215637

CIVIL DIMENSION OF SECURITY 167 CDS 05 E Original: English NAT O Pa rl i a me n t a ry As s e mb l y CHEMICAL, BIOLOGICAL, RADIOLOGICAL OR NUCLEAR (CBRN) DETECTION: A TECHNOLOGICAL OVERVIEW DRAFT SPECIAL REPORT LORD JOPLING (UNITED KINGDOM) SPECIAL RAPPORTEUR* International Secretariat 26 September 2005 * Until this document has been approved by the Committee on the Civil Dimension of Security, it represents only the views of the Rapporteur. Assembly documents are available on its website, http://www.nato-pa.int

167 CDS 05 E i TABLE OF CONTENTS Page I. INTRODUCTION ................................................................................................................... 1 II. GENERAL CONSIDERATIONS ABOUT THE DETECTION OF CBRN THREATS: .............. 2 III. CHEMICAL AND BIOLOGICAL THREATS: DETECTION MECHANISMS ........................... 4 A. BIODETECTION........................................................................................................... 5 B. CHEMICAL DETECTION............................................................................................ 10 IV. RADIOLOGICAL AND NUCLEAR THREATS: DETECTION MECHANISMS ...................... 12 A. PHYSICAL SECURITY OF NUCLEAR AND RADIOLOGICAL MATERIAL AND THE PREVENTION OF TRAFFICKING IN NUCLEAR SUBSTANCES .............................. 14 B. DETECTION AT PORTS OF DEPARTURE AND PORTS OF ENTRY....................... 15 C. PROTECTION OF CRITICAL INFRASTRUCTURES ................................................. 18 V. CONCLUSIONS .................................................................................................................. 19 REFERENCE DOCUMENTS ....................................................................................................... 21

167 CDS 05 E 1 I. INTRODUCTION 1. In the past various terrorist groups have employed or threatened to employ chemical, biological, radiological or nuclear (CBRN) agents. However, despite widespread publicity about the threat, there have been few actual attempts by terrorists to cause mass civilian casualties using CBRN agents. Exceptions have been the salmonella poisoning of 751 people (no fatal cases) by the Rajneesh sect in Oregon, USA in 1984; and the various attempts by the Aum Shinri Kyo in Japan to use both chemical and biological agents, the most “successful” of which resulted in June 2004 in seven dead and 200 hospitalised in Matsumoto, and 12 dead and 1,000 hospitalised in Tokyo. Unsubstantiated threats have been far more common, hoaxes or relatively low-level incidents causing few, if any, casualties. 2. There have also been a small number of attacks on nuclear power facilities worldwide; numerous unsubstantiated threats to trigger nuclear explosive devices; and at least one reported case of radiological materials being used by terrorists to a very limited degree when Chechen rebels planted a cesium capsule in a park in Moscow. 3. However, as information and capabilities become progressively more widespread via the Internet etc, it is becoming increasingly difficult for the authorities to distinguish between a mere hoax and the real thing. This raises a number of difficult questions about the appropriate responses to such threats, which not only have the potential to be extremely disruptive to normal, day-to-day activities, but also may provide individual terrorists and terrorist groups with a potent instrument against society, even in the absence of a real capability or willingness to carry out an actual attack. In any event, an attack using unconventional weapons would certainly cause serious economic and social disruption. According to a recent study by the Organisation of Economic Co-operation and Development (OECD), the cost of a single attack could range between $50 billion and $250 billion. 4. Governments and the general public alike view the potential threat of CBRN weapons being in the hands of terrorists with growing concern. In the autumn of 2001, the anthrax attacks in the US started the warning bells ringing. Recent terrorist attacks directed at public transportation in Madrid or in London brought the threat to Europe and signalled the need to prepare for an even worse scenario in Europe too. At the end of August, information regarding plans by an Al-Qaeda cell to carry out a sarin gas attack on the British House of Commons, and an incident in May 2004, when condoms full of purple flour were thrown at Prime Minister Tony Blair during a Question and Answer session in the House, highlighted the vulnerability and lack of preparedness of national parliaments. 5. All these incidents crudely demonstrate the crucial need to understand the extent of the threat and to adopt adequate measures. The question is, how easy would it be for an individual terrorist or terrorist group to manufacture or to obtain such weapons, and yet more importantly: how easy would it be for such weapons to be delivered, dispersed or used? 6. At all levels governments have been prompted to reconsider their readiness, resources, and capabilities to mitigate the impact of those threats to society. Several international initiatives adopted in recent years have also contributed towards a global awareness of potential threats. At the Kananaskis Summit in Canada in June 2002, G8 countries adopted a “Global Partnership Against the Spread of Weapons of Mass Destruction”, committing themselves to spending up to $20 billion over 10 years on preventing terrorists, or those who harbour them, from acquiring or developing CBRN weapons, missiles, equipment, technology and related materials. 7. The 2003 report on “Civil Protection: a general overview” by Ms Wohlleben (Germany) assessed general threats and policy approaches, and as a follow-up to this, this year Rapporteur Lord Jopling (United Kingdom) decided to focus upon the means available to detect potential

167 CDS 05 E 2 CBRN threats. Despite the fact that recent events, such as the tsunami in Asia, hurricane Katrina in the US and the avian flu pandemic have focused public attention upon natural disasters and emergencies, your Rapporteur strongly believes that these events should not divert efforts away from the equally serious threat of CBRN terrorism, especially as improved CBRN detection mechanisms could enhance civil protection capabilities in general, and further protect us from natural disasters. 8. A Committee visit to the US in September 2005 provided valuable insights for this report and your Rapporteur would also like to thank the British and American delegations for their valuable input to this report. 9. After general considerations of the detection of CBRN terrorism, this report will review the technology currently available for detection of each kind of weapon, as well as the general orientation of related Research and Development. II. GENERAL CONSIDERATIONS ABOUT THE DETECTION OF CBRN THREATS: 10. Detection mechanisms are a fundamental aspect of any successful CBRN civil protection policy. Generally speaking, detection aims at establishing the release or presence of a CBRN agent in a given area/location. Detection is usually associated with prevention. In reality, detection mechanisms are needed at the three stages of a CBRN incident, i.e. before, during and after. Before an incident occurs, CBRN detectors allow for continuous monitoring to either prevent a CBRN incident or to allow for early warning in the event of its happening. These two options are sometimes referred to as detect-to-protect and detect-to-treat. During the incident, detectors are required on the spot in order to allow first responders to identify the precise nature and extent of the release and to organise the response accordingly. Lastly, once the incident has occurred, detectors are indispensable in order to confirm the results of early identification, collect evidence and confirm that the area has been decontaminated. Monitoring, warning, identifying and consequence assessment are thus all core functions of detection. In other words, detection contributes to at least four of the main objectives of civil protection, i.e. prevention, protection, response and recovery. 11. However, detection does not provide a comprehensive and perfect solution in any of these cases. Efficient civil protection requires a holistic approach, of which detection is only one component, along other policies and actions. The difficulty is that in most countries, there is no one overarching and supervising institution in charge of civil protection. The US, with their Department of Homeland Security, is in that sense rather an exception. 12. Co-ordination and integrated efforts and policies are thus among the central challenges to effective adaptation of civil protection policies. They are also crucial in terms of detection. As mentioned above, detection is required at all different stages of an emergency. Detectors also tend to employ “dual-use technologies” that can be useful medical and industrial tools as well as in the field of homeland security. Finally, and maybe most importantly, co-ordination is needed because detection mobilises multiple other policies (intelligence, R&D, health, energy, foreign policy, etc.) and stakeholders (government departments, local administrations, first responders, academic community, industry, international organisations, etc.). 13. Therefore detection should not merely be considered a technological issue. It is also political and involves priority setting, assessment of needs, threats and capabilities and a balancing of competing objectives. It is fundamental to find a balance between both security needs and budgetary constraints, and security and the protection of human rights and individual freedom. For example freedom of trade is affected when detectors are installed at departure ports or entry ports

167 CDS 05 E 3 and slow down the free flow of trade. Privacy is also reduced when technology such as video- surveillance or biometrics are used for counter-terrorism purposes. 14. A genuine detection policy would entail guidelines regarding the development, deployment, use, assessment and adaptation of detectors. 15. Regarding the development of detectors, governments need to define: 1) what detectors are needed for, i.e. the threat they should detect and their objective; i.e. should they aim to detect the threat in time to prevent an attack (detect-to-protect) or should they only allow for early warning once the threat is real (detect-to-treat); 2) who will use the detectors: civilian or military personnel; experts - lab technicians - or novice users - first responders; 3) who will develop them and with which financial resources, i.e. public programmes / public funding, public-private partnerships, civilian-military partnerships. 16. A detection policy also needs to define guidelines for the deployment of detectors. Critical infrastructure must be identified and monitored for the presence of CBRN agents (government buildings, public transportation facilities, postal sorting offices, water supplies, chemical and nuclear plants, etc.). The policy must be applied nationwide, especially in the case of environmental monitors. 17. Furthermore, a concept of use of the detectors also needs to be developed, i.e. the detectors’ properties and the ways in which generated data will be interpreted and used in decision-making. In particular, this means that a proper chain of command and control must be clearly identified to decide upon the information generated by detectors. 18. Finally, policies should be adopted to allow for the validation of detectors, the assessment of their performance and their adaptation. Validation means an official authority would ensure that privately produced detectors meet all specifications. As the Committee learned during its visit to the Port Authority of New York New Jersey (PANYNJ), such controls are fundamental, as tests conducted by the Department of Homeland Security’s Office of Systems Engineering and Development in partnership with the PANYNJ have demonstrated that only a limited fraction of the broad range of commercially available detection devices meet manufacturer’s advertised claims. 19. Adaptation of detectors is also a critical challenge. Most existing technology is flawed and no one country can claim full territorial or infrastructural coverage. The risk is that partial deployment of imperfect technology would create a false sense of security. Nevertheless, the recent and growing focus on CBRN threats has led to technological advances in all categories of detectors and new technology is constantly being developed and tested. 20. Detectors currently on the market differ according to the agent they are intended to identify. The objective of having an all-purpose detector is still a fairly unrealistic prospect. Detectors also differ in their mode of operation; the main distinctions are between standoff and point detection, and between fixed and hand-held detectors. Today’s detection devices include various combinations of these characteristics. Stand-off detectors are stationary systems or mobile units designed to monitor large areas remotely. Point detection refers to hand-held devices which can be pointed at a suspect area or be a point source for detection. Fixed detectors are installed, automatic instruments designed to be used at checkpoints or critical facilities to monitor a continuous flow of persons, vehicles, luggage, cargo, or air samples. Hand-held devices are lightweight instruments, which can be used to detect, locate and sometimes identify a CBRN agent. 21. The choice between these different types of detectors is usually dictated by considerations regarding the purpose of the detector. In other words, detectors must be adapted to their intended purpose. Fixed detectors are ideal when nodal points or critical infrastructures can be identified as

167 CDS 05 E 4 they are most highly sensitive. Hand-held instruments are particularly useful in widely dispersed areas such as airports or seaports, or in targeted search situations. 22. Detectors also differ as to their key function: simple warning when the presence of an agent is detected; identification of the agent (this is a more complex function, which sometimes requires scientific reach back, for example laboratory analysis); mapping/localisation/assessment of the contamination (a capacity some hand-held detectors have). The current drive is towards the development of new, cheaper, easier-to-use, hand-held, highly sensitive detectors, combining different types of technology to both detect and identify CBRN agents, whilst routinely covering large areas. 23. However, as mentioned above, one important question is how affordable these technologies are and what budget governments can and will dedicate to the detection of CBRN terrorism. Although it is sometimes difficult to obtain precise figures, some indicators indicate current spending trends in two leading countries in this field. The United States’ effort is certainly the greatest. According to a recent New York Times article, since 9/11 the US has spent more than $4.5 billion on screening devices to monitor the nation’s ports, borders, airports, mail and search for guns, explosives, nuclear and biological weapons. The US President requested an overall budget for homeland security of $47.4 billion in fiscal year (FY) 2005. In comparison, the UK’s overall spending on counter-terrorism and resilience has increased from £950 million in 2001, to £1.5 billion in 2004 and by 2007-2008 this is expected to become £2.1 billion. The research and development (R&D) budget is also a good spending indicator. In the US, the Department of Homeland Security’s R&D budget for anti-terror technology should amount to $1 billion in FY 2005. In the UK, the Home Office’s CBRN Science and Technology Programme is financing R&D projects to improve terrorism preparedness capabilities. It launched a bid for projects in January 2005 and will finance short-listed projects of an approximated value of £10 million, part of which will finance the development of new detection technology. The bidding process will be repeated every year. 24. These general considerations need to be kept in mind while presenting and assessing CBRN detection technology in use at present. Confronted with a changing threat, our societies must adapt their efforts constantly. Detection mechanisms should also be adapted. How this is done will depend upon the priorities set by each individual society and consequently upon their capabilities. The overview of CBRN detection technology provided in this report is intended to assess efforts made to date. Your Rapporteur also hopes it will foster knowledge of and promote dialogue on how, to which extent and with which limitations, detection technologies can help ensure that our societies are prepared for the worst. III. CHEMICAL AND BIOLOGICAL THREATS: DETECTION MECHANISMS 25. There have recently been reports of new or renewed interest in obtaining chemical and biological weapons being shown by a number of traditional, international terrorist groups. Senior US government officials have publicly asserted that the terrorist financier Osama bin Laden has been actively seeking chemical, biological, and nuclear weapons for use against Western targets. The WMD Terrorism Research Program at the Monterey Institute keeps a listing of reports on al-Qaeda's involvement with CBRN weapons between 1997 and 2004, which currently contain about 80 entries. Although the authors themselves admit that the reliability of sources varies, the mere existence of such a table is in itself alarming. The recent apparent resurgence of the Aum Shinri Kyo in Japan is also troubling, given the technical knowledge of some of its remaining followers, and the possibility of yet-undiscovered stocks of CB agents or precursors. 26. The ideal chemical or biological sensor would fulfil a host of criteria. It would be inexpensive, easy to use, rapidly deployable (hand-held), able to detect all dangerous pathogens, capable of

167 CDS 05 E 5 detecting those pathogens in real time; able to detect them from diverse sample types. It woud be usable, ‘stand-off’ detection; and, most importantly, would always be correct. 27. The dangers of false-positives (detecting a non-existent threat) and false-negatives (failing to detect a real threat) are obvious. To guide policy-makers and reassure a concerned public, there must be faith in the equipment used. Designers must balance the need for sensitivity with the danger of false alarms, with all the consequences they provoke. 28. To date a perfect sensor does not exist. A number of different technologies have been developed to detect chemical and biological agents, and technology is becoming increasingly innovative and sophisticated, but there are still flaws. A. BIODETECTION 29. The US and Europe have become ever more concerned by the threat of bioterror. Whilst September 11 was a watershed in security assessments, the anthrax attacks on the US postal system in September and October 2001 served as an additional wakeup call, leaving 5 dead out of the 22 diagnosed cases, and the perpetrators were never caught. Early diagnosis certainly contributed to a reduction of the overall number of casualties, underlining the utility of detect-to- treat systems. However, more than 27,000 employees of United States Postal Services (USPS) had to be treated and clean-up costs of only two facilities amounted to $300 million. 30. In response, both sides of the Atlantic have explored how to detect a biological attack. The US effort has been the greatest with President Bush’s launch of a new comprehensive initiative called “Biodefense for the 21^st Century” in April 2004. According to one study, after September 11 the total US budget for civilian biodefense increased 16 fold, from $305 million in FY 2001 to approximately $5 billion a year for FY 2004, 2005 and 2006. The increased funding of biodefence research by the National Institute of Health’s is even more striking. This experienced a 34 fold budget increase from FY 2001 to FY 2006. In comparison, the British government allocated £260 million to bio-release countermeasures in FY 2003. 31. Biological agents attract terrorists because of their virulence, toxicity, transmissibility and lethality. They are relatively cheap to produce, sometimes readily available, and are also relatively easy to store and to transport. Moreover, besides naturally existing pathogens, terrorists could try to use engineered organisms. Experts believe that up to 1,000 toxins could be made of natural or genetic sources, although not all of them would be suitable for use as biological weapons. Pathogens are difficult to detect: they are colourless and odourless and have incubation periods, ranging from 48 hours for respiratory anthrax, to 21 days for Q-fever. This incubation period is both an asset and a challenge: an asset because it provides a window for quarantine and treatment of the victims and vaccination of others; a challenge because identification of the disease is often difficult. In early stages many diseases present flu-like symptoms and patients are thus likely to go on with their normal lives, which could cause widespread contamination in the case of transmissible diseases. Treatments and/or vaccines exist for most diseases caused by biowarfare agents (see information document 186 CDS 05 E). Timely detection of an attack is crucial to allow for the deployment of response mechanisms, including medical countermeasures. 32. “Detect to protect” biological detection technology is currently unavailable. Available instruments are usually large, slow and expensive. The more reliable the detection instrument, the longer it takes to identify the defined threat. Thus, the main goal of biological detection is to provide sufficient warning for responders to use the time to minimise casualties; in other words, the goal is to buy time.

167 CDS 05 E 6 33. Several technologies and tools exist, but individually they do not provide sufficient protection against an attack. Biodefense strategies thus tend towards the combination of several layers of detection. A first layer is composed of standoff detectors; a second layer of protection is provided by the use of point detectors; lastly, the collection of epidemiological data can support and complement the use of biosensors. 34. Sensor technology is the most obvious example of biodetection. The fundamental challenge is that biological agents have different properties and many sensors are pathogen-specific; each test must be tailored to recognise a specific pathogen. Moreover, in some cases, even a very small quantity of pathogen will cause disease. In those cases (as for example, in the case of the tulameria pathogen, which requires as few as 10 organisms to infect), sensors need to be sensitive enough to detect even a minimal presence of pathogens. 35. If a biological attack were to be undertaken through the release of a biological agent into the air from a distance, advance detection would be a crucial asset to warn of the attack and allow for an organised response. Early warning is the purpose of standoff detectors. Several technologies, such as Doppler RADARs, LIDARs (Light Detection and Ranging) or LIBS (Laser-Induced Breakdown Spectroscopy), can be used for standoff detection of biological agents. They rely on radio waves or light reflectance techniques to screen clouds for airborne pathogens. However, applications for these technologies are mostly military and their efficiency is still limited. 36. Recent developments in civil protection technology against bioterror focus on the development of new or more efficient point detectors to a far greater extent than standoff detection. The goal is to have detectors allowing for on-the-spot detection and identification of biological agents, where an attack is suspected to have occurred. JASON, an independent scientific advisory group providing defence science and technology consulting services to the US government, identified three broad types of biosensors in its 2003 study on biodetection, based on their mode of operation. 37. Environmental Monitoring refers to the continuous automatic monitoring of the environment in fixed locations. Sensors collect air samples that are then filtered, concentrated and analysed. Environmental monitors are not equipped for definitive identification of pathogens and in the event of detecting an abnormal presence, further tests are essential. Although relatively cheap, they are limited in the number of parallel tests they can perform at once. R&D in this area has focused mainly on detection of anthrax, because, unlike other pathogens, contamination by anthrax is only possible at relatively high levels of concentration. Airborne anthrax is thus comparatively easy to detect. 38. Sample Collection refers to the process of collecting and then analysing samples, either on the spot or in a laboratory. Filter paper is often used to collect the data. The sample is then submitted for a DNA-test intended to identify the biological agent used. Typically, polymerase chain reaction (PCR) is used as an identification procedure. Tests can result in very specific identification of pathogens, but this makes the system inherently less able to detect novel biological agents. Sample collection is also a labour-intensive and costly process. More effective detectors are currently under development, combining sample collection and on-the-spot PCR analysis. 39. The Biological Aerosol Sentry and Information System (BASIS) is a typical sample collection system. BASIS collects air samples at defined locations at specified time intervals, to help determine both the time and place of the release. Aerosol collection hardware continually collects, time-stamps, and stores samples. Samples then need to be transported to a fixed or mobile field laboratory for analysis. The samples are then submitted to the DNA-based PCR analysis for identification. BASIS devices were deployed in Salt Lake City, Utah, for the 2002 Winter Olympic Games.

167 CDS 05 E 7 40. The BioWatch initiative of the US Department of Homeland Security features elements of the BASIS technology but, unlike BASIS, BioWatch is deployed nationwide and, instead of using a mobile laboratory, uses laboratories that are part of the federal Laboratory Response Network, operated by the Center for Disease Control and Prevention (CDC). BioWatch was created in 2003 as a nationwide early warning system to rapidly detect the presence of biological materials in the air. It operates as a network of sample collection facilities, coupled to the network of pollution sensors deployed by the Environmental Protection Agency. BioWatch monitors are placed at key locations nationwide and operate around-the-clock. It currently covers strategic locations in more than 30 major cities. 41. The BioWatch programme was the first attempt at creating a large-scale network of bioterror detectors. However, it has been criticised for its high cost ($53 million in the first year of operation), limited coverage and for the choice of sensor location. As part of a new Bio- Surveillance Program Initiative announced by the Bush administration for FY2005, the Department of Homeland Security announced an overhaul of BioWatch in response to this criticism. BioWatch will receive $55 million in FY 2005, intended in part to modernise its detectors, extend coverage and start networking sensors and integrate them with other monitoring mechanisms. 42. In the United States, new technologies are currently being developed and tested with DHS funding and could be deployed to replace first generation BioWatch detectors. A second generation device is the Autonomous Pathogen Detection System (APDS) developed at Lawrence Livermore National Laboratory (within the US Department of Energy). The APDS can perform both detection and primary identification of at least 11 agents on the field. BANDs are other devices currently being tested (estimated cost: $25,000 or less). These are rugged, semi- autonomous detectors, able to identify at least 20 pathogens and toxins and detect as few as 100 organisms, or as little as 10 nanograms of a toxin. They sample the air more frequently than current BioWatch detectors, test themselves internally and report on results every three hours of this initial screening. However, positive samples still need to be brought to a lab for a secondary inspection. The Rapid Automated Biological Identification System (or RABIS) presents many of the same features as the BANDs detectors. However its mode of operation is different: these detectors could be attached to building heating or air conditioning systems, detect biological agents in under two minutes and shut down ventilation in the event of a release. Since RABIS units are expected to be fairly expensive (the target price is $50,000 a unit), they would probably be limited to the protection of critical infrastructure, such as government buildings. 43. The US Postal Service’s Biohazard Detection System (BDS), which was shown to the Committee during its visit to the US, also uses technology which combines sample collection with PCR-based DNA testing. Following the anthrax attacks of 2001, the U.S. P.S. installed biohazard detection devices (price per unit: $250,000) at 191 of its 282 mail processing facilities (complete coverage is expected by December 2005). This system screens mail for traces of anthrax. The BDS uses automated systems based on rapid on-site PCR analysis of aerosol samples collected during one of the earliest stages of mail processing. The equipment collects air samples as the mail moves through a stamp-cancelling machine. It absorbs the airborne particles into a sterile water base. The liquid sample is then injected into a cartridge, and tested for DNA match and the results are available on-site within less than an hour. 44. Most sample collection devices require that at least part of the identification process be made in a laboratory. Regular PCR procedures take time (usually 2 to 4 hours), are expensive, labour-intensive and require trained users. Moreover, DNA-based techniques cannot be used for detection of toxins, which have no DNA. New PCR or other amplification techniques are being developed to accelerate the process and improve the efficiency of the detection. Research into the miniaturisation of detection devices is also underway.

167 CDS 05 E 8 45. Mass spectrometry is an alternative technique for identification of pathogens. The technique is sensitive and efficient, but detectors are still bulky, expensive (prices range from $30,000 to $150,000 per unit) and require operation by trained personnel. 46. Rapidly-Deployable Sensors are mobile, often hand-held, detectors, which have the obvious benefit of being deployable to the area of a suspected incident. However, their usefulness is limited as they are often pathogen-specific and therefore unable to recognise a broad range of agents. Demands for reduced size and greater mobility also obviously affect the effectiveness of these machines. The Committee was surprised to hear from officials of the New York City Office of Emergency Management that the city has banned the use of such hand-held biodetectors by its responders due to their suspected inefficiency. 47. Basic and non-discriminatory sensors - that is sensors that cannot precisely identify the pathogens detected - include protein detection kits and Aerosol Particle Sizers (APS). Other basic and commercially available sensors are immunoassay kits or tickets with colorimetric indicators. Immunoassays are detectors that mimic the human body’s immune system, based on antigens and antibodies. These sensors are cheap ($20 for each disposable test strip for example) and easy-to-use, but they are pathogen-specific, have high false-positive rates and a low level of sensitivity – i.e. they require more organisms to trigger detection than to infect. Similar tickets or hand-held devices which use DNA-based assays also exist. 48. An example of this type of DNA-based sensor is the Hand-held Advanced Nucleic Acid Analyzer (HANAA), developed at Lawrence Livermore. The system was designed for emergency response groups, such as fire fighters and police and is about the size of a brick. Each hand-held system can test four samples at once, either the same test on four different samples or four different tests on the same sample. HANAA can provide results in less than 30 minutes, compared with the hours to days that regular laboratory tests usually take. It uses the PCR technique to amplify agent-specific DNA fragments to a detectable level. 49. R&D efforts are currently directed towards the creation of field, miniaturised, lab-on-a-chip biosensors, which could be used on the spot by first responders. These sensors combine immunoassays, or DNA-based assays, with signal transduction on a chip to provide direct and quantitative electronic readout. Such sensors would present a host of advantages: they would be cheap, easy-to-use, fast; and could integrate several functions in a single mass-produced device. Another version of these biosensors would involve using living cells from humans, animals or plants. The underlying idea of these cell biosensors is that the detector would respond to the presence of a biological agent just as the intended target would, but more quickly. They would also be able to detect novel engineered pathogens. 50. Beyond sensor-based detection systems, which remain an imperfect science as noted above, there are other means of improving a “detect-to-treat” biodetection architecture, based on the involvement of the health care community. 51. Hospitals and other medical services are likely to be the first institutions to identify the victims of a biological attack. Appropriate and timely responses to an attack might thus depend on accurate diagnosis of contaminated patients. Health care professionals thus constitute another layer of detection, their observations can either replace or complement results obtained through detection devices. The development of Syndromic Surveillance or Bio-Surveillance in several countries aims at making the best use of this crucial asset. 52. Syndromic surveillance refers to the process of collecting and analysing statistical data on health trends, particularly symptoms reported by people seeking care in health care facilities. By focusing on symptoms rather than confirmed diagnoses, syndromic surveillance aims to detect bioterror events earlier than would be possible with traditional disease surveillance systems (see

167 CDS 05 E 9 timeline in information document 186 CDS 05 E). Syndromic surveillance systems regularly monitor a range of existing data for sudden changes or anomalies that might signal a disease outbreak. These data may include school and work absenteeism, sales of OTC products, calls to nurse hotlines, counts of hospital emergency room (ER) admissions or reports from physicians about certain symptoms or complaints. 53. Such systems are already used in both the US and the UK. In the UK, the National Health Service and Health Protection Agency run the programme called NHS Direct Syndromic Surveillance Project. NHS Direct is the only national syndromic surveillance system in England and Wales. It is a nurse-led telephone help-line, which provides health information and advice to 6 million callers a year. This network was originally created to improve detection of influenza outbreaks. In December 2001, it was expanded to provide an early warning for potential deliberate release of biological or chemical agents. It currently provides surveillance of 10 syndromes (cold/influenza, cough, diarrhoea, difficulty breathing, double vision, eye problems, lumps, fever, rash and vomiting). If any anomalies are detected from historical trends, i.e. exceedances, the system triggers a public health alert. The direct annual cost of operating the NHS Direct surveillance system is an estimated $280,000. This system is thus relatively cheap and is considered timely, representative and reasonably efficient. 54. In the US, the Center for Disease Control (CDC) runs the only national syndromic surveillance system, BioSense. This programme has received considerable funding since its creation in FY2003. President Bush’s proposed 2005 federal budget included over $100 million for BioSense. Many cities and state public health agencies have also recently invested substantial funds into syndromic surveillance systems. For example, the New York City Department of Health and Mental Hygiene (DoHMH) runs one of the most advanced syndromic surveillance programme, at an estimated annual cost of $150,000, this including maintenance and routine follow-up of signals. This programme was presented to the Committee during its visit to the US. 55. Syndromic surveillance is considered an attractive tool to detect deliberate and naturally occurring disease outbreaks. It is relatively cheap, because it uses many existing networks and institutions and can serve purposes other than the sole detection of bioterrorism, such as the detection of influenza outbreaks. However, syndromic surveillance is also flawed. A recent study by RAND’s Center for Domestic and International Health Security assessed the use of such surveillance. It recognised the inherent risk of false-positives and the chances of environmental distortion such as the flu season and concluded that, being a relatively untested methodology, health departments should be cautious about investing in costly new syndromic surveillance systems immediately. 56. Sentinel Organisms, meaning the use of animals and even plants for detection, offer another potential source of information. A dog, for example, has an olfactory (sense of smell) capacity that is four times larger than that of humans. In another example, the US Army recently used pigeons in the invasion of Iraq as its first line of detection of chemical and biological agents since they are more sensitive to certain agents than humans. The potential in this area is broad and studies are currently underway to find a means of incorporating such detection into the overall architecture. These range from simple monitoring of veterinary data patterns to advanced bioengineering of plant cells to indicate the presence of certain agents. 57. Biological detection is probably one of the most challenging areas of CBRN detection. No currently available technology used alone is sufficient to protect a population. In the event of a deliberate disease outbreak, the availability of vaccines and treatment is thus crucial. Yet, faced with the relative scarcity of medical countermeasures for biowarfare agents, governments have adopted very different policies, particularly regarding stockpiles of vaccines and population categories which should receive routine vaccinations. In the U.S., the $5.6 billion BioShield

167 CDS 05 E 10 programme aims at building large stockpiles of cutting-edge drugs, vaccines, and other medical supplies for biodefense, but implementation efforts have been recently scaled down. B. CHEMICAL DETECTION 58. Hazardous chemical materials that may be used in attacks include chemical warfare agents, common toxic industrial chemicals, and special purpose chemicals. Fears of chemical terrorism usually focus on chemical warfare agents; these include blister agents (sulphur mustard or HS and lewisite or L), nerve agents (GB and VX), blood agents (hydrogen cyanide or AC, cyanogens chloride or CK, and arsine or SA), and choking agents (chloropicrin or PS, chlorine or Cl2, and phosgene). 59. Chemical weapons’ detection has traditionally been a military matter and current detection capabilities have largely arisen from the military. Chemical agents are less difficult to detect than biological ones. It is already possible even to detect the presence of single molecules, but only in sophisticated laboratory environments. Current detection systems still fall short of the ideal needs for civilian detection purposes. They are either insufficiently sensitive, not mobile, or require a trained user. 60. A market survey of commercially available detection equipment conducted five years ago identified 148 detection devices available for the military and first responders. Typically prices range from $10,000 for hand-held detectors to $20,000 for fixed instruments. The following are some of the more widely used examples and none offer a perfect solution. First responders generally use chemical detection paper, or, in a few cases, ion mobility spectrometer (IMS) devices or combined IMS / surface-acoustical wave detectors for early warning. In the event of positive results, further confirmation is needed through the use of more sensitive lab technology, which takes between 6 and 48 hours. Gas Chromatography (GS) combined with Mass Spectrometry (MS) is the standard method of identification and quantification of chemical agents. Some GC-MS hand-held devices exist and further efforts at miniaturisation are underway. 61. Colorimetric Indicators, based on enzymatic detection techniques, are at the most basic end of the chemical detection scale. They are available to first responders and are cheap, fast and simple to use. They contain an acid–base indicator that changes colour when exposed to specific agents in liquid or aerosol form. These indicators are highly prone to false-positives from various everyday substances, even smoke. They are essentially an early warning system that must be confirmed by further laboratory testing. The same colorimetric principle is also used in detection tubes, through which vapour or gas is pumped. They are agent-specific, requiring a different tube for every agent, chosen from a range of more than 160 substance-specific reagent tubes. Colorimetric detection tubes are nonetheless familiar to the first responder community, because of their low cost and simplicity of use. 62. The US military, as well as specialised HAZMAT teams, use M8 and M9 detection paper. M8 paper is blotted on liquids that arouse suspicion. It identifies agents by changing colour within 30 seconds of exposure. M9 paper has adhesive backing that allows it to be attached to clothing and equipment. 63. Ion Mobility Spectrometry is another means of point (hand-held) detection. It uses an electric field to recognise differences in the velocity of ions and has been miniaturised to the point that it is used in mobile detection without diminished resolution. This process is used in many current detection systems, mainly because it is fairly resistant to contamination and false-positives. Generally the response time is very short (minutes), but dependent on agent concentration. IMS detectors typically cost around $10,000.

167 CDS 05 E 11 64. The Finnish M86 and M90, the Improved Chemical Agent Monitor (ICAM), or the APD 2000 all use IMS technology and are available to civilian first responders. They can detect and identify the most common chemical warfare agents. Stand-alone detectors also exist, allowing for very precise detection and identification. 65. Surface Acoustical Wave Detection is a popular choice for first responders, due to the relatively low cost. It can also detect multiple agents simultaneously. These SAW devices use piezoelectric quartz crystals coated in polymers, which absorb certain chemicals. Using an array of sensors provides a response pattern that is unique to a chemical agent. The limit of this absorption process in turn limits the sensor’s sensitivity, other molecules being inadvertently absorbed can also undermine the process. 66. SmallCAD is a lightweight, hand-held and battery operated chemical vapour detection instrument, combining IMS and SAW for higher sensitivity and lower false alarm rates. It can detect and identify a range of chemical agents and provide concentration levels in less than one minute. It is commercialised at a price of $30,000. 67. Mass Spectroscopy, usually used in conjunction with Gas Chromatography (GC-MS) involves breaking apart a molecule before accelerating the charged fragments and bending their paths in a magnetic field. Although highly sensitive and able to tackle mixed samples, the technology is not sufficiently small at present to be incorporated into mobile systems. It is also expensive and requires sample preparation before testing, which needs trained personnel. It is thus not used in detection systems available to civilian first responders. The accuracy of the technology is reflected by the fact that it is the only approved technique for CWC (Chemical Weapons Convention) inspection on-site analysis. 68. Infrared Radiation is employed in various chemical agent detectors. Chemical agents each have a unique infrared fingerprint based on their vibrational wavelength. Passing infrared light through gases or vapours results in absorption of specific wavelengths of light. Infrared spectroscopy measures the quantity of light absorbed at given wavelengths in order to identify the agent. It can be used for standoff detection, usually in military applications, or point detection, which is more appropriate for use by first responders. 69. As well as these oft-used detection techniques, a host of others exist which all have various shortcomings in field or mobile usage. Examples include Flame Photometry, which burns a sample in a hydrogen flame and identifies it from the resulting emission, or Photoionisation, which uses ultraviolet light to ionise vapours or gases and then monitors the change in electrical current. 70. As this presentation of chemical and biological detection technologies has demonstrated, there is no one perfect or universal detector for biological and chemical threats and all existing detectors suffer flaws. Governments are thus faced with difficult choices as to their civil protection strategies. According to the JASON study mentioned above, ensuring blanket coverage of the whole population with detectors would be very expensive and not necessarily the most efficient strategy. Using the city of Lincoln, Nevada, as a model, the study estimated that each sensor node would cost approximately $100,000, with an annual maintenance cost of approximately $10,000 (2003 prices), or an amortised cost of $40 per person per year, that is $10-15 billion nationally. 71. A rational, multi-layered biological and chemical defence architecture, combining and integrating currently available and tested tools is a more realistic and preferable short-term option. This approach relies on the constitution of vertical and horizontal webs or layers of detectors. A rational distribution of detectors at potential target points can constitute a first horizontal layer of protection. However, this supposes that governments define priorities and are willing to decide which infrastructures are critical. The vertical component calls for the use of different layers of

167 CDS 05 E 12 detectors, from the less sensitive and precise (usually the detectors that allow for continuous and indiscriminate detection) to the more accurate (usually using more labour-intensive and complex technologies). The consecutive use of detectors will allow a move from mere suspicion to certainty, as well as a reduced chance of false-positives and false-negatives. A further objective is to improve the compatibility and synergy between different detectors. Integration of detectors with other indicators or sources of information – intelligence, syndromic surveillance data, etc. – should also be a priority. When the Committee visited the United States in September 2005, it was briefed on some of the public and private initiatives currently being developed to create integrated emergency management systems. One such example in the area of biodefence is the U.S. National Biosurveillance Integration System (NBIS), which, when fully operational, should integrate data collected from sensors throughout the country (BioWatch), information from health (BioSense) and agricultural surveillance and terrorist-threat information from law enforcement and intelligence communities. 72. Other prevention or response policies also need to complement this detection architecture. In particular, prevention initiatives include controlling access to hazardous material, such as deadly pathogens or dual-use chemicals. Security standards for labs and other facilities involved in sensitive chemical and biological-related activities should include both national and international initiatives and should engage the private sector. Several initiatives are underway to develop codes of conduct for scientists engaged in such research activities. Exports control mechanisms and threat reduction programmes in the Commonwealth of Independent States (CIS) region, for example, help improve global bio- and chemical security. The Biological Weapons Convention regime, the Organisation for the Prohibition of Chemical Weapons (OPCW), as well as the World Health Organisation (WHO), also contribute to the prevention and response to biological and chemical terrorism. 73. Investment in R&D for new related technology is crucial to ensure that protection mechanisms are constantly adapted to new threats and needs. This can be achieved by various means. Public funding is important, but alone it cannot provide for the whole research effort. Civilian-military partnerships have allowed for the development of new technology and adaptation to the needs of first responders. Governments have also explored ways of fostering public-private partnerships. These are potentially very efficient tools, as long as governments enforce appropriate standards and oversight. In the US, the Homeland Security Advanced Research Projects Agency (HSARPA) is the Department of Homeland Security’s arm for engaging industry, academia, government, and other sectors in innovative R&D, rapid prototyping, and technology transfer to meet operational needs. In April 2004, this agency awarded contracts to 14 teams amounting to a total budget of $48 million for the development of a new generation of biosensors, including detect-to-treat and detect-to-protect technologies. 74. Finally, it should be kept in mind that efforts towards the development of new technology also prepare our societies for other kinds of non-deliberate events and broaden policy objectives, such as the advancement of science in the field. A very timely example is the protection against agroterrorism, that is the contamination of field crops, animals, food items or water supplies. Monitoring agriculture is not only protecting against terrorism, it is also protecting against natural disease outbreaks, such as, for example, avian influenza. IV. RADIOLOGICAL AND NUCLEAR THREATS: MECHANISMS FOR DETECTION 75. The International Atomic Energy Agency (IAEA) has categorised four potential nuclear security risks: theft of an existing nuclear weapon; radiological hazards caused by an attack on, or sabotage of, a nuclear facility or transport vehicle; acquisition of nuclear material and preparation of a primitive or improvised nuclear weapon; malicious use of radioactive sources, particularly in a

167 CDS 05 E 13 so-called “dirty bomb”. Preparedness scenarios have focused mainly on the last two categories – primitive nuclear weapon or “dirty bomb”. 76. Terrorist groups armed with radiological weapons are one of the gravest risks our societies faces. Unlike nuclear weapons, radiological dispersal devices (RDD), or “dirty bombs”, are not very hard to acquire, transport or build. A “dirty bomb” does not trigger a nuclear reaction or involve a nuclear explosion. It consists of a high explosive, e.g. semtex, dynamite or TNT, some incendiary material, e.g. thermite, and some radioactive material. The detonation of the conventional explosive would spread radioactive material and contaminate personnel, equipment, facilities, and terrain. The fire caused by the incendiary material would carry the radioactivity up into the air, further spreading contamination. 77. A “dirty bomb” is likely to result in some immediate deaths and serious injuries, caused by the explosion of the conventional explosive rather than by exposure to radiations. Effects on the health of those exposed to radioactivity depends upon how long they remain in the contaminated area, the size of the particles released by the explosion, and the type of radioactivity emitted. While such weapons would bring about far less damage than a nuclear explosion, which would result in hundreds of thousands of casualties, RDDs have enormous power to intimidate and also have the potential to cause serious social, psychological and economic disruption. Decontamination would be very costly and would last for weeks, if not months. According to one estimate by the Center for Homeland Security and Defense (CDHS), a terrorist attack on a major port could result in losses of $1.5-2.7 billion per day for the first few days, $5 billion a day for the next two weeks, and could then rise exponentially thereafter. 78. A simpler RDD would aim at spreading radiological material without the use of an explosive, for example in water or food supplies, or by simply placing radioactive material in a public location, e.g. a trashcan on a busy street, to contaminate people passing by. Although such a device would probably have limited effects, it would also be difficult to detect before a significant number of contaminations occur. 79. An estimated ten million radioactive sources exist around the world, with several hundred thousand sufficiently radioactive to pose a health threat. Potential radioactive sources for an RDD include Cobalt-60, Cesium-137, Iridium-192, Strontium-90, Americium-241, Californium-252, and Plutonium-238. The most typical areas where radiological materials are used are hospital radiation therapy (Iodine-125, Cobalt-60, Cesium-137), radiopharmaceuticals (Iodine-131, Iodine-123, Technetium-99, Thalium-201, Xenon-133), nuclear power plants spent fuel rods (Uranium-235), universities and laboratories (see information document 186 CDS 05 E). Radiological material is also used in smoke detectors (Americium-241). Other common radiological materials are Iridium- 192 and Plutonium-239. Among these, a “dirty bomb” using strontium-90 (a highly radioactive isotope found in old Soviet power generators), HEU or spent nuclear fuel from a nuclear power plant, would have the most devastating consequences. Plutonium is about ten times more toxic than nerve gas and the dispersal of as few as 3.5 ounces of plutonium would kill everyone in a large office building. However, access to significant quantities of plutonium would be difficult. 80. Another threat could come from a different type of terrorist attack, using a primitive or improvised nuclear weapon rather than a “dirty bomb”. Unlike “dirty bombs”, a primitive nuclear weapon – also called improvised nuclear device (IND) - would actually imply the explosion of a nuclear device fabricated with stolen or illegally acquired civil plutonium. The damage caused through such a device would be great, even if the nuclear explosion induced were relatively limited. The explosion of the high explosives would cause the unfissioned plutonium to be widely dispersed, potentially contaminating large areas. Such an apocalyptic scenario should not be considered completely unrealistic. When the Aum cult prepared its attack on the Tokyo underground in 1995, for example, its initial plan was to fabricate a nuclear weapon and members of the group who were nuclear scientists, had been recruited to acquire fissile material.

167 CDS 05 E 14 81. In the event of a release of radiological material, three types of radiation-induced injury can occur: external irradiation, contamination with radioactive materials, and incorporation of radioactive material into body cells, tissues, or organs. More specifically, there are four types of radiation that are emitted: 82. Alpha Radiation is the heaviest and most highly charged of nuclear particles, however alpha particles are only able to travel a short distance in the air and cannot penetrate the skin. Materials emitting alpha radiation can only harm humans if inhaled, swallowed or absorbed through open wounds. As a consequence, clothing and turnout gear can keep alpha emitters off the skin. Various instruments are available to detect alpha radiation emitting materials, but special training is essential to make accurate measurements. One example is the palm hand-held precision Geiger-mueller meter that detects and measures alpha, beta, gamma and x-ray forms of radiation. Such instruments are designed for emergency responses, domestic preparedness, hazardous material safety, law enforcement, and compliance verification applications, allowing their users to determine whether a particular area is a nuclear or radiological “hot zone”. 83. Beta Radiation occurs when high-energy electrons are emitted from the nucleus of an atom during radioactive decay. Beta radiation can travel in air and is moderately penetrating. Skin injury can occur if beta-emitting materials remain on the skin for a prolonged period of time. If deposited internally, beta contaminants may also be harmful. A survey instrument (such as a Geiger counter CD V-700) can detect beta radiation. Clothing and turnout gears provide some protection to the skin. 84. Gamma Radiation is high-energy photons emitted from the nucleus of atoms. They easily penetrate body tissue and many other materials, and are potentially lethal. Thick layers of dense materials, such as lead, can protect from gamma ray exposure. Clothing and turnout gear provide little shielding from penetrating radiation. Gamma rays can be detected with survey instruments, including civil defense instruments. A standard Geiger counter can measure low levels of radiation, while an ionization chamber is able to measure high levels of gamma rays. The most appropriate instruments to measure accumulated exposure to gamma radiation are pocket chamber (pencils) dosimeters, film badges, thermo luminescent, and other types of dosimeters. 85. X-Rays are an invisible and highly penetrating electromagnetic radiation of much shorter wavelength (higher frequency) than visible light. As with gamma rays, only thick layers of dense materials can defend against x-rays. 86. The threat arising from terrorists trying to smuggle illicit radioactive materials or nuclear fission weapons has forced governments to embark on programmes to protect, control and account for material of proliferation concern. Current efforts to prevent and detect the use of RN material by terrorists have three major aims: 1. securing sensitive material where it is found – i.e. nuclear facilities, medical and industry environments using radioactive sources; 2. monitoring international borders for attempts at cross-border trafficking in RN material; 3. domestic deployment of networks of detectors, to cover critical infrastructures in particular. These three aspects are examined below. A. PHYSICAL SECURITY OF NUCLEAR AND RADIOLOGICAL MATERIAL AND THE PREVENTION OF TRAFFICKING IN NUCLEAR SUBSTANCES 87. The events of September 11, 2001 have intensified concern that terrorist groups will attempt to steal weapons-usable nuclear material in order to build a nuclear weapon. Although stocks of these materials - plutonium and highly-enriched uranium (HEU) - exist in many countries around the world, the largest inventory is held in the Newly Independent States of the former Soviet Union (NIS). Owing to economic and political turmoil, this material is vulnerable to theft. A close

167 CDS 05 E 15 examination of open source evidence reveals 14 confirmed cases of theft or attempted theft of weapons-useable material from NIS facilities between 1991 and 2001, mostly highly enriched uranium. Even in the US and Europe, it has been reported that thousands of radioactive sources have been lost or stolen. According to the IAEA, between 1993 and 2004, there were 650 confirmed cases of illicit trafficking of nuclear and radiological substances worldwide, of which a significant number involved material that could be used to produce either a nuclear weapon or a “dirty bomb”. Networks of illegal transfer of nuclear technology, such as the one set up by Pakistani nuclear scientist Abdul Qadeer Khan, the exact reach of which is still unclear, are also a serious concern. 88. Dozens and dozens of instances of profit-motivated nuclear hoaxes have been reported in the media in the past two decades. Such hoaxes involved sellers offering weapons-usable or weapons-grade nuclear material and instead deliver some other bogus radioactive, or in some cases, non-radioactive substance. Such scams increased when economic conditions in the former Soviet Union and Eastern Europe declined in the late 1980s and early 1990s. The region’s economic decline coupled with weakened security and enforcement mechanisms and a growing interest on the part of both state and non-state actors to illegally obtain nuclear materials all created favourable conditions for nuclear trafficking scams. 89. All these cases demonstrate the acute need to combine detection policies with effective policies to control the spread of nuclear and radiological material, and nuclear technology in order to limit the risk of terrorists accessing them. More generally, the physical security of nuclear material in all sensitive facilities, as well as the security of nuclear facilities themselves – nuclear power plants, storage sites, etc. - need to be reinforced along common lines. Current initiatives in this area include both national efforts, bilateral and international co-operation. The US leads several threat reduction programmes with CIS countries. The IAEA itself adopted a Code of Conduct on the Safety and Security of Radioactive Sources. Further international initiatives include the G8 Action Plan on Securing Radioactive Sources, adopted at the Evian Summit of 2003 and the Proliferation Security Initiative (PSI) agreed in May 2003, which aim to foster international co-operation and halt illicit shipments of WMD or WMD-related material. Several export control groups, such as the Australia Group or the Nuclear Suppliers Group, are also active in the regulation of transfers of sensitive material and technology. 90. Broader and longer-term efforts to thwart the proliferation of nuclear material and technology could also include the re-shaping of current non-proliferation regimes. However, the failure of the 2005 NPT Review Conference in New York to reach agreement on the further strengthening of IAEA safeguards demonstrates the difficulty to build consensus on such sensitive issues. B. DETECTION AT PORTS OF DEPARTURE AND PORTS OF ENTRY 91. Terrorists intending to smuggle radiological materials into target countries aim to exploit weaknesses of the control mechanisms at ports, terminals, border crossing and airports. Both the UK and the US have embarked on ambitious programmes to install hundreds of detectors at major points of entry. The UK Cyclamen programme, agreed in April 2003, provides for the introduction of routine screening of cargo, vehicles and people entering the UK to check for the illicit importation of radioactive materials. An extensive trial and assessment of radioactive screening equipment was conducted at selected ports in 2002. Drawing on the results of these tests, Cyclamen will procure fixed and mobile detections units. The aim is to screen all air, sea and Channel Tunnel traffic, including container and road freight, post and fast parcels, vehicles and passengers. 92. As 90% of all traded goods travel by sea on approximately 72 million sea containers a year, port detection mechanisms are of paramount importance. In this respect, national authorities must try to guarantee security without harming commerce. Here again, governments face a strategic

167 CDS 05 E 16 choice between a policy aiming at screening 100% of incoming goods at the risk of slowing down trade flow, and one that only screens “suspicious containers”, at the risk of overlooking others. 93. Current technologies to detect radiological and nuclear threats are fairly mature. Typically, a detection architecture would combine fixed and hand-held detectors. Fixed detectors placed at ports of departure or ports of entry can help detect radiological or nuclear materials or weapons before they reach their destination. They also contribute to the fight against trafficking of RN materials and weapons. Hand-held devices can also be used at ports for detection or confirmation of the presence of RN material. Additionally, they can help monitor large areas and be used by responders to monitor contamination and decontamination. 94. A combination of active and passive detection can also improve detection capabilities. Passive detection systems are relatively simple and safe to employ, but they can be evaded by shielding. Active systems allow for enhanced detection, also of shielded material. They use detectors that x-ray or irradiate an object with neutrons or high-energy electrons, to either get a “picture” of the contents of a container or “interro gate” these contents by setting off physical reactions. However, active systems are often more costly, inconvenient, complex and are potentially harmful to humans. 95. Prices of RN detectors range from $150,000-$250,000 for radiation portal monitors to $50,000-$80,000 for a large, laboratory-type spectrometer, and as little as $2,000 for a hand-held detector. Recent efforts have involved the development of non-intrusive technology, i.e. devices that do not necessitate manual inspection of the contents of a container or vehicle. These are ideal for quick detection of a great number of containers or vehicles in strategic transit points, such as sea ports. 96. Radiation Portal Monitors (RPMs) are a popular choice for nuclear and radiological detection at ports of entry. It is a passive, non-intrusive and quick technology. RPMs can screen trucks, cargo containers, rail cars, passenger vehicles, and other conveyances and detect the presence of various types of radiation. The monitors, which typically consist of an array of detectors in one or two vertical pillars with associated electronics, capture energy emitted by radioactive sources and set off an alarm whenever such a source is detected. RPMs are deployed at major ports and border crossings worldwide. In the US alone, more than 400 RPMs are deployed at 22 major ports of entry to scan the 7 million cargo containers entering the US every year from abroad. The President’s budget for FY2006 includes $125 million to continue the deployment and enhancement of WMD Detection Technology at US ports. 97. However, RPMs have been criticised for their limitations. Firstly, they do not identify the exact origin of the radiation and consequently tend to produce a high number of false-positives, responding to naturally occurring radiation materials (NORMs) or medical and industrial isotopes that do not pose a threat to human health. Moreover, they are not sensitive to fissile material, such as uranium-235, which only emits low levels of radioactivity, about one hundred-millionth of the radioactive material that might be used in a “dirty bomb”. They are also less efficient in detecting nuclear or radiological material when shielded in lead or other heavy metal. RPMs must therefore be complemented by other, more accurate technology. 98. Among other non-intrusive technologies are active large-scale imaging systems, which use X-rays or gamma-rays to produce images of the content of a cargo container or vehicle within 2-3 minutes. 166 such systems are currently deployed in the US. The Vehicle and Cargo Inspection System (VACIS) produced by the Science Applications International Corporation (SAIC). based in San Diego, California, is an example of gamma-ray imaging technology (price per unit: about $1 million, plus $500,000 a year operating costs). During its visit to the US, the Committee saw presentations of the VACIS, as well as other detection devices, at the SAIC virtual Emergency Operations Center in McLean, VA.

167 CDS 05 E 17 99. Various gamma and neutron detectors are available commercially, which can identify and distinguish specific radioisotopes. Radiation Isotope Identifiers, for example, are used in conjunction with RPMs to allow for identification of a radioactive source. These hand-held battery-powered gamma-ray spectrometers are capable of detecting gamma and neutron emissions from radioactive sources and identifying the exact source of alarm within minutes, based on the spectrum of radiation or radiation signature. More than 500 of these systems are deployed in the US to equip Customs and Border Protection officers. One manufacturer commercialises RIIs for about $10,000 each. However, these devices have also been criticised for their high rates of false-positives and false-negatives, as well as for their limited sensitivity to the most dangerous materials. 100. During its visit to the PANYNJ, the Committee received a presentation on a new detector developed by Sandia National Laboratory and tested by the PANYNJ. Known as SMART (Sensor for Measurement and Analysis of Radiation Transients) and mounted on a Jeep, this system uses sodium iodide detectors and special software to distinguish between NORMs and other kinds of radiation. This technology is easy-to-operate, mobile and considered fairly accurate. It can be used to complement other radiation detection devices. 101. Recent research into nuclear and radiological detectors focuses on the identification of the neutron and gamma-ray signatures of radioactive sources. Some of these detectors combine gamma-ray imaging and radiation detection. One of the most advanced detection mechanisms currently under development is a neutron generator sensor. Neutrons, unlike gamma rays, can pass through lead or other metal, allowing the sensor to detect shielded nuclear material. The generator bombards a container with neutrons, producing nuclear fissions in materials when in contact with uranium or plutonium. The container is then scanned by detectors, which analyse gamma rays produced by the fission. Specific energy levels correspond to each substance, permitting identification of the substance concealed in the container. This technology would be used as a secondary test when other non-intrusive technologies have revealed an anomaly. Neutron spectrometers function along similar principles, but they identify materials based on the spectrum produced by the scattering of neutrons when bombarded at the material, rather than a gamma-ray spectra. Ultra-high resolution neutron spectrometers are currently under development. 102. Neutron and gamma-ray detection are also the basis for development of glass optical fibre detectors by the Pacific Northwest National Laboratory. These have been commercialised by NucSafe of Tennessee and used by various U.S. and European governments. A light is emitted at the end of the fibres when they are hit by a neutron or gamma ray emitted by radionuclides such as plutonium. Ionising radiation interacts with the scintillating fibres and produces light. Fibre detectors can be used to monitor large areas for illicit nuclear material. Typically, fibre sensors are embedded in roads at border crossings to detect nuclear material smugglings. 103. Detection technology is thus relatively widely available to protect ports and other national points of entry. However, some governments feel that waiting for nuclear or radiological material to reach a nation’s ports is an excessive risk and a late detection. To enhance detection of attempted transfers of nuclear and radiological material, as well as to reduce delays and costs, cargo containers should be inspected once only, preferably at ports of departure, and then sealed by electronic systems to ensure that they are not opened en route to their destination. This is the purpose of several bilateral or multilateral cooperative programmes. A major initiative in this area is the US-led Container Security Initiative (CSI). This initiative, launched by U.S. Customs in January 2002, aims to protect containerised shipping from exploitation by terrorists. For this, a team of US officers is deployed to work with host nation counterparts to target and pre-screen all US-bound containers that pose a potential threat. As of June 2005, CSI covered 37 ports in 20 countries at various levels of implementation. The World Customs Organisation, as well as the European Union, have expressed support for the programme and called for its expansion. Initiatives also exist to engage the private sector through voluntary frameworks, such as the

167 CDS 05 E 18 Customs-Trade Partnership Against Terrorism (C-TPAT). In addition and as a complement to the CSI, the US State Department run programmes to install RPMs in more than 20 countries abroad with the support of the Departments of Energy and of Defense. $500 million were spent on these programmes between FY1994 and FY 2005. 104. Finally, besides monitoring ports and other points of entry for the illegal importation of radiation emitting materials, the entry of illegal asylum seekers or migrants, some of whom could be potential terrorists, must be controlled. C. PROTECTION OF CRITICAL INFRASTRUCTURES 105. Many of the devices used at ports of entry can also be used to protect critical infrastructures throughout national territory. For example Radiation Portal Monitors can also be placed at international mail and package handling facilities to screen for radiation. Glass optical fibre detectors can be embedded in major roads. 106. Recent progress in miniaturisation of low power electronics have also made the development of compact gamma and neutron detectors possible. These can be broadly distributed to different categories of personnel for routine use. These instruments are similar to message pagers. They are small, hands-free, low-power instruments which can be worn by law enforcement or customs officers for continuous monitoring. At about $1,600-2,000 each, they are also relatively cheap 107. Such radiation pagers have been used in the US since 1998 and equip more than 10,500 customs officers and border patrol agents. However, their performance is generally poorly rated. In any case, radiation pagers cannot function as independent detection devices and need to be coupled to other more sensitive sensors, in the event of a positive alarm. A more recent technology, called RadNet combines a cellular telephone, a personal digital assistant with Internet access, and a global positioning system (GPS) locator with a radiation sensor. The RadNet detector is fairly inexpensive (about $2,000), lightweight, able to operate at low power and is precise enough to eliminate background radiation emitted by food, medical devices or soil. 108. R&D in new technology is crucial to enhance current systems and compensate for flaws. In the US, several initiatives and programmes aim at supporting research into new detection technologies and ensuring that private as well as public manufacturers respect adequate standards. An example of this is the project to test and assess new radiation detection systems for air, sea and land established by the Department of Homeland Security’s Environmental Measurement Laboratory and the Port Authority of New York New Jersey. 109. The project, which was presented to the Committee during its visit to the US, has successfully tested commercially available cargo radiation monitors, hand-held instruments and prototypes of the next generation of detection systems. The Department of Homeland Security has requested $227 million in FY 2006 as part of its internal reform to initiate and coordinate a national effort to develop improved radiation detection technologies, fostering both short-term improvements of existing technology and a long-term transformational R&D programme. 110. Globally, current R&D efforts are directed towards ease of use and integration of several systems for increased efficiency. “Sensor fusion” is the keyword of this trend, that is a combination of data collected by different kinds of sensors to produce the most accurate results. For example, integrated systems would combine information from a portable radiation detection system with that of hand-held detectors and video cameras, or information from gamma-ray detectors, with neutron detectors and detectors that take visual images.

167 CDS 05 E 19 111. Further integration should also be achieved through the combination of data from detectors with information from other sources, such as intelligence. The creation, in April 2005, within the US Department of Homeland Security, of the Domestic Nuclear Detection Office as the primary entity to supervise all efforts aimed at the prevention of nuclear and radiological terrorism is clearly intended as a response to this need for integration. V. CONCLUSIONS 112. This paper intends to list the various types of devices already in existence, or being developed to identify CBRN agents as early as possible. In the event of attempts to import these agents, or in the event of an actual release of CBRN agents, the most urgent step is to identify them in order that appropriate measures be taken to protect the civilian population. Time is crucial in preparing for CBRN terrorism. 113. In an ideal world, one might wish for a complete range of devices to be available for use in heavily populated areas. This would be incredibly expensive, as costing estimates in this paper demonstrate. But in the event of an actual CBRN attack, it is almost certain that current capabilities would be insufficient, leading to strong criticisms of both national and local government by politicians, media and public opinion. One crucial challenge of civil protection lies in this very difficult, and yet crucial political assessment of how much is “enough”. Only a few common standards can guide this assessment which remains fundamentally country-specific. In operational terms, however, co-operation can be crucial. Euro-Atlantic partners share a common interest in the fight against international terrorism and therefore need to develop common actions, based on shared experiences and resources. 114. Some lessons can be drawn from the review of currently available detection technologies in this report and the way in which they are used. Tools need to be developed to allow for monitoring of large areas and/or critical infrastructure. Devices should also be adapted to the needs of their intended users. First responders in particular need quick and easy-to-use devices, which can detect, and, if possible, identify the source of a contamination. Currently available technology is far from meeting these ideal standards, despite the occasional claims of certain unscrupulous manufacturers. It is therefore fundamental for governments to adopt and enforce strict standards for the use of detection technology, while at the same time continuing to invest in R&D in new devices. 115. The main and most general lesson of this report is that, to be effective, a CBRN detection policy has to make the most of existing technologies by adopting a comprehensive multi-layered approach. Horizontal and vertical networks of detectors need to be built and integrated with other available information sources, be these medical surveillance in the case of a biological attack or, more general, intelligence sources. Priority should also be given to the co-ordination of detection policies with other policies. Although this paper focuses on technology, it should be clear that technology does not provide an exclusive answer to terrorist threats. Intelligence is crucial to helping us understand the threat and direct the use of necessarily limited resources. Prevention policies, particularly those addressing the root causes of terrorism should also be developed. 116. More generally, the technological dimension of terrorism preparedness efforts should not overshadow the human dimension. Raising the population’s awareness, informing and educating it will help make people an integral part of the detection and response architecture. This is certainly not an easy task. Transparency and security are far too often considered as conflicting objectives. Here again, governments must find a balance based on their national traditions, needs and structures. Another fundamental aspect of the human dimension is the need to train those categories of personnel who will be using the various technologies reviewed in this report, i.e. first

167 CDS 05 E 20 responders, health care professionals, law enforcement and customs officers, etc., and teach them about both the uses and limitations of detection technologies. 117. In all these areas, international co-operation helps improve global preparedness. Preventing terrorism is our common responsibility. During its visit to the US, the Committee was particularly surprised to hear that some non-governmental experts in Washington feel that NATO has not yet demonstrated a firm interest in engaging in civil protection policies. Your Rapporteur strongly believes that the Alliance could go beyond its existing programmes and reflect upon the positive role that it could play in support of member countries as they prepare for and respond to CBRN terrorist attacks. 118. It would clearly be foolish for us publicly to seek to identify what measures have already been taken, thus, by implication, drawing attention to the gaps. Therefore, the purpose of this paper was to highlight what could be done in advance to protect civilian populations. This will hopefully encourage politicians to enquire what preparations have already been made in their own countries and thereafter to urge their governments at national and local levels to do as much as is financially feasible to fill the gaps. Our civilian population is entitled to expect no less of us.

167 CDS 05 E 21 REFERENCE DOCUMENTS Armstrong, Cooper and Prior, ‘Report on Activities and Programmes for Countering Proliferation and NBC Terrorism’, Counterproliferation Programme Review Committee Report to Congress (May 2004) ‘Making the UK safer: detecting and decontaminating chemical and biological agents’,The Royal Society, (April 2004) JASON, ‘Biodetection Architectures’,The Mitre Corporation (February 2003) Stoto, Michael A et al, ‘Syndromic Surveillance: An Effective Tool for Detecting Bioterrorism?’, RAND Center for Domestic and International Health Security (February 2004) Stoto, Michael A., ‘Syndromic Surveillance’, Issues in Science and Technology Online (Spring 2005), http://www.issues.org/issues/21.3/stoto.html Armstrong, Robert et al, ‘Looking for Trouble: A Policymaker’s Guide to Biosensing’, Centre for Technology and National Security Policy, US National Defense University (June 2004) Davis, Griffin and Gabor Kelen, ‘CBRNE– Chemical Detection Equipment’ (June 2004), Section 3, http://www.em22pedicine.com/emerg/topic924.htm http://www.globalsecurity.org/military/systems/ground/icam.htm Kosal, Margaret E, ‘The Basics of Chemical and Biological Weapons Detectors’, Centre for Non- proliferation Studies (November 2003) ‘Science and Technology: A Foundation for Homeland Security’,White House Office Science and Technology Policy (April 2005), http://www.ostp.gov/html/OSTPHomeland.pdf ‘Fighting Bioterrorism: Tracking and Assessing US Government Programs’, Chemical and Biological Arms Control Institute (December 2004) Aloise, Gene, ‘Combating Nuclear Smuggling. Efforts to Deploy Radiation Detection Equipment in the US and Other Countries’, Testimony Before the Subcommittees on the Prevention of Nuclear and Biological Attack and on Emergency Preparedness, Science, and Technology, Committee on Homeland Security, House of Representatives, US Government Accountability Office, (21 June 2005) http://www.gao.gov/new.items/d05840t.pdf