Risks of Terrorism at Chemical Facilities
Nature of Hazards.
Potential terrorist acts against chemical facilities might be classified roughly into two categories: direct attacks on facilities or chemicals on site, or efforts to use business contacts, facilities, and materials (e.g., letterhead, telephones, computers, etc.) to gain access to potentially harmful materials. In either case, terrorists may be employees (saboteurs) or outsiders, acting alone or in collaboration with others. In the case of a direct attack, traditional or nontraditional weapons may be employed, including explosives, incendiary devices, firearms, airplanes, or computer programs.
In obtaining chemicals, a terrorist’s intent may be to use them as weapons or to make weapons, including explosives, incendiaries, poisons, and caustics. Access to chemicals might be gained by physically entering a facility and stealing supplies, or by using legitimate or fraudulent credentials (e.g., company stationary, order forms, computers, telephones or other resources) to order, receive, or distribute chemicals.
Risks of Chemical Terrorism.
The validity of any risk assessment depends on how much is known about the hazard, risks (probabilities), adverse effects, events and conditions that lead to or modify adverse effects or risks, and populations or environments that influence or experience adverse effects. The most accurate, and therefore the most useful, risk assessments generally are for familiar, frequently occurring hazards and events with impacts that are experienced with some regularity, such as severe storms or floods. In contrast, the risk of terrorist activity is unfamiliar (at least in the United States), rarely experienced, and likely to vary significantly over time, depending on rather unpredictable social and political phenomena.
The risk of terrorism targeting chemical facilities is particularly difficult to assess for at least three reasons:
- There are few prior examples of terrorists targeting chemical facilities;
- Numerous factors theoretically may increase or decrease risks; and
- Interactions among factors influencing risks are dynamic and changing.
In part, these difficulties stem from the nature of terrorism and the terrorists’ deliberate efforts to do what is least expected — that is, to defy prediction. For these reasons, most experts have not tried to quantify risks; existing analyses of chemical terrorism risks in the open literature are speculative and qualitative. Until the mid to late 1990s, reports focused on the acquisition and use of chemical weapons, such as sarin or mustard gas. One of the most comprehensive of these reports was a 1995 review of the open literature on terrorism that was prepared for the Canadian Security Intelligence Service. According to this review of the literature, “[t]hose authors who have speculated about the future terrorist use of chemical agents in particular have generally rated its likelihood as quite high.”
According to some, the risk also appears to be increasing. Many experts today believe that factors that might have inhibited proliferation and use of chemicals as weapons in the past are eroding. For example, some experts hypothesized several years ago that the combination of chemical and strategic skills necessary to create and deploy chemical weapons would prevent the lone terrorist from using them.
Security experts now believe that lack of personal expertise no longer limits chemical weapon use, because there is a tendency for terrorists with similar extreme views to affiliate loosely with others with complementary skills and abilities. Moreover, the rising level of education worldwide means that more people have the requisite training in chemical engineering, and the Internet has simplified communications, training, and cooperation within geographically dispersed terrorist groups.
Others have argued that chemical attacks would be unlikely, due to the difficulties of producing and effectively delivering chemical agents in sufficient amounts to produce mass casualties. However, while this may be true with regard to military use on a large scale, where weapons are delivered by advanced systems, it is not necessarily relevant to terrorists who may have more limited ambitions. A 1999 report by the U.S. General Accounting Office (GAO, now the Government Accountability Office) summarized the situation —
... many conflicting statements have been made in public testimony before Congress ... concerning the ease or difficulty with which terrorists could effectively disseminate a chemical or biological agent on U.S. soil and cause mass casualties.
GAO studied the threat and concluded that the ease or difficulty for terrorists to cause more than 1,000 casualties depends on the chemical or biological agent selected. The report stated —
Experts from the scientific, intelligence, and law enforcement communities told us that terrorists do not need sophisticated knowledge or dissemination methods to use toxic industrial chemicals such as chlorine. In contrast, terrorists would need a relatively high degree of sophistication to successfully cause mass casualties with some other chemical and most biological agents.
On the other hand,
“terrorists with less sophistication could make a chemical or biological weapon and disseminate agents, but these would be less likely to cause mass casualties.”
Other factors that might have inhibited chemical use by some terrorists in the past might not apply to loosely affiliated terrorist groups. For example, some experts argue that terrorists supported by nation-states have been reluctant to use chemical weapons for fear of offending other nations and neutral parties, particularly if the sponsors were signatories of the Chemical Weapons Convention. Another possible deterrent to chemical use, fear of retaliation, probably is of little concern to attackers with no identifiable homeland or headquarters. Lack of a homeland might also lessen concern about environmental damage that may be associated with chemical production. Finally, one must presume that occupational safety would be of limited concern to terrorists who are not accountable to a government, and who are willing to sacrifice their own lives for a religious, political, or social cause.
However, many experts believe that the relative risk of terrorism involving chemical weapons remains small. This point was stressed by John V. Parachini, a senior associate at the Center for Nonproliferation Studies, Monterey Institute of International Studies at a 1999 hearing before the U.S. House of Representatives, Committee on Government Reform, Subcommittee on National Security, Veterans Affairs, and International Relations. Referring to the risk of any use of chemical or biological weapons he stated:
... attacks with chemical and biological weapons are strikingly infrequent and the number of fatalities and casualties are far lower than those caused by conventional explosives. According to an analysis of 105 U.S. incidents featured in the Monterey Institute database from 1900 to 1998, only one fatality resulted from a [chemical or biological weapon] attack. This incident involved a 1973 assassination of an Oakland, California school superintendent by the Symbionese Liberation Army.
Severity of Harm.
It is generally agreed that chemical agents are likely to be the least lethal of the three “weapons of mass destruction.” In part, this judgment reflects the difficulty of producing and delivering large quantities of a lethal chemical to the target area prior to release. On the other hand, industrial chemicals and pesticides are readily available for purchase, and are stored in large quantities in thousands of locations throughout the United States, often near population centers.
A key question for chemical facilities then is “How much damage could terrorists do using existing stationary chemical manufacturing, processing, distribution, and storage facilities?”
There are two key sources of information for answering this question: accident reports and hazard assessments conducted by facility personnel or outside experts. There is no comprehensive database for either kind of information, but various groups have used publicly available data to estimate hazard potential, usually limited to accidental releases of chemicals from chemical facilities.
The Washington Post reported March 12, 2002, that a classified study conducted by the U.S. Army Surgeon General dated October 29, 2001, found that a terrorist attack resulting in a chemical release in a densely populated area could injure or kill as many as 2.4 million people. According to the news article, the study found “even middle-range casualty estimates from a chemical weapons attack or explosion of a toxic chemical manufacturing plant are as high as 903,400 people.” The worst-case estimate of 2.4 million casualties from a chemical release was roughly half the surgeon general’s estimate for casualties due to widespread use of biological weapons, according to the report. The Army Surgeon General recently explained that the estimate of 2.4 million casualties is of “the number of people who might request medical treatment during a total
release of a large industrial chemical manufacturing plant, in a densely populated
area, and under ideal weather conditions for maximum exposure.” As in most studies of this kind, some question the magnitude and likelihood of the casualty estimates.
In 2004, the Department of Homeland Security used EPA data to estimate the
number of potential fatalities that might result if all the various chemicals at a facility were released suddenly. The purpose of the exercise was to allow DHS to prioritize chemical facility sites for inspections. Assuming that released chemicals would move in the direction of the prevailing winds, DHS determined possible fatalities
within a wedge-shaped zone. It identified two facilities that threaten at least one
million people downwind. DHS selected 360 facilities for its attention in the near
term based on these estimates.
In July 2004, the Homeland Security Council issued 15 national planning scenarios to guide federal, state, and local homeland security preparedness activities. Included in these scenarios are two that refer to industrial chemical releases. One describes a terrorist assault on a petroleum refinery while the other treats the release of a large volume of chlorine from an industrial facility. The planning figures cited for the hypothetical refinery attack include 350 fatalities and an additional 1,000 casualties. For the chlorine release, 17,500 fatalities, 10,000 severe injuries, and 100,000 additional casualties are postulated.
More recent calculations by DHS, based on more sophisticated models, have reduced hazard estimates.
In our best estimate, based on an incredible amount of modeling that we’ve done, the highest-risk facility in the United States would produce under 10,000 potential fatalities and less than 40,000 people that would demonstrate some effects in terms of anywhere from a near-death experience from exposure to inhalation of the toxic chemical to a minor skin blemish caused by irritation through contact with the chemical.
Chemical Site Vulnerability.
CRS identified two publicly available reports that assess site security at U.S. chemical plants. In addition, investigative reports published in newspapers or documented with video recordings indicate that reporters have been able to visit various facilities without being supervised. The studies and selected newspaper accounts are summarized below.
Prior to September 11, an assessment of chemical plant site security by the Agency for Toxic Substances and Disease Registry (ATSDR) was considered by many to be the most comprehensive analysis that was publicly available. ATSDR researchers reviewed national statistics on domestic terrorism compiled by the FBI in 1995, and interviewed security staff from facilities and potential targets in one community with numerous chemical plants. ATSDR researchers concluded:
- “security at chemical plants ranged from fair to very poor;”
- chemical plant security managers “were very pessimistic about their ability to deter sabotage by employees, yet none of them had implemented simple background checks for key employees such as chemical process operators”; and
- “none of the corporate security staff had been trained to identify combinations of common chemicals at their facilities that could be used as improvised explosives and incendiaries.”
The full ATSDR report was never made public, but a DOJ report noted that:
... among the ‘soft targets’ that the ATSDR identified as potential terrorist sites were chemical manufacturing plants (chlorine, peroxides, other industrial gases, plastics, and pesticides); compressed gases in tanks, pipelines, and pumping stations; and pesticide manufacturing and supply distributors.
The DOJ released a study April 18, 2000, describing the risk of terrorism aimed at chemical plants. It concluded that “the risk of terrorists attempting in the foreseeable future to cause an industrial chemical release is both real and credible.” The study also noted that security at many industrial facilities generally is “not as substantial as the security at other comparable potential terrorist targets.” In April and May, 2002, six to seven months after September 11, 2001, the Pittsburgh Tribune-Review published a series of articles describing an investigation of plant security conducted by the paper’s reporters. On April 7, 2002, the newspaper stated that “anyone has unfettered access to more than two dozen potentially dangerous plants in the region” (referring to western Pennsylvania). The author of the report continued:
The security was so lax at 30 sites that in broad daylight a Trib reporter — wearing a press pass and carrying a camera — could walk or drive right up to tanks, pipes and control rooms considered key targets for terrorists.
The report was based on reporters’ trips to 30 plants in western Pennsylvania which have filed risk management plans under the Clean Air Act, Section 112. Two of the plants were among the 123 plants nationwide that projected potential risks to more than 1,000,000 residents in the event of a worst-case accident or attack. The 30 companies constituted more than half of the 61 sites in the region required to file risk management plans. Fifteen of the sites to which reporters gained unchallenged access were water treatment facilities in Pennsylvania and Maryland. In May, another Tribune-Review article described a similar investigation of 30 additional plants in Houston, Baltimore, and Chicago. The report concluded that security was lax at some of “the potentially deadliest plants” in all three cities; access was easy to some sites owned by corporations with large security budgets; employees, customers, neighbors, and contractors “not only let a stranger walk through warehouses, factories, tank houses and rail depots, but also gave directions to the most sensitive valves and control rooms”; and access to 19 sites was allowed due to “unguarded rail lines and drainage ditches, dilapidated or nonexistent fences, open doors, poorly angled cameras and unmanned train gates.”
Chemical manufacturers and users contacted by reporters said that they had bolstered security recently. Several site managers reported that they made immediate changes in procedures or construction plans in response to security breaches by the reporters. But security cannot be ensured “overnight,” according to the president of the Pennsylvania Chemical Industry Council, and it can be expensive. For example, the newspaper reported that U.S. Steel spends more than $1 million each year to equip, train, and hire its own hazardous chemicals response team, firefighters, paramedics, and gate guards at its coke factory. The American Chemistry Council, which represents large chemical manufacturers, has reported that since September 11, 2001, its members have spent over $2 billion at about 2,000 facilities (about $1,000,000, on average, per facility).
Television crews again entered and photographed chemical storage areas in November 2003. Robert Full, Chief of the Allegheny County Department of Emergency Services in Pennsylvania testified February 23, 2004, before the Subcommittee on National Security, Emerging Threats, and International Relations, House Committee on Government Reform, that there continued to be facilities in his county “that one could walk straight in under the guise of darkness and cause significant damage and public danger.” He stated, “Some of the facilities have no more security than maybe perhaps a padlock or a chain.”
In mid-2004, surveys were distributed to 189 U.S. chemical facilities where workers were represented by the Paper, Allied-Industrial, Chemical and Energy Workers International Union (PACE). Of the 133 surveys returned, 125 were from facilities where workers agreed that there were quantities of hazardous materials on site large enough to cause a catastrophic event if they were released. Responses to the survey indicated that surveyed workers believed nearly three-quarters of the plants had improved systems to guard toxic chemicals and had conducted drills to respond to an intrusion by terrorists. On the other hand, according to employees who responded to questionnaires, fewer than half had improved communications, emergency response training, warning signals, or protective equipment, or contacted local first responders about the hazards on their sites. Nearly two-thirds of the plants had not discussed terrorist concerns with neighbors, according to surveyed workers.