American Experience during Hurricane Katrina and Superstorm Sandy
Karl Kim began his presentation by describing the National Disaster Preparedness Training Center (NDPTC), which is a Federal Emergency Management Agency (FEMA) center and part of the National Domestic Preparedness Consortium, which he directs. The consortium was created after the Oklahoma City bombing and was greatly expanded after the September 11, 2001 (9/11), terrorist attacks.1 After Hurricane Katrina, the University of Hawaii was added as one of the consortium’s members. The NDPTC at the University of Hawaii develops and delivers training courses on natural hazards and various tools related to disaster risk reduction, response, and recovery. The center also works on threat and hazard identification and risk assessment tools to integrate both a national perspective and state and local perspectives. Geographically, the center focuses primarily on the Asia-Pacific region, one of the most dynamic in the world, in which there are many different regional entities and groups organized for security, trade, environmental issues and so forth.
Security of the American homeland arguably started in the 1920s, focusing on civil defense and in the 1970s FEMA was created, and there was an increased awareness of natural hazards. Terrorism events—the Oklahoma City bombing, and the attacks of 9/11—broadened the scope of what the organizations that became the Department of Homeland Security covered. The responsi-
1The U.S. Federal Emergency Management Agency. National Domestic Preparedness Consortium. “Preparing the Nation through Training.” April 29, 2014. Available at: https://www.ndpc.us/pdf/2014%20NTE%20PRESENTATION%20TEMPLATE_CS3%20NPDC%20Final.pdf; accessed September 21, 2014. See also, “Letter from the Chairman.” NDPC News, Vol. 3, Issue 3, Summer 2008. Available at: https://www.ndpc.us/pdf/NDPCNews3.3.pdf; accessed September 21, 2014.
bilities are very broad and that scope tends to be event-driven; one of the largest recent events in the United States was Hurricane Katrina, which caused FEMA to further evolve. The current framework for emergency support functions in the United States is laid out in the Presidential Policy Directive 8, the Stafford Act, and the National Preparedness Goal, and recently the emergency support functions have widened to include recovery support functions.2
The Asia-Pacific region also has the greatest rates of urbanization in the world, and some of the fastest rates of change are occurring in countries like China and India. Kim believes that cities should be units of analysis for studies related to emergency response because they are critical leverage points, and because cities allow one to maintain perspective better than if the unit of analysis was a large geographic area such as a country or the world. Cities are also a source of solutions, not just sources of challenges.
Kim then turned to a comparison of two significant events in detail: Hurricane Katrina and Superstorm Sandy. As Hurricane Katrina passed over the Bahamas, it intensified and first made landfall in Florida on August 25, 2005, and then made landfall in Louisiana on August 29, and it actually weakened when it hit Louisiana. Superstorm Sandy was different. It also developed in the Caribbean, and then its peak intensity occurred prior to landfall in Cuba when it became a hybrid storm. It made landfall in the United States on October 29, 2012. Hurricane Katrina was a very compact and intense but big storm, whereas Superstorm Sandy was much larger and in many ways diffuse, causing a variety of effects. It caused wind in some areas, but it also brought snow. Hurricane Katrina had more significant winds, whereas Superstorm Sandy had everything from blizzards in the Appalachian Mountains to flooding in Manhattan. With both Hurricane Katrina and Superstorm Sandy, there was an increase in the seawater level, causing a storm surge. More than 1,800 people were killed as a result of Hurricane Katrina, and damages totaled approximately $108 billion, which made it the costliest hurricane in U.S. history.3 As a result of Superstorm Sandy, 147 people died and damages totaled approximately $50 billion, but there were also
2U.S. Department of Homeland Security. National Preparedness Goal. 1st ed., September 2011. Available at: http://www.fema.gov/media-library-data/20130726-1828-250459470/national_preparedness_goal_2011.pdf; accessed September 21, 2014. And, U.S. Department of Health and Human Services. Public Health Emergency, Emergency Support Functions. “Emergency Support Functions is the grouping of governmental and certain private sector capabilities into an organizational structure to provide support, resources, program implementation, and services that are most likely needed to save lives, protect property and the environment, restore essential services and critical infrastructure, and help victims and communities return to normal following domestic incidents.” Available at: http://www.phe.gov/preparedness/support/esf8/pages/default.aspx; accessed September 21, 2014.
3CNN Library. Hurricane Katrina Statistics Fast Facts. Available at: http://www.cnn.com/2013/08/23/us/hurricane-katrina-statistics-fast-facts/; accessed September 21, 2014.
significant effects on peoples’ lives: an estimated 8.5 million people were without power, 650,000 homes were destroyed, and so forth.4
Kim was part of a FEMA reconnaissance team deployed immediately after Superstorm Sandy. The storm significantly affected two types of areas: urban areas and coastal areas. In urban areas like New York City, Lower Manhattan was flooded, basements were flooded, power systems were knocked out, and subway tunnels were flooded. On the coastal areas, the significant storm surge and flooding destroyed many buildings and structures. Disproportionately, elderly, people with disabilities, and poor people were negatively affected in this storm. An evacuation order should have been given in Kim’s view, due to how quickly the water level rose and the threat of casualties in the New York City area. There was an assumption that Superstorm Sandy was going to behave a lot like Hurricane Irene, but it did not.5
Rae Zimmerman at New York University, together with the American Institute of Architects in New York, is mapping the locations of a variety of types of urban infrastructure, such as electrical power plants, wastewater systems, and transportation systems located below 10 feet or in coastal areas prone to flooding.6 A plan was proposed by Jeroen Aerts of the Free University of Amsterdam to build a floodgate system to protect Manhattan for between $11 billion and $23 billion.7 The question is, where will that water go if it is prevented from going into Manhattan?
Kim then compared Superstorm Sandy to previous hurricanes in terms of the number of homes destroyed, the damage in dollars, the number of people evacuated, and the number of fatalities. When compared to Hurricane Camille (category 5, one of the biggest storms),8 and to Hurricane Ivan, a category 3
4Blake, E., et al. “Tropical Cyclone Report, Hurricane Sandy.” National Hurricane Center, National Oceanic and Atmospheric Administration, February 12, 2013. Available at: http://www.nhc.noaa.gov/data/tcr/AL182012_Sandy.pdf; accessed September 22, 2014.
5Rice, L. “An Analysis of Public Perception and Response to Hurricane Sandy” (master’s thesis, Univ. of South Florida, 2014). Available at: http://scholarcommons.usf.edu/cgi/viewcontent.cgi?article=6310&context=etd; accessed September 22, 2014.
6Zimmerman, R., and Faris, C. “Infrastructure Impacts and Adaptation Challenges.” In New York City Panel on Climate Change 2010 Report, pp. 63–85. Available at: http://macaulay.cuny.edu/eportfolios/bird2012/files/2012/07/Infrastructure-Impacts-Adaptation-Challenges.pdf; accessed September 22, 2014. See also: NYC Planning. Coastal Climate Resilience, Designing for Flood Risk. Department of City Planning City of New York, 2013. Available at: http://www.sustainablenyct.org/news/NYCDCP_DESIGNING%20FOR%20FLOOD%20RISK_DRAFT-LOW.pdf; accessed September 22, 2014.
7Aerts, J., et al. “Cost Estimates for Flood Resilience and Protection Strategies in New York City.” Annals of the New York Academy of Sciences. New York: New York Academy of Sciences, 2013. doi: 10.1111/nyas.12200. Available at: http://www.ivm.vu.nl/en/Images/nyas12200_tcm53-364896.pdf; accessed September 22, 2014.
8Associated Press. “Hurricane Camille’s Peak Winds at Coast Downgraded.” Washington Times, April 9, 2014. Available at: http://www.washingtontimes.com/news/2014/apr/9/hurricane-camilles-peak-winds-at-coast-downgraded/; accessed October 20, 2014.
storm,9 many more people evacuated during Superstorm Sandy. Hurricane Katrina was not particularly damaging along the coastal area of Mississippi, but with the effects of flooding and the breaking of the levy in New Orleans, many more people were killed. Kim also compared Hurricane Katrina to a high-end typhoon in the Philippines, which caused 6,200 deaths.10 The damages were much lower in terms of dollar value, only about $15 billion, but many more people were displaced and adversely affected in the Philippines.11
Kim is working on a federal highway project that involves assessing effects on transportation by examining different risks, hazards, and threats, then focusing on the inventory itself and what needs to be protected. The same kind of logic that applies to security assessments applies to this analysis as well. In order to conduct long-term planning for prevention and response, Kim and his colleagues are modeling various effects of sea-level rise, hurricanes, storm surge, tsunamis, and river flooding on cities to determine the impact of an overall rise of 1-meter by 2100, which most experts agree will occur. They are interested not only in average sea levels, but also in the increase during extreme events; therefore they have been examining the effects of recent flooding that is not related to storm surge or surf. Flooding occurred at Waikiki Beach in recent years and in the winter of 2013, which caused significant damage. The water drainage system is backed up with sea-water and the resulting effect is being seen.
One of Kim’s colleagues at the University of Hawaii has shown that as the sea level rises, the freshwater level also rises and produces increased flooding in urban areas. Therefore, much more of the urban area will be flooded due to the rise of groundwater levels, as well as the intrusion of seawater. As a result, part of the mapping involves trying to understand the combined effects of sea level rise and tsunamis, and sea level rise and other hazards. Further, Kim and his colleagues are using asymmetric mapping techniques to identify the exposed population by flooding depths and then applying that information to an assessment framework to attempt to predict possible events, focusing specifically on criticality. Which place will receive the worst flooding? Which place will be most negatively affected? Where do seniors live? Where are the most socioeconomically vulnerable people located? Where are critical infrastructure systems located? It is very difficult to predict the locations, but we know the most critical
9CBS Miami. “Florida Marks the 10th Anniversary of Hurricane Ivan.” September 16, 2014. Available at: http://miami.cbslocal.com/2014/09/16/florida-marks-the-10th-anniversary-of-hurricane-ivan/; accessed October 20, 2014.
10CNN Staff. “Typhoon Haiyan Death Toll tops 6,000 in the Philippines.” December 13, 2013. Available at: http://www.cnn.com/2013/12/13/world/asia/philippines-typhoonhaiyan/; accessed October 20, 2014.
11“Of the 15 million people in the Philippines affected by Typhoon Haiyan, around 4 million have been displaced from their homes, including more than 94,000 living in 385 evacuation centres.” United Nations Children’s Fund. Available at: http://www.unicef.org/infobycountry/philippines_71516.html; accessed September 22, 2014.
systems in terms of demand, so combined risk scores are developed to understand how to apply the framework.
Kim focuses specifically on trying to understand travel behavior. For example, based on travel demand surveys, Kim can determine that all of a person’s trips in a given day are in the flood zone, and then he maps trip origins and destinations by flooding depths and models travel demand during hazardous events. It is very difficult to simulate behavior during an event itself; therefore, what he has been trying to do is determine how to integrate global forecast information and other data into the framework.
Another application of Kim’s research is to integrate public alert systems including not just radio and television, but also smart phones and other devices to assist in evacuation modeling. All of the sirens and warning detection systems are designed primarily for tornadoes and tsunamis, less for flooding in vulnerable communities. One of the problems he found was that many of the evacuation centers or shelters are located in the flood zone. Which responses, including detection, warning, and evacuation are most effective? What is the range of strategies that can be used to improve responses?
Kim and his colleagues have been developing a tsunami preparedness model that focuses on nuclear power plants and associated issues. Kim noted that in the Japanese case, multiple precautions were taken, and almost every one of them failed. Kim’s future work involves refining this information and focusing on critical lengths or segments of urban environments because obviously with sea level rise, more areas will be flooded. He and his colleagues are moving away from the paradigm of “fail-safe” to “safe-to–fail” approaches, rethinking zoning or no-build zones.
In conclusion, Kim underscored that urban technology is getting faster and smaller, and has increased connectivity. The notion is that humans are now the ultimate sensors through personal electronic devices such as smart phones, from earthquake detection to information about storms. These technical changes also change notions of security, and today, almost everyone has a camera on a smart phone. Even though there is some resistance to security systems, video cameras are prevalent. This is actually an old concept, called the panopticon, which was a system of construction designed by Jeremy Bentham for surveillance in prisons, schools, and factories. Bentham’s idea was that one could, by design, create an environment where the people are knowingly always observed, resulting in complete security and no privacy.12
Indian Experience during the 2013 Himalayan Tsunami
Shri Vinay Kajla prefaced his remarks by stating that he is a member of the Central Industrial Security Force (CISF), which handles security for India’s
12University College London. “The Panopticon.” Available at: http://www.ucl.ac.uk/Bentham-Project/who/panopticon; accessed October 20, 2014.
airports, seaports and all Indian nuclear power installations. CISF was one of the organizations that responded to the Bhuj earthquake in 2001 by addressing response needs at the airport. Kajla led a small team of approximately 320 people from his fire department in that response. Their first observation was that there was a large number of aircraft with a great deal of equipment but there was no one to unload it. Initially, there was reluctance by the air force to accept help from Kajla’s group, but eventually his company of 320 men unloaded nearly 120 aircraft. The International Red Cross wanted to put up a 400-bed hospital and with the help of Kajla’s team they were able to get it going. The CISF team also helped a 50-bed Israeli hospital become operational. Rescue teams arrived from several countries, including Denmark, Switzerland, and the United States, bringing specialized equipment that India did not have.
The severe floods of June 2013 affected five districts13 and were caused by extensive rainfall that fell in just 48 hours (see Figure 5-1). Normal annual rainfall is approximately 26 millimeters, whereas in this period alone there was 400 millimeters of rainfall. The heavy rainfall triggered a large number of landslides and even the movement of a glacier. Kajla noted that in the most affected town, Kedarnath, nearly 5,000 people died, including numerous tourists. In addition to the loss of human life, many mules that carried tourists to temples also perished. Large construction projects had been built all along the riverbed itself, which also contributed to the vulnerability of the riverbanks to landslides. The local temple was located on a foundation of large, heavy stones, but was washed down the hill within 3 minutes, starting at 6:37 a.m. on June 16, 2013, leading to the total destruction of the towns below. A brand new building unfortunately no longer exists because it was constructed too close to the river bank. This is a clear example of the importance of zoning and adherence to zoning regulations.
Given the destruction of nearly all of the roads leading to the site, the only way to evacuate victims was via one of a few Mi-17 helicopters that can carry 25 people or perhaps 30 if the helicopter is carrying less fuel weight. This limited capacity meant that there was a large number of people waiting to be evacuated. Although relief flights rapidly responded with deliveries of food and other necessities, during the initial 3 days the response from the state governments proved to be insufficient to meet the needs. Furthermore, heavy rainfall continued, and prepositioning of fuel at the airstrip below was inadequate to allow flights to keep up with demand. When combined, these operational challenges limited the effectiveness of the relief response. The response effort proved dangerous not only to those who were attempting to evacuate, but also to the National Disaster Management Authority (NDMA) responders. In one case, an Mi-17 helicopter crashed, killing 20 people, of whom 9 were Kajla’s colleagues from NDMA. Reflecting on the implementation of response efforts during this
13The five districts were: Bageshwar, Chamoli, Pitho-ragarh, Rudraprayag, and Uttakashi.
FIGURE 5-1 Heavy rains on June 16 and 17, 2013. The floods of June 2013 affected five districts in India; the town of Kedarnath was most affected with 5,000 deaths. SOURCE: Kajla, 2014.
disaster, Kajla concluded that there was fairly good coordination among the Indian Army, Central Paramilitary Forces, Indo-Tibetan Border Police (ITBP), and National Disaster Response Force (NDRF). ITBP responded with five battalions. NDRF took in 450 responders and evacuated a large number of people over specially constructed bridges, and using their helicopters.
Among the challenges India faces in preparing for and responding to such incidents and disasters is the equipment procurement process. As a result, good communication equipment such as mobile towers at the site of a disaster, and portable reverse osmosis drinking water systems unfortunately are not available to first responders.
The fifth meeting of NDMA, held in October 2013, strengthened their commitment to disaster preparedness by, for example, creating a network of available helipads in advance of an incident. They will try to earmark funds for better observational and forecasting capabilities, improvements in communication infrastructure, restoration of roads, and the building of multidisaster shelters. These are the key areas on which the government of India would like to focus in the next 2 to 3 years to improve preparedness in the high-altitude and hill states.
The government is also encouraging those responsible for private structures, such as religious sites, to prepare for disaster in advance as well. There are temples designed for a capacity of 1,000 people, but at times hold more than 7,000 people. This creates significant vulnerabilities; in one case, a stampede occurred and 250 people died while trying to evacuate a temple in the midst of a natural disaster.
Kajla concluded his remarks on the flash floods known as the Himalayan Tsunami by stating that it will be a long process, but India is making great strides in improving its preparedness and response capabilities.
Indian Experience during Cyclone Phailin
Kajla provided a brief presentation on improvements in disaster response as seen during the category 3 hurricane, Cyclone Phailin, that hit in October 2013.14 This hurricane was the second most powerful hurricane in Indian history after the Super Cyclone of 1999.15 Despite its force, only 25 casualties were reported as a result of Cyclone Phailin as compared to 9,894 casualties caused by the Super Cyclone.16 This success in the number of lives saved is attributable to the fact that the advanced evacuation plan took a large number of people out of harm’s way, and those who remained were protected in cyclone shelters. Kajla noted that this is where technology aided the government’s preparedness efforts.
Accurate forecasting by the India Meteorological Department significantly contributed to responders’ ability to concentrate preparedness efforts in the areas of greatest need. Teams were able to be positioned with necessary equipment well in advance of the storm. Specifically, the cyclone flood shelter was staffed with volunteers and had appropriate equipment. The Puja holiday of Dussehra is an important festival and fell just a day prior to Cyclone Phailin and so all activities were canceled by the Odisha State Disaster Management Authority, so as to take proactive steps. Free food stations were established and the evacuation process intensified and was concluded over 4 days. Section 34 of the Disaster Management Act empowers authorities to force people to evacuate wherever they are reluctant to move to safer locations.17 In the final stages of preparedness, this statute was invoked by local officials helping to ensure the low fatality rate.
A large number of people responded and were well-coordinated with well-established communications. NDRF responded with 53 teams totaling 2,500 people, the largest-ever deployment in India to date. When the cyclone did hit, formal reports of damage in 17 districts indicated that there was a great deal of
14Cyclone Phailin made landfall near Gopalpur with a wind speed of 200 km/h.
15The 1999 Odisha Cyclone was the strongest tropical cyclone ever recorded in the North Indian Ocean. It developed over the Malay Peninsula and on October 28, it became a severe cyclone and hit India the next day as a 155 mph (250 km/h) cyclone.
16The cyclone dumped heavy, torrential rain over southeast India, causing record breaking flooding in the low-lying areas. The storm surge was 26 feet (8 meters). It struck the coast of Odisha, traveling up to 20 km inland. It caused the deaths of about 10,000 people, and heavy to extreme damage in its path of destruction.
17For the purpose of assisting, protecting, or providing relief to the community, the district authority may control and restrict the entry of any person into, his/her movement within, and departure from a vulnerable or affected area.
infrastructure damage in numerous villages. An estimated 90,000 houses were destroyed, but human casualties were low. Estimates of the cost of the damage were lower than in other disasters and are attributable in part to the circular design of some structures. Highway and energy infrastructure damage was largely repaired fairly quickly. Restoration of the water supply did take time but the Odisha state government addressed this issue quickly.
NDRF helped with removal of fallen trees immediately after the storm cleared. At 6:30 the next morning, before people woke up, NDRF was already cleaning up the highway so that infrastructure maintenance and repairs could proceed. NDRF also established medical camps at the site, and used disaster management equipment to open the doors as quickly as possible.
Looking forward, the World Bank and the Asian Development Bank conducted a study that was presented to the government of India, in which they suggested using underground cabling because India still has a lot of overhead cabling in coastal states. Following the “build back better” principle, India will transition to underground cabling for power and telecommunication systems.
Nancy Jo Nicholas framed the discussion period by underscoring the importance of thinking about the link between counterterrorism and natural disaster planning and resilience, and adaptability and sustainability. She believes that definitions of terrorism also include attempts to terrorize, frighten or scare people; however, increasing levels of preparedness counters these threats.
A workshop participant asked about the distinction between preparedness and response to terrorist incidents versus natural events. How does one make those distinctions and where is it useful to make those distinctions?
Kajla stated that NDMA was constituted to deal with natural hazards; however, terrorist incidents that lead to a similar kind of mass fatality situation will also require a coordinated response by all agencies, including the Ministries of Home Affairs, Defense, and other core ministries.18 Any incident, whether man-made or naturally occurring, will require a humane approach to solve the problems of those affected. If agencies learn to cooperate and integrate, he said, most problems can be solved. Unfortunately, whatever lessons are learned along the way, they often take a long time to implement.
Kim replied that there are clearly differences in terms of the level of harm that can be caused by different types of hazards, storm surges, earthquakes, and terrorist attacks, and it is important to understand these differences. The concept
18Ministries and agencies include: Civil Aviation, Earth Sciences, Water Resources, Agriculture, Mines, Environment and Forests, Department of Atomic Energy, Heath and Family Welfare, Railways, Road Transportation and Highways, Urban Development, and the India Meteorological Department under Earth Sciences.
of criticality—the critical infrastructure systems that one cannot afford to lose for whatever reason—is essential.
Kim also pointed out that there is another type of hazard separate from a natural hazard or a man-made terrorist hazard: the hazard of poor planning or bad governance. Governance sets the context for how a risk assessment will be conducted, how the detection and warning systems will operate, and how communication with vulnerable populations will be addressed, which will ultimately determine the effectiveness of the response, especially in how quickly one can recover from these incidents.
A workshop participant noted that during and immediately after major disasters, space technology has a significant role to play, both in providing satellite imagery and in supporting communications for emergency terminals, satellite phones, and so on. In this context, one of the mechanisms for international collaboration that has been established over the years is the International Charter on Space and Major Disasters in which India and the United States are among 11 partner countries. The charter was activated for both Superstorm Sandy and Hurricane Katrina. Useful images were provided immediately after these events to assess the damage and also to inform appropriate response actions. Also, there was mention of the use of satellite imagery in the case of the Himalayan Tsunami. What is the effectiveness of the use of these images in tackling emergency situations?
Kajla replied that satellite images help in assessing what is happening, but from an operational point of view, real-time decision making has to happen on the ground. If there is a 13,000-foot mountain in front of you, images are good but you still need to walk up the mountain and take patients down from wherever they are stuck in the landslide. It also helps to assess aftereffects for reconstruction and rehabilitation, and long-term planning. In the immediate term, one has to be on the ground. Smaller unmanned aerial vehicles (UAVs) proved helpful in quickly assessing the situation, together with eyes on the ground. In one example, people were trapped between two landslides and even NDRF could not reach them and the army helicopter could not come close enough to the mountain. Finally, in the evening, a civilian helicopter managed to bring them out. For short-term issues, people are needed on the ground, but for long-term issues space images and satellite telephones are very important, because it will take days before normal communication returns.
Kim replied that there have been tremendous improvements in satellite imagery and access to this information. He pointed out that the imagery is very good for tracking and following a storm, but that is not the same as predicting the storm path. Even though we have become much better at predicting where the storms will make landfall, there was a controversy during Superstorm Sandy concerning predictions of the U.S. model and the European model. It is very easy to make predictions in the near term, but as one goes farther and farther out in time, predictions become more challenging. Kim’s group at the University of Hawaii is one of the few programs to conduct minisatellite launches and they are
also attempting to put multispectral sensors on UAVs or smaller, shorter-range devices.
Another participant asked, what are the lessons learned for improving training for emergency responders after these disasters? Kajla answered that training for a wider group of people is a key issue because the emergency responders from NDRF will take time to reach the site of a disaster. Therefore, training of state-level forces is critical because the community response is the first line of defense. Unless communities are better prepared, we will continue to suffer fatalities. There are practical issues with regard to various facets of disaster response, which unfortunately still have to reach the grassroots level; hopefully, technology will be able to provide other alternatives. The only problem is that training courses are often so boring that people sleep through most of them. Disaster management by itself is a very difficult topic, and it is something that people do not want to consider. There is a common thought, “If I did without a seat belt for the last 30 years, why do I need to wear one now?”
Kim replied that for first responders, there are really only two different types of training required. One addresses incident command systems, such as the U.S. National Incident Management System. Standard operating procedures are essential, as is the link between training and equipment. At another level, there must be decision making in periods of uncertainty and decision making under stress. How does one create an environment where responders can make the appropriate decisions when they have limited information? How do they make very quick resource allocation decisions? That level of training, creating that kind of culture, is much more difficult than training on equipment. Kajla added that there are other characteristics that are required, such as experience in disaster management situations, in, for example, handling large crowds.
A workshop participant asked: How can mobile devices and similar technologies, which tend to be in the hands of wealthier members of the population and not as readily available to other populations, benefit the whole community? Kim said that there is such rapid evolution in digital technology that it is very difficult to anticipate and predict all of the innovations that may be forthcoming. Technology is evolving, and many vulnerable populations now have access to mobile devices, so this provides one way to reduce some of the disparities. Kim added that one of the truly remarkable things that FEMA has done is adopt a “whole community initiative.” It is based on the recognition that the government cannot be all things to all people everywhere, and it is important to learn from communities and to empower them to take care of themselves. In the event of an emergency, chances are a resident or a community member would be the first on the scene. Also, there is a lot of indigenous knowledge, particularly about hazards and threats that exist in a community. In the race to adopt satellite communications and cell phone technologies, we should not abandon all of the wisdom and knowledge that exists with cultures that are particularly close to the natural environment and that have an understanding of the history and the place.
Further, what are the priorities if one has a limited amount of time but not a limited budget? What are some priorities for U.S.-Indian cooperation? Kajla
noted that it is difficult to identify priorities for international collaboration; however, a key area is training for communities. One can help people become aware of the simple steps that they can take. This has worked in the states of Assam and Bihar, which are prone to floods and where people wanted to save their communities. Therefore, on their own initiative, they made low-cost, low-floating aid devices that save their animals, resources, and lives and now have a very well-established system. Conversely, Kajla has been grappling with the problem of how to provide better water treatment systems during floods. The government does not own these systems but can procure them thereby enabling communities to recover more quickly from an event. This is an area that is open for cooperation.
A workshop participant shared experiences from the response to the 2004 Indian Ocean earthquake and tsunami. Nongovernmental organizations (NGOs) had a lot of money to rebuild infrastructure, but most built two-story buildings; Thai people in that particular region, however, were not keen to inhabit the second stories of the buildings. They used only one story and the second story often remained empty. Further, one of the NGOs, a heavily funded group, had a standard list of materials to be used, but the particular type of wood specified was not available, so funds were not made available for response to the tsunami.
Kajla replied that people in India had experienced the same challenges with international donations. In many Indian communities people do not wear jeans and tops although this is what was provided by international NGOs. Whatever we do in the spirit of help may not actually be helpful at the site of a disaster. It is important, therefore, to let donors know what they should do to help. Reconstruction is slow, and it is not a very publicity-driven process. Most people lose interest in a disaster site after a short while because the next disaster has already happened. However, a local NGO from the affected area will remain engaged because it is in its long-term interest to develop that area faster and better. Though committees are formed at the national level comprising all ministries, ultimately decisions will be made by the local people.
A workshop participant responded that perhaps local people may not have a very good decision making process. In most cities devastated in the Indian Ocean tsunami, people built their homes on the very same place, and the government is not stopping them. Kajla agreed that, typically, it is good to suggest that people rebuild in safer areas away from the water, but many will eventually return to their original location because they have no other place to go, there is a source of employment on the water, and they do not like to be far from their fishing boats.
Technologies for Countering Terrorism
K. Sekhar framed his remarks in the context of significant changes in India that began after 9/11. There was recognition that terrorism, left-leaning ex-
tremism, insurgents, and people with various frustrations needed to be addressed more effectively. Given the Defense Research and Development Organization’s (DRDO) research and development base, a decision was made to try to convert available technologies and to develop new technologies to help counter those threats more effectively with the singular objective of reducing the loss of lives. With this objective, a variety of paramilitary organizations were gathered to draft a program that identified the kinds of technologies that could be given immediately to fighting forces within a short period of time, approximately 6 months. It also addressed the questions: What are the technologies that can be customized over a period of 1 to 2 years and potentially be provided to the fighting forces? Within major, longer-term programs, where major technology breakthroughs might be made within 2 to 4 years? To address these questions, an assessment of possible threat scenarios is essential: The areas of vulnerability range from urban areas to forested areas, mountain regions, coastal areas, and desert areas. Indian air space is also vulnerable to infiltration by terrorists; hostage scenarios are of particular concern.
Sekhar listed recent terrorist events in India, including the 2008 Mumbai attacks and the landmine blast in Sandiwara, the Bengal blast, and the April 2010 Dantewada Maoist attack. In the process of examining these incidents and means of countering future terrorist attacks, Sekhar stated that efforts must start with surveillance and reconnaissance. For example, how can one improve the ability to fight at night, which is a particular challenge because most encounters with terrorists occur at night. Another challenge is establishing secure communications as well as gaining the ability to jam adversary communications. What kind of arms and ammunition are specifically required when engaging terrorists in open fighting? What are the nonlethal weapons that could be used to capture terrorists alive, and take them into custody for interrogation? Explosives detection and diffusion is also a very big challenge. One may detect an explosive; but diffusing it is more challenging. Personnel protection to support fighting forces and combat support systems are also necessary.
To address the surveillance and reconnaissance challenges, Sekhar stated that India already has developed a capability like the unmanned Nishant aerial vehicle that can be deployed, the Netra UAV, and the quadrotor helicopter, which is now being used very extensively by police personnel for crowd control, and battlefield surveillance radar, which can be used very effectively to control infiltration. Mountain radars, such as Bharani, also serve a similar purpose. Fixed-wing mini-UAVs are under development, and soon they will be operational.
Another immediate equipment challenge, especially in hostage scenarios, is through-wall imaging radar. The Israeli’s have developed a radar that is being explored by Indian experts. It is not very effective, although it provides an idea of how many people are inside a building. Police would like more information about what is transpiring on the other side of the wall than can be obtained with technology today. DRDO asked a laboratory in Bangalore with expertise in radar to develop technology that can monitor the movement of people based on
their heartbeat and breathing patterns. This could distinguish very clearly, for example, among people who are moving and people who are seated. This is anticipated probably in a year or two. Through-wall emitting radar will be one of the first known radars for monitoring movements inside closed rooms. This would have been a considerable advantage had it been available during the Mumbai attacks. In conjunction, a “corner-firing weapon” would also be helpful. Similarly, multimode grenades are useful both for offense and for defense. Air-bursting grenades can be launched from a distance of 200 to 300 hundred meters and they can explode precisely at the desired altitude.
The topic of surveillance is something that India has been working on for quite some time. Balloon-based surveillance is valuable, especially in areas that are not easily accessible over land. India has had considerable discussion with some foreign countries, especially Sweden, to work with SAAB, a pioneer in foliage-penetration radar. India is also working to develop moving-target indication to detect people moving under the foliage. This is a big challenge in India, especially if one looks at the central part of the country. The movement of insurgents and left-leaning extremists takes place under these conditions; therefore, this could become a very important technological tool.
Sekhar noted that there has been a great deal of discussion in India about the detection of explosives. One of DRDO’s laboratories, along with industry in Bangalore, is working on a point detector based on amplifying fluorescent polymers. They have demonstrated with many of the conventional explosives, like RDX, TNT, HMX, and PET, that even at very low levels of concentration, ppblevel detectors are sensitive to ammonium lactate. The major advantage with this type of detection is that it does not have a radioactive source and indicates the presence of explosives through a warning light. This technology is now ready, and within in 6 months to 1 year, it should be ready for deployment. Stand-off detection of explosives is another challenging area that many people have been working on, and another technology in development is based on laser photoacoustic spectroscopy, where the target is penetrated with tunable fiber-optic lasers, and the acoustic signal generated by the target is detected and analyzed. A sense of the target is gained based on the type of acoustic signals detected. This technology will likely take a few years to complete.
Left-wing extremists are a significant threat in central India. They have used hard-wired connection explosives rather than remote detonators and have laid wire for hundreds of meters and buried explosives at a depth of up to 2 meters. In some cases the explosives have been buried for months and years. Unless one has a method of detecting these explosives, there is always a risk. Terrorists see people coming and they sit about a hundred meters away physically connected to the explosive via hard wire. The moment people come to the point where the explosives are buried, the device can be detonated. To counter this threat, DRDO is trying to use well-known equipment called ground-penetrating radar, which is used to locate cables and pipes for construction activities. How can one effectively use ground-penetrating radar to detect explosives? DRDO’s future plans include examining different kinds of explosives, different contain-
ers, different depths, different kinds of signals, and different kinds of materials. They will then prepare a large database with this information so that when an actual signal is detected in the field there is a greater likelihood of successful prediction of the actual kinds of explosives buried and at what depth. Sekhar hopes that in the next couple of years, they will be able to have a reliable databank.
Detection of the explosive is only the first step, however. After detection, how does one handle the explosive? Obviously one needs a remote vehicle, and India has developed a remotely operated vehicle called Daksh, which is capable of entering a building, climbing stairs, picking up an object and taking it to a safe place, and placing it down. Cameras and other equipment can also be mounted to the remotely operated vehicles. Then, having detected a significant amount of explosives, how should disposal occur? A laser ordinance disposal system was developed to address this challenge: using a laser beam, an explosive can be deflagrated at a distance of 50 meters.
Next, Sekhar explained, they examined communication systems to ensure that they are secure and also so that they can jam the adversaries’ communication. They have developed an S-band mobile satellite terminal that is light weight (3 pounds) and that can be used anywhere. They have developed vehicle-mounted improvised explosive device jammers that have become a regular part of all BVA convoys in India. They have also developed a human-portable jammer so that people can enter areas not accessible by vehicles. In terms of acoustic eavesdropping, DRDO is developing technology to listen to conversations and detect the movement of terrorists hiding in forests or in remote buildings. They have developed a network of wireless acoustic sensor nodes placed over a large territory and then they are able to communicate to form networks. Significant intelligence capabilities have been incorporated into this network to have the ability to recognize specific voices. This technology is also in the process of being developed, and some positive results have been obtained. The nodes are able to communicate to a base station of the network, and the host at the base station conveys the communication to the control room. Should one or more nodes be damaged, the system will form another network and continue to communicate. Another technology is the optoelectronic-based eavesdropping system. With this technology, a laser beam is aimed at a suspected object like a window, and the voice inside modulates the laser beam, and the reflected beam provides an idea about the type of conversation. This has been successful up to approximately 20 meters in houses and cars, but the technology needs improvement so that it can be used for more applications over long distances.
Less lethal weapons are also very important, especially to flush out terrorists from hideouts. Oleoresin-based grenades were developed almost 6 or 7 years ago and are now being used extensively by India’s ground forces, especially in the border areas where terrorists are known to be hiding. The CR-based tear gas grenade is an advanced version of a lachrymatory agent commonly known as tear gas. The advantage of CR gas is that it is less toxic, but more effective. Normally, one can protect oneself by placing a wet cloth over one’s eyes and
nose; however, with CR gas a wet cloth aggravates the eyes and nose. CR will increasingly become a common element in tear gas.
Mine-protected vehicles are always being improved to make them as effective as possible. For example, today, Sekhar said, India is trying to develop vehicles capable of protecting people against explosions of up to 50 kilograms of TNT. However, the adversary may use 75 to 100 kilograms. This is why it is important to keep developing technologies, to send a strong message of confidence to the forces fighting the terrorists and to the terrorists about their capabilities. Also, bulletproof vehicles have become very important, and DRDO has developed vehicles that can withstand AK-47 fire as well as 7.2 SLR fire. Sekhar also noted the importance of personnel protection. Security forces need good bulletproof jackets. This was a serious issue during the Mumbai attacks because the armed forces, did not have the proper bulletproof jackets capable of giving them the required protection. With regard to chemical, biological, radiological, and nuclear emergencies, a nuclear-biological-chemical suit is essential. India has already developed such suits, and the armed forces are using them. Now they have developed a new version of the suit. Additional combat support includes life detectors for fighting forces. These detectors are sensitive acoustic detectors for people trapped under debris. If they can detect people, they can save lives.
Another incident that Sekhar touched upon was the Moscow hostage incident.19 Fentanyl was used to incapacitate the terrorists, however the problem with fentanyl is that it is an anesthetic and hence the concentration levels are very important. Government forces pumped fentanyl through the air-conditioning ducts and people who were close to the ducts had much higher levels of exposure which caused casualties. In India, they have worked on gases equivalent to fentanyl, synthetic analogues, like remifentanil and carfentanil, which are much less lethal. The median lethal dose, or LD50 levels are much lower and are as effective as fentanyl. India is exploring different methods of dispersing this anesthetic, including UAVs, which look like birds. They are being positioned at various places where people congregate in large numbers. These vehicles will operate a valve and fentanyl can be dispersed, but the concentrations are much lower, avoiding casualties.
Finally, biosensors are being developed based on nanotechnology to detect biological and chemical agents. India hopes to develop sensors that are small and cost a few dollars that can be put in a number of areas around sensitive points. If there is an attack, these sensors will detect the type of agent that has been used, at what time, and in what concentrations. This will enable proper protective measures and counter-measures.
Sekhar concluded by stating that if they can accomplish these tasks, they can protect sensitive installations from chemical and biological attacks. India can collaborate with other interested nations to enhance the combat readiness and effectiveness of forces engaged in countering terrorism.
Developing Comprehensive Training for Future Explosives Experts
Byron Gardner spoke about security at globally strategic facilities—nuclear, chemical, energy, and liquid fuel—and incidents that can have a significant impact on the national security of a country or on global security. The problem he said is that many of the security systems are not integrated and they do not work. In the United States, there have been some incredible mistakes at the Department of Defense, the Nuclear Regulatory Commission, at nuclear power reactors, and at the Department of Energy (DOE). Gardner has had the opportunity to work on some of the world’s most important facilities, and he finds these problems everywhere.
During his presentation, Gardner proposed a solution to these problems: to ensure that security professionals really understand the problem from a system engineer’s perspective, which they often do not. People managing security systems or major facilities tend to be in one of two categories. They are exceptional policeman, have had great experience and are placed in a high-security job, or they are exceptional people from the facility itself and yet they do not have a systems engineering background to run all of the complex systems that are under their jurisdiction.
To begin, Gardner provided a summary of the problem. Brand new, multimillion dollar security systems have failed during attacks or performance tests. These systems typically cost around $10,000 per meter to install and operate around sensitive facilities. That is expensive when perimeters are a couple of miles long. These systems are designed and installed in full compliance with government regulations, including double fences, cameras, lights, and alarms, but when they are tested often they do not work, and in some cases when terrorists attack, the alarm does not sound. Why not?
High-technology systems are rarely integrated, although these kinds of systems should be synergistic. The alarm systems should generate an assessment that in turn generates a response that is able to neutralize terrorists before they can destroy the target. If the alarm systems are not integrated, one piece might work, but the system itself does not work. A large part of the problem is that humans have to be involved in the system; however, there is pressure to take humans out of the loop. In Gardner’s opinion, that is a big mistake. Facility operators usually do not understand the capabilities and limitations of the security systems. That includes the government and system operators. There is an absence of design basis threat in many critical industries. United States critical energy facilities have security systems designed to prevent the theft of steel and copper, but not to prevent sabotage that may shut the power off to major regions
of the country. Further, systems are often installed that do not take into consideration the environment in which they are operating: it makes a difference whether the environment is hot, cold, snowy, or windy. Security vendors will sell anything, and they will say a piece of equipment works and charge a lot of money, although it might not perform as expected. Further, systems are often not installed and maintained properly.
Gardner’s assessment of the problem is that the people running these facilities lack hands-on experience, and it takes many years before people are fully productive. Security management professionals and government oversight inspectors often do not understand much of the new equipment; rather some feel that if they spend the money, everything is going to be fine. Further, the high-security expertise needed to run integrated systems, at least in the United States, does not reside in one institution, whether it is colleges or universities or national laboratories. Therefore, Gardner and his colleagues attempted to create a security program at Arizona State University and New Mexico State University, mainly based on good textbooks, but it failed because they did not incorporate the practical side of security requirements.
Traditional industrial security solutions do not work for major energy facilities and their support elements, whether it is power generation, desalination, or liquid fuel transport. The threat is great: terrorists may cause horrible, catastrophic consequences to economies or kill many people. The other issue that security management often does not understand is that cyberdistributive control systems and supervisory control and data acquisition (SCADA) threats are truly out there. Many security managers do not understand anything about cybersecurity of SCADA networks, therefore it is just pushed off to another organization and not integrated and as a result the systems do not talk to each other. Too often the perception is that kids are hacking in a basement, but there are vulnerabilities of a combined cyber and kinetic attack or physical attack.
Gardner then provided some examples of systems that cost on the order of $50 million each. He showed a perimeter along a globally strategic facility. The assessment zone is about 400 meters long and to adequately determine if someone is penetrating a facility, ideally it should be about 100 meters long. Therefore, if someone crosses over that facility, guards too often ignore the alarm. In an exercise, Gardner recounted that six people went across a perimeter, and the command center could not tell what was happening. In another example, the system was installed improperly, resulting in the inability for security personnel to see the perimeter clearly and to see if anyone was hiding. Typically, if someone runs across an alarm zone, unless there is a pre-alarm recording, an intruder can actually hide within the zone before someone sees him or her. The alarm might go off and by the time the operator puts down his or her work and looks at the screen, the intruder may be off the screen; therefore, the operator does not see anything and declares the incident a false alarm. At another globally strategic facility, a brand new, $50 million security system was installed, but the operator could not see what was going on in the area under observation because the lights were not installed properly. Another problem that occurs all over the
world has to do with vehicle barriers. Sometimes a trench will be put around a facility but still have a high avenue of approach; therefore, even with a nice big trench, a regular Toyota can just jump across it. At another facility, Gardner recounted that a new, $50 million security system was installed with a vehicle barrier at the entrance but not at the exit.
In the United States, “Jersey barriers” (cement obstacles) are often placed around government buildings, diplomatic facilities, and nuclear and chemical facilities. If a vehicle hits the barrier driving fast, the vehicle will be destroyed, but if a heavy truck just pushes on the barrier, it will easily be moved out of the way and the truck can drive right into the facility. That is common. In yet another case, people just drove right through the trees. Gardner provided another example where the physical barriers are outside of the alarm area. There is no benefit to delaying a terrorist if there is no detection beforehand. In this case, the vehicle barriers and the concertina wire are outside and the sensors are inside and if they go off, the terrorist is already running up to destroy the target. Still another common problem is that the security forces are deployed at the main gates, and their security commanders often say, it is impossible for terrorists to go over our fences because we have concertina wire. However, a person with good clothing hopped over a fence with concertina wire in about 11 seconds. This often causes the site commander to acknowledge that they need to change the response plan. Yet another $50 million security system had a good drainage ditch coming out of the facility and a terrorist could enter through the drainage ditch, go right underneath the security system, and pop up inside the facility.
Another common problem involves vibration cables on fences, but if they are installed in a windy area the alarms sound all the time. At another $50 or $60 million facility, with alarm-system vibration and fiber-optic vibrations, a sturdy ladder for a terrorist to climb up was placed against the wall so that the cable would not vibrate; a terrorist could crawl right over without an alarm sounding. In another case, a brand new, $50 million system had 3,000 alarms in 3 days. Do you think the guards will assess an alarm and respond to it in these conditions? They just turn the alarm off. One can throw all of those millions of dollars away. With another example, Gardner explained that there is a maritime nexus to many facilities, such as ports or critical energy facilities. These systems need to be tested as well. Gardner showed an example of a performance test team swimming up to a sensor, swimming around it, and banging on it with a wrench, and it did not go off. This was a brand new, multimillion dollar system. What was wrong? Someone did not know how to performance test the system.
What is the solution? For Tier-I, priority A, globally strategic, very important facilities, we have to do something to kick-start an awareness of how these systems should run so that governments and private industries do not waste money and create vulnerabilities that could affect our societies. Gardner said that it baffles him as to why 50-kilovolt transformers and the radiators that support them are not hardened. Not many transformers are made in the world, but the new ones that are made should have ballistic hardening that is oriented to
where they could be vulnerable in order to render standoff attacks from the outside ineffective.
Effective systems analysis involves response forces, deployment, and equipment. There is a tremendous propensity in the security industry to ignore the guards. Many believe that if they buy fancy equipment, the guards will respond. However, guards have to be integrated as well. Security systems inspections and evaluation are also critical. The Y-12 National Security Complex incident was, in Gardner’s view, a result of a weakened inspection program that did not have anyone trying to maintain awareness of what was happening.20
Gardner and his colleagues are trying to create a multidisciplinary, rotational graduate internship program to target early-career government and industry personnel. The program will cover issues that professionals need to know to question what is occurring in their facilities. Topics will include sensor and equipment security evaluation and testing, ballistic protection and critical infrastructure components. There is an agreement among several entities to create the multi-disciplinary program Gardner and his colleagues envision. The first one is Lawrence Livermore National Laboratory (LLNL). Experts at LLNL would discuss systems effectiveness, meaning vulnerability analysis; emergency preparedness and response, response force training and industrial security operations. Sandia National Laboratory (SNL) is going to be a part of the program and teach how to evaluate sensors, how to test them, how to look at the environmental conditions, and then they are going to have part of the National SCADA Test Bed Program that will allow people to actually gain some hands-on experience with cybersecurity. Texas A&M Transportation Institute will do live, full-scale vehicle crash tests with this group of students, and every rotation will do their own crash test. Then they will also go through emergency preparedness training at Texas A&M. Each rotation of students will be able to put blast waves across an infrastructure of interest to their industry or their country and also perform ballistic penetration tests on infrastructure. A theme running through this course is the idea that students will be able to understand the physics requirements, and be able to evaluate how to procure, maintain, and operate a security system that will provide the necessary protection. A utility company in California, Pacific Gas and Electric, will also participate in the program and provide training on industrial security planning and emergency preparedness. Pacific Northwest National Laboratory (PNNL) will provide training on industrial safety and cybersecurity, and then they will provide an instructional block on maritime security. Carnegie Mellon’s Computer Emergency Response Team, will provide an instructional block on cyberdefense.
In a 1-year rotation, students will go to LLNL for an introduction with all the national labs and then they will split into groups and rotate to SNL, New
20See, for example, U.S. Department of Energy, Office of Inspector General, Office of Audits and Inspections. Inquiry into the Security Breach at the National Nuclear Security Administration’s Y-12 National Security Complex. Special Report, August 2012. Available at: http://energy.gov/sites/prod/files/IG-0868_0.pdf; accessed September 25, 2014.
Mexico Tech, Texas A&M, and PNNL. They will go on field trips to the Strategic Petroleum Reserve and the DOE National Training Center, and they will have a block of instruction on response-force training for Tier-I facilities. They will also visit U.S. Coast Guard facilities and attend the American Society of Industrial Security conference. They will write a thesis at the end of the program and will be evaluated by their own country or by the United States.
Gardner and his colleagues want participants to receive certification and accreditation for the course, and both New Mexico Tech and Texas A&M are very interested in providing certification for the entire program. The program will be implemented on a pilot basis initially, with 20 participants from the United States and other countries. The first pilot program is just about to be funded. Gardner ended by saying that hopefully the program will start having an impact. It is just a start, but these facilities are too important, and the impact on the global economy, and maybe hundreds of thousands of lives, is just too important to delay.
S. Gopal opened the discussion by asking Gardner to clarify if the cost of $50 million for security systems included the cost of the equipment. Gardner said that the cost included equipment purchase and installation. Gopal said that normally one conducts performance assessment trials before such very expensive equipment is purchased. Gardner replied that Gopal’s question gets to the crux of the problem. The questions are: How does it happen that after the equipment is purchased and installed, there are performance problems? Why does the system not work? Do the checkout procedures match what the threat should be? How should the systems be performance tested? This requires not just a functionality check, but an actual performance test. Does the alarm system allow for an assessment in sufficient time to generate a response that can actually interrupt the adversary to keep sabotage from occurring?
An Indian participant recalled his experience at an ammonium plant three decades ago. Although they had very good tools, the operators received a flood of alarms because there were so many signals triggering the alarms to go off, and the net result was that guards ignored the alarms. Now Windows is starting to supply alarm management software; however, technology that one does not understand is by itself a threat.
Gardner was then asked about the cost of the graduate program. He replied that it is not as expensive as one might think. It is as expensive as a regular graduate program, but the universities will charge the tuition rate for a 9-hour class for each rotation. The universities are much more cost effective than the labs: A full-scale vehicle crash test for Texas A&M is about $40,000, including the truck and all the instrumentation that goes along with it. The explosive training at New Mexico Tech is nearly free, and the university is providing a range to do it. Finally, salaries are another expense. Salaries at the national labs are high, and the program has a couple of experienced people as mentors. Further, one
may believe that this type of program is sensitive from a security perspective, but as long as we are using infrastructure that is generic worldwide or applicable to the country or industry of interest, the U.S. Department of State has approved the program.
A workshop participant asked about cooperation with industry through the program and the agenda of industry in participating. Gardner replied that the goal is to inform the management of facilities and those who provide government oversight that security and therefore the training has to address the problems and perform to address these problems. The program is a great first step for cooperation with industry. This could really be a good way to have some cooperation across these Tier-I industries and across countries. For example, the United States has trouble with its southern border with Mexico. High-tech sensors were installed and a great deal of money has been spent, but the problem still is not solved, because people still have not grasped the problem fully.
John Holmes added that, having accepted many security systems that were not functional, in many cases the problem is that testing has been done on pieces of the system and not the whole system. Often part of the system is finished and tested, and then the next part of the system is finished and tested. Sometimes the full system is never functionally tested. Also the people accepting these systems are not necessarily professionals in that area of security, and performance testing is a very complex issue. Therefore, if the person who makes a test plan is not familiar with performance testing, sometimes the system is less than functional because the person does not know how to test the entire system properly. Holmes worked with the U.S. National Academy of Sciences on several projects, and he came to understand this challenge much better. This is a very specific area of study.
Gardner agreed and added that this is why the program he described includes 2½ months of performance testing. Although the facility manager is not personally going to be doing performance testing, he or she needs to be available to evaluate the test plan, observe it, and ensure that the security system is working.