Discussion of Terrorist Threats to Urban Infrastructure and Relevant Science and Technology Responses
M.K. Narayanan and Richard Garwin,
M.K. Narayanan, discussion moderator, stated that in his experience science and technology help in efforts to strengthen key aspects of urban infrastructure, but there was still a problem of matching technology with intelligence; the latter always seemed to follow, rather than precede, a terrorist event. There is a major divide between the world of the intelligence practitioner and that of the scientist and technologist. Except for electronic intercepts, Narayanan stated that he was not sure how helpful technology had been to the intelligence community. Intercepts and phone taps and other intrusive devices have improved over the years, and India has used electronic jammers to protect very important persons. However, these uses are few and far between, and more are needed. Narayanan noted the assistance of the Defense Research and Development Organization (DRDO) in developing profiling techniques, in distinguishing between the kind of explosives used by different groups, their preferred weapons, and so forth. In all this we received a great deal of assistance, but there are areas of vulnerability. The Indian estimate is that in about one-third of the cases, airport baggage screening is inadequate, and profiling of passengers has proved to be extremely inadequate. Even detection of fake passports has proven to be inadequate.
Narayanan noted that for road protection, mentioned by Lawrence Papay, technology has not noticeably assisted in the reduction of terrorism, as is evidenced by the six or more attacks on the Banihal tunnel on the strategic Jammu-Srinagar highway. Newer threats include maritime shipping, where it is estimated that nearly 90 percent of all the transport of goods is by sea. India knows, via both human intelligence and intercepts, that terrorists are paying more and more attention to shipping lines, which have generally been outside their purview of major attacks—only the Liberation Tigers of Tamil Eelam (LTTE) have done this. As for container ships, this is another problem: when will we take steps to deal with the problem, and how many container ships are going to be blown up before that? There has to be a closer marriage between science and technology and container safety, but the biggest problem is that there is a certain
unwillingness of scientists and technologists to associate with the intelligence community, and even more so with the police community. There is also a feeling of intellectual jurisdiction: the technical missions depend on the relationship between the scientific advisor and maybe the director of the intelligence bureau, but it stops with them; beyond that, nobody is really interested.
Narayanan observed that in India there was concern about the threat from radiological dispersal devices (RDD), which may be a bigger threat than is commonly believed. India has very few emergency operations centers, and those that exist are ill-equipped to deal with an RDD. While India’s nuclear installations are well protected, the tracking of radioisotopes, which can be used in a dirty bomb, in hospitals and other places is lax, partly because of the cost, and there have already been several instances of theft of radioactive materials from hospitals and other facilities.
Moving from the radiological to the biological, Narayanan observed that there was a total vacuum in understanding about probable threats. Most people do not really know what constitutes a biological weapon, and it frequently takes some time to decide whether an attack has occurred because of natural causes or a terrorist effort. Biosecurity is certainly one of the areas where Narayanan thought there was a clear case for marrying technology with the protective arms of the state, the intelligence services, and the security forces.
Narayanan observed that as for the role of the private sector in both biosecurity and radiological security, there was a very important role, but he doubted whether any of the private laboratories or research centers were fully equipped to respond properly. We need the help of science and technology, but how much of it is available within the next few weeks or the next few months? This is really the question. As for cyberterrorism, he did not believe that India was fully prepared to respond.
Richard Garwin, discussion moderator, focused on two problems: the development of the “smart” container and the vulnerability of the electrical grid.
He noted that existing technology, such as global positioning systems, bar codes, and Radio Frequency Identification (RFID) could allow for comprehensive tracking of containers around the world. Today, few containers coming into the United States are tracked at all (in the sense that the container does not have a very visible number on it). Containers have a numbered seal, and before the container is taken off the ship or sent via another mode of transportation, the seal is painstakingly read and compared with the manifest, which is sent to the United States before the container arrives. About 11 million containers a year enter U.S. ports; each of them costs on average about $1,500 to manufacture. A refrigerated container costs about $3,000, and a container is used about 50 times over its 10- or 15-year life. So the amortized cost, if you take a refrigerated container, is only about $50 to $100 per transit. The cost of transatlantic or transoceanic transport ranges from $500 to $1,500, and has fluctuated by $500 during the last few years. Nevertheless, people in the industry complain that if an additional charge of $50 per container transport were imposed, it would ruin the industry. This is nonsense, as the rates and costs fluctuate by as much as 10 times that amount per transport without any effect on the industry. Indeed, a U.S. importer, Tommy Hilfiger, uses refrigerated containers not because the shipped clothing would suffer without it but simply because
refrigerated containers receive expedited treatment, as all refrigerated containers are unloaded first. These refrigerated clothes arrive a few days ahead of time, and that is worth the extra cost.
What ought to be done? Citing former Coast Guard Commander Stephen Flynn, Garwin stated that for $300-$500, someone could make the kind of tracking unit that Lawrence Papay described. It would include a seal, a tag, and an interior sensor that would indicate if there has been any tampering or entry. Getting around these devices would require more sophistication than has so far been exhibited. With this smart container approach, shippers who have signed up to provide effective verification of the manifest at the time that the container was loaded, and subsequently tracked, would get preferred shipping. For example, they might be able to enter ports where other containers could not; their containers would be offloaded more rapidly, while other shippers’ containers might be shunted for unloading and inspection at the shipping ports rather than at the ports of entry. In this way shippers should be able to reduce the costs associated with shipping even though they pay more for the container initially. If the $300 or $500 is amortized over 50 shipments it is a negligible cost, but the amortization costs must be included because the tracking devices will become technologically obsolete in a couple of years, like ordinary personal computers. Even though the container might last 10 or 15 years and might continue to do the jobs for which it was bought, it would not be kept that long.
Addressing the problem of electric grid security, Garwin agreed that India and the United States had different vulnerabilities. In India, disruption occurred frequently, and there were ways of coping with the problem. However, in the United States, so much of the excess had been trimmed from the system in the interest of profitability that the United States was very vulnerable, especially to simultaneous disruption in several places. The United States needs to return to something simpler: rather than optimum control, it needs control that is good enough. This could be achieved by converting the system into a set of islands, an island being a generating capacity, and a corresponding load system. This concept is commonly called Distributed Generation, and is being employed in some areas currently.
Because there is only a finite amount of generating capacity, there is a degree of energy in the spinning reserve in the short run (the kinetic energy of the rotors, the connected loads are also sources of energy). This is on the order of milliseconds to a fraction of a second. Beyond that, power at electronic speeds is obtained from elsewhere in the grid. In a Direct Current system, this happens automatically. In an alternative current (AC) system, it is much more complicated than that. Rather than have the system go down, it would be far preferable to cut connections and shed load instantaneously, so that whatever live-generating capacity is locally available feeds a corresponding amount of load. After that, resynchronization must occur—a complicated problem. However, a large fraction of the system will continue to operate. So either the customer’s facilities must have commandable load shedding or it will have to be done with switches belonging to the utility or the transmission-distribution system, block by block or over large areas in the environment. That is something that needs to be examined if we are going to face either natural disruption or a terrorist attack.
Transformers are a choke point, Garwin observed. Transformers for power plants are very efficient because a gigawatt power plant produces $300 million worth of product
per year at a few cents per kilowatt-hour times 6,000 gigawatt-hours of electrical energy per year. One-half percent of this $300 million will be $1.5 million per year, which will amortize a $15 million transformer. Extra High Tension transformers of up to 1,000 megavolt-amperes are available; 500 megavolt-amperes are common. They are very expensive but very efficient. One three-phase transformer is more efficient than three single-phase transformers because the core is used to better effect, but it would be better to lose one single-phase transformer than to lose the generating capacity of a three-phase transformer for months or for a year or more. Thus, it is far better to replace a transformer with one that is 95 percent efficient (single-phase) instead of one that is more than 99 percent efficient (three-phase). A simple analysis, taking into account the cost of electricity, the cost of the transformer, the variation of transformer cost with efficiency (if the cost of a transformer is proportional to one over the inefficiency, so a gigavolt-ampere transformer may cost $5 million dollars at 99 percent efficiency and $10 million at 99.5 percent efficiency) then we find that the optimum efficiency is about 99 percent and that the cost of a transformer can be about $10 million in order to minimize the expenditure overall. However, it would cost one-tenth this much to replace that transformer with one that was 90 to 95 percent efficient. This will require more intensive cooling because of the loss of a lot of power in that transformer, but if they are stacked in modular fashion out of single-phase transformers, they would be a lot cheaper to stockpile for emergency use. This kind of analysis would suggest that somebody should be early into the market to start making them so that they can be stockpiled and rapidly transported, and they would be much lighter than the highly efficient transformers, easier to transport and easier to erect. We need to do this kind of analysis jointly if we take seriously the damage that can be caused by terrorists to our infrastructure.
General Paul shared his analysis of the September 11, 2001, disaster and the response of those in New York. Under the auspices of the National Institute of Advanced Studies, he spent 3 days at Ground Zero some 2 years after the event, and visited Washington, D.C. He was impressed by the way in which the Federal Emergency Management Agency, the state Emergency Management Agency, and the New York City disaster organization made decisions on how to respond to the communities needs; they had a structure there, on the ground.
Paul also suggested that innovative thinking had to take place about the evacuation of high-rise buildings, such as those found in New York City and Mumbai. People in such buildings have to know whether they should go upward (perhaps to helipads on the tops of such buildings) or downward, or to some innovative lateral evacuation system. These are areas where engineers and science and technology can be of help.
He also pointed out that the Indian response to disaster or terrorism did not seem as efficient as that of the United States, where there are standby task forces with adequate equipment, or some European countries, such as Germany, which offers disaster management service as an alternative to military service.
S. Gopal wondered about the utility of science and technology in coping with terrorism. For example, despite the fact that all required technology was in place to avert an incident such as the September 11, 2001, attacks (radar coverage, awareness of the deviation from flight paths, and so forth), the attack was not able to be averted due human failure. He asked whether technology would be able to mitigate and compensate for
human failures? Gopal was also not sure that biometric detection would be 100 percent foolproof. He felt that the possibility of harassment of innocent people needed to be avoided by fine tuning technology
As for Narayanan’s concern about the theft of radio isotopes and their possible use in dirty bombs, it would seem, according to Gopal, that sensors and other techniques should be sufficient to take care of this. In India the occasional cases of the loss of radioisotopic material have been from carelessness, but regardless, even if somebody steals some radioactive isotope materials, many of these isotopes, such as Carbon 14, are not really effective for making a dirty bomb. With a little care and lots of sensors this problem can be mitigated.
Papay responded to these points by noting that more and more attention and investment is now going into technologies such as interceptors. This might not be as useful in a rural environment, but it is helpful in thwarting international terrorism. As for the use of biometrics or improving passports, they are at an early stage. As many as 35 percent of all passports have been falsified, which shows that passport technology lags behind that, for instance, of credit cards.
B. Raman developed his ideas on the difference between the Indian approach and the U.S. approach to counterterrorism. The difference stems from the impact of September 11, 2001, on the U.S. mindset. These attacks affected U.S. thinking more than the April 1995 Oklahoma City bombing and the February 1993 attack on the World Trade Center in New York City. Until September 11, 2001, Americans believed that nothing much could happen there; afterwards they began to plan on the assumption that anything could happen. The U.S. approach is to identify all areas of the infrastructure that could be vulnerable to a terrorist attack and take whatever action is required to protect them, even if there is no specific intelligence information of an impending threat from a terrorist organization to that infrastructure. The Indian approach is to identify various aspects of the infrastructure that are vulnerable, identify those that have to be protected, whether there is intelligence of an impending terrorist attack on them or not (for example, the nuclear infrastructure, transportation, civil aviation) and for the rest of the infrastructure, take protective action only if there is specific intelligence of a likely threat from a terrorist organization.
Raman cautioned that Indians needed to think more about the threats and needs to counter them in the medium and long term. Some of the things the United States is now worried about may not be relevant to India today, but they could become relevant in 5 years or so, and to avoid a nasty surprise, as the United States had on September 11, 2001, India has to learn from U.S. experience. Instead of being complacent that this kind of attack is not relevant, that it could not happen here, we should plan on the assumption that it could happen to us tomorrow. From that point, Raman observed, it is important to identify the areas where we lack science and technology capability, and take action to build them even though they are not required today. India does not have the same level of financial and technical resources as the United States. The United States was able to respond quickly after September 11, 2001, but for India it will take much longer.
Roddam Narasimha commented that these thoughts led him to two suggestions for projects where he thinks India might be able to contribute significantly. Biometrics is one, the other is data mining, fusion, and management. This is central to the operation of intelligence services. There are many Indian experts in these areas, but their link is often
stronger with foreign customers than with Indian customers and in particular with public sector customers.
Devises such as electronic interceptors and jammers are crucial. They can play an extraordinary role in fighting terrorism in India, and while electronic systems are now in use, Narasimha does not believe that they exploit the potential Indian strengths in this area. Surveillance is another area worthy of future discussion because it is an extremely important issue for counterterrorism in India. Based on some Indian strengths, particularly in subareas of intelligence, and those of the United States, such as in technology, joint projects could be very beneficial. There seem to be certain areas where India has strengths that might become more evident through joint projects.
In response to Garwin’s comments about grids, islanding, load shedding, and so forth, Narasimha added that Indians have a great deal of expertise in these areas, because they are forced to live with a transmission system that is very rational, given that power transmission companies do not make a profit on the power they supply to farmers. Therefore, they are not interested in improving the reliability. The more money they put into the system, the more money they lose. So if we examine the policies of the State Energy Commission in Karnataka, we find that the policies being followed are very rational in view of what is in their interest. Therefore, they have developed all these methods of living with an unreliable system; a lack of reliability is actually profitable for them. There might be considerable U.S. interest in these Indian methods.
Paul made the very important point that although in India natural disasters are frequent—in the last few years, there has been at least one major disaster every year— India is not yet well prepared to address them. This is an area where Indian experts could learn a great deal from the U.S. system of disaster management.
Garwin observed that there were two aspects of surveillance as it pertains to the movement of people: authentication and identification. Authentication is not so difficult—there can be a picture, a retinal scan, or a fingerprint—but this raises the problem of an adequate database. Some people advocate fingerprints; others, retinal scans; others, automated photo-identification, and so forth. Yet if we utilize only a single modality, that freezes the system so that it cannot evolve. Perhaps it would be best to have two biometric indicators on a passport or identification card; fingerprints are very stable and have been reliably used for many years, but they may be replaced by retinal scans or iris scans. Authentication is the process by which the identity of a person is verified: finding the biometric, that is, a face picture or fingerprint, should determine who the person is. First, that person has to be in the database. Second, the accuracy and validity of the database have to be very much greater to identify one in a million or more. Fingerprints are good enough, but facial recognition is far from adequate now.
Garwin also commented on sensor grids for the detection of radioactive sources that could be used in a radiological dispersal device. RDDs are not necessarily bombs, because explosives are not an efficient way to distribute radioactivity unless the radioactive material is a gas. A much better way of distributing radioactivity is with an atomizer (nebulizer), and this is also much less obvious. Most of the sources that are available in industry and medicine are gamma ray sources, which are difficult to shield, especially expeditiously. The U.S. Department of Energy has revealed that at the end of 2003 it deployed large teams from national laboratories with search sensors, some of them in grids, and they found one radioactive source as a result. It turned out that a
homeless man had found a stainless-steel source 4 years before and kept it with the rest of his worldly goods in a lockup self-storage system where he slept during the day, with his source under his pillow. This was a radium source for the treatment of uterine cancer, that was probably designed to produce about 50 Rem/hr1 at a distance of about 3 centimeters, hence about 1.2 Rem/hr at 20 centimeters.55 It is much more difficult to find some sources that would be particularly suitable for radiological dispersal devices, that is, alpha emitters, that are not detectable from a distance and could be more easily shielded, but they are far less widely used. It is a good idea to share experience in this area as well.
S. Rajagopal raised the question of environmental sampling and moving detectors. In response, Garwin noted that these were explored several years ago, and a grid of sensors, or having them placed on buses (for early deployment), made sense. However, the market for sensors is very small and they cost about $1,000 each, but the price could be significantly reduced if they were made in China, Singapore, or India.
Narayanan clarified his position. While science and technology played a role in counterterrorism, especially in electronic intercepts, we do not know what is available unless scientists offer their expertise. Narayanan stated that this ought to be a major theme of the workshop: scientists should explain what they have and for what use, and this will encourage greater deployment of science and technology to counterterrorism.