogeneity).  However, equivalent weapons targeting specific agricultural varieties is a very real possibility.


Agriculture is highly vulnerable to genotype-specific weapons


Agriculture, particularly in many developed countries, has several properties that make it vulnerable to attack with genotype-specific weapons. Typically it employs monocropping of large acreages with genetically identical cultivars, and high-density husbandry of genetically inbred animal strains. These agronomic practices reduce the genetic variability that makes populations resistant to genotype-specific weapons, and it creates conditions (large, dense populations) that facilitate disease spread.


High-tech agent design is an option available only to sophisticated players—eg states and multinational corporations.


At least for the near future, the construction of novel toxin weapons or genetically engineered microbes as genotype-specific weapons would require a substantial scientific infrastructure, ranging from molecular biology to genomics to agronomy. It is unlikely that most terrorist groups could assemble the necessary expertise and materials for this. Thus genotype-specific weapons are likely to remain accessible only to states (and possibly their sub-state clients), and to large agricultural corporations.



The states most vulnerable to economic attack on the agricultural sector are those with several or most of the following attributes:


--high-density, large area agriculture

--heavy reliance on monoculture of a restricted range of genotypes

--free of specific serious animal and plant pathogens or pests

--major agricultural exporter, or heavily dependent on a few domestic agricultural products

--suffering serious domestic unrest, or the target of international terrorism, or unfriendly neighbor of states likely to be developing BW programs

--weak plant and animal epidemiological infrastructure


For such at-risk states, the threat of biological attack against its agricultural sector should be taken quite seriously, and preventative and punitive measures put in place.

How might attack on the agricultural sector be deterred?


Enact appropriate legislation


Enactment of legislation implementing the BTWC is required of all States Parties; however, many have not yet done so. Such legislation can be a significant deterrent to biological attack on the agricultural sector. The legislation should, among other provisions, provide for substantial criminal penalties for the hostile use anywhere of biological agents against plants or animals as well as people, and it should provide for extradition for anyone charged with using such agents against the agricultural sector of another state.


States that already have enacted such legislation should review its provisions to ensure that it adequately covers biological attack on plants and animals.


Insure effective epidemiological investigation to determine origin of outbreaks


Biological attack on the agricultural sector is likely to be covert if one ever happens. Such attacks will be options for perpetrators only to the extent that they maintain plausibility as natural events. Increased epidemiological capacity, especially in strain identification from molecular sequence data, makes it increasingly difficult to escape detection, and thus would act as a substantial deterrent.


Negotiate an effective BTWC Protocol


A BTWC Protocol that establishes effective measures to deter  States from developing or possessing biological weapons would provide a powerful tool in making progress towards the goal of complete biological disarmament. This would reduce the likelihood of BW in regional conflicts, and the chance that state-supported terrorist organizations would ever get bioweapons. Provision for internationally sponsored epidemiological investigation of possible agricultural attacks would deter covert use in the same manner as national epidemiological capacity.


Reduce reliance on monoculture and expand the diversity of genotypes cultured


States that engage in high intensity agriculture of a limited range of varieties could reduce their vulnerability to both deliberate and natural disease outbreaks by increasing the use of intercropping, expanding the diversity of genotypes utilized, reducing the size of plots, and a variety of other agricultural changes designed to reduce susceptibility to disease outbreaks. However, these constitute substantial changes in established practice, and are probably not likely to be instituted without sustained and forceful political leadership.



The analysis suggests the following recommendations for consideration by the Review Conference:


With respect to Article I


1. The RC should reaffirm that the general purpose prohibitions of Article I apply to plant and animal diseases, and to all agents and toxins regardless of their means of production, including artificial toxic analogues of natural transmitters, modulators, etc, generated by genomic or other methods.


2. The RC should affirm that insect pests are covered by the BTWC.


3. The RC should affirm that the development of biocontrol agents by a State Party, intended for use within its borders is a permitted peaceful use. It should consider whether the same is true if the only possible use is in other states and the developing state does not have the explicit approval of the country in which they are intended for use.


With respect to Article IV


4. The RC should urge all States Parties that have not yet passed implementing legislation to do so; and all States Parties with implementing legislation in place to review it to insure that it adequately covers biological attack on the agricultural sector.


With respect to Article IX


5. The RC should recognize that novel chemical weapons will be made possible by the revolutions in genomics and proteomics, and should communicate this concern by appropriate means to the first RC of the CWC.


6. The RC should affirm that the General Purpose criteria expressed in Article I in both the BTWC and the CWC between them effectively cover every conceivable chemical compound and microbial agent. The RC should invite the First RC of the CWC to join them in this affirmation, and to jointly devise a mechanism of cooperative jurisdiction in ambiguous cases.



Ban, J., 2000. Agricultural biological warfare: An overview. The Arena, Number 9, June 2000. Washington: Chemical and Biological Arms Control Institute. 8 pp.


Brown, C., 1999. Agro-terrorism: a cause for alarm. The Monitor: Nonproliferation, Demilitarization, and Arms Control. Winter-Spring 1999, pp 6-8.


Frazier, T. W., and D. C. Richardson (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 (Annals of the New York Academy of Sciences, Vol 894). 233 pp.


Gordon, J. C., and S. Bech-Nielsen, 1986. Biological terrorism: a direct threat to our livestock industry.  Military Medicine 151, 357-363.


Kohnen, A. S., 2000. Responding to the Threat of Agroterrorism: Specific Recommendations for the United States Department of Agriculture. BCSIA Discussion Paper 2000-29, ESDP Discussion Paper ESDP-2000-04, John F. Kennedy School of Government, Harvard University, October 2000. 43 pp. Available online at


Rogers, P., S. Whitby and M. Dando, 1999. Biological warfare against crops. Scientific American, June, pp 70-75.


Van der Plank, J. E., 1963. Plant diseases in biological warfare. Pp 212-222 in J. E. van der Plank, Plant Diseases: Epidemics and Control. New York: Academic Press.


Wheelis, M., 1999. Biological sabotage in world war I. Pp. 35-62 in E. Geissler and J. E. v. C. Moon (Eds.) Biological and Toxin Weapons: Research, Development and Use from the Middle Ages to 1945. Oxford: Oxford University Press.


Wilson, T. M., L. Logan-Henfrey, R. Weller, and B. Kellman. 2000. Agroterrorism, biological crimes, and biological warfare targeting animal agriculture, p. 23-57. In C. Brown and C. Bolin (eds.), Emerging Diseases of Animals. ASM Press, Washington, D.C.


Whitby, S. and P. Rogers, 1997. Anti-crop biological warfare—implications of the Iraqi and US programmes. Defense Analysis 13, 303-318.






First presented at the Annual Meeting of the Association of Politics and the Life Sciences in Atlanta, Georgia, September 1999; Expanded version presented as a working paper at the 14th Workshop of the Pugwash Study Group on the Implementation of the Chemical and Biological Weapons Conventions, Geneva, Switzerland, November 2000. Last update: November, 2000