Preventive engineering involves tackling a problem at its root cause or causes. In contrast to alleviating detrimental consequences which have occurred, preventive engineering techniques aim to anticipate detrimental consequences and avoid their occurrence. In reference to Figure 2-1, the focus for society’s activities with respect to natural disasters would be on mitigation and prevention activities. Preventive engineering tends to be the best approach to engineering problems, since a problem tackled at its root often yields the most effective, sustainable solution.
As discussed in section 4.4, especially section 4.4.3, there are many difficulties in correctly anticipating all potential natural disaster situations, and so response/recovery and preparation/planning activities are still necessary components of society’s activities. Proper mitigation/prevention techniques, though, can be powerful in reducing the need for, and resource use of, these other activities, hence it is important to focus a significant portion of efforts on mitigation/prevention.
From Chapter 2 and Figure 2-1, the root cause of natural disasters is based in risk, the combination of natural hazards and vulnerability. Therefore, there are two potential areas to implement preventive engineering: preventing natural hazards (section 5.2) and preventing vulnerability (section 5.3).
5.2 Preventing Natural Hazards
Society’s activities continually influence natural hazards, both intentionally and unintentionally. Sometimes natural hazards are caused or augmented and at other times natural hazards are curtailed or prevented.
5.2.1 Preventing Astronomical Hazards
Preventing astronomical hazards involves astronomical challenges, and there have not yet been formal attempts. Manipulating stellar properties to prevent flares or geomagnetic storms would be an immense engineering undertaking and excessive care would have to be taken to ensure that solar radiative properties were not affected too much. There is a strong movement to monitor space for celestial objects on a collision course with the Earth, with at least four long-term, continuous programs in operation (see Part II in Gehrels, 1994), but such actions do not prevent the hazard. The idea which most closely resembles collision prevention is using technologies such as nuclear weapons and lasers to destroy or alter the course of incoming objects, but such ideas are still in the conceptual stage (see Part VII in Gehrels, 1994).
5.2.2 Preventing Biological Hazards
Habitat destruction and species extinction is one method which society has used, deliberately and unwittingly, to eliminate biological hazards. Sometimes the extinction is local (e.g., wiping out malaria from the southern U.S.A. by exterminating the Anopheles mosquito vector) and sometimes it is global, as in the case of small pox. Engineers often play prominent roles, through contributions such as developing production processes and equipment for manufacturing, transporting, and applying biocides.
This approach to preventing biological hazards is inappropriate for managing vulnerability to biological disasters. The disadvantages of anthropogenic habitat destruction and species extinction are well-documented in practical and theoretical ecology (see, for example, Begon et al., 1990; Jeffries, 1997; Noss, 1995; Wilson, 1988) because areas such as food production, medicine, recreation, and materials development are impacted detrimentally (see also Bell, 1993; Dobson and Carper, 1993). The biological hazard itself is often advantageous to society, such as venom from a saw-scaled viper used to produce drugs which combat heart attacks (Eliot, 1998a), the poisonous nightshade (Atropa belladonna) used to detoxify PCBs (Eliot, 1998b), and the large amount of meat and hide which can be obtained from large and often dangerous game animals.
In the microbial realm, many preventive techniques are developed with significant engineering input, but medical scientists, politicians, doctors, social scientists, and geographers, among others, play prominent roles too. Standard preventive techniques are developing vaccines (which often are developed using genetic engineering) and implementing vaccination programs; preparing other public health measures such as modern health facilities, quarantine plans, and educational programs; considering disease prevention as one factor in appropriate urban and rural planning; and supporting socioeconomic policies which aim for poverty reduction.
Encompassed in these techniques are the engineering areas of providing a clean water supply and proper waste treatment and disposal. Examples are:
•Following separate cholera epidemics in the 19th century, London and Paris constructed their first sewage systems in order to prevent further outbreaks of this biological hazard (Viessman and Hammer, 1993).
•Deficiencies in Peru’s drinking water treatment system (exacerbated by bickering between Peruvian and American governmental agencies) permitted the Vibrio cholerae bacterium, which causes cholera, to establish itself in Peru in 1991, whereupon it spread throughout South and Central America (Anderson, 1991).
•Stagnant water collected by discarded tires and coconut husks are used for egg-laying by some Aedes species, the mosquito vector of the dengue fever virus, and so appropriate waste management of tires and coconut husks would be a factor in preventing incidences of dengue fever (Innes, 1995).
5.2.3 Preventing Geological Hazards
Methods of using technology to control seismic events were often discovered accidentally. An underground nuclear bomb test in Nevada in 1968 induced an earthquake of Richter magnitude 6.3 and aftershocks up to Richter magnitude 5.0, which led to suggestions of inducing more frequent, lower-magnitude earthquakes with underground explosions in order to prevent a single high-magnitude shock (Waltham, 1978). The influence of underground nuclear explosions on seismicity is still a subject of debate with some authors claiming that the correlation can be skewed by tides or atmospheric pressure (see Console and Nikolaev, 1995).
Filling reservoirs for inducing low-magnitude earthquakes, or draining a reservoir to prevent a large earthquake, represents another earthquake prevention technique. In Arizona in 1935, the artificial reservoir Lake Mead on the Colorado River was accidentally discovered to induce earthquakes through the weight of the water on fault lines (Smith, 1996; Waltham, 1978). Reservoirs also increase the groundwater pressure which would reduce friction along a fracture, causing low-magnitude shocks. Such “lubrication” of a fault can be achieved by manipulating the groundwater levels in any fashion, including the injection of water into deep boreholes. In the 1960’s, using injection to induce earthquakes was discovered inadvertently near Denver and afterwards, the United States Geological Survey (USGS) conducted reasonably successful experiments in Colorado (Bolt, 1993).
The amount of damage reduction which can be achieved through altering earthquake patterns is not clear. Structures can fail due to fatigue from numerous, small shocks as well as failing due to one large shock, and it is disputable as to which scenario would actually cause more damage or how vulnerability would be influenced by each scenario. This concern illustrates the separation between an initial natural hazard event and the design load experienced from the natural hazard event, as discussed in section 4.4.2. The fear of inadvertently causing a major earthquake is another powerful restraint on implementing techniques of earthquake prevention. The Koyna Reservoir near Bombay, India was filled in 1967 and caused an earthquake which killed 177 people (Waltham, 1978).
Volcanically-induced seismic events could conceivably be controlled by providing an artificial outlet for the gases or magma which cause earthquakes. Excavating surface material or drilling holes are possible techniques for creating the outlet. The amount of material needed to be moved and the depth of the holes required would often be far beyond that which could be achieved quickly and cheaply, and so success might be limited. There is also the danger of inducing an unwanted volcanic explosion since complete control of the volcanic processes could never be attained.
Although engineers currently do not have techniques for stopping or controlling volcanic eruptions, affiliated hazards can often be controlled. Lahars (mudflows) and water from jökulhlaups (flash floods from a glacier, sometimes caused by a volcanic eruption melting ice) can be directed by hydrological engineering techniques, including dams, levees, and floodways. Lava flows can also be diverted with analogous barriers and diversions. Barriers can be built, but can also be created by blowing up solidified lava walls thereby blocking the lava with debris. As well, an advancing lava flow can be doused with water, cooling it and forcing the lava to spread sideways. Even if such techniques do not divert the lava flow, they usually slow it down, providing more time for evacuation and for natural cooling of the lava. Barriers were both built and created with explosives during a 1983 eruption of Mount Etna in Sicily, successfully protecting tourist areas from damage (Coch, 1995). On the island of Heimaey in Iceland in 1973, seawater was pumped onto a lava flow, possibly preventing the blockage of the town of Vestmannaeyjar’s harbour (National Geographic, 1997).
Barriers and landscape engineering could be used to prevent landslides, rockslides, and poison gas emissions. The effectiveness of these solutions cannot be entirely assured, particularly in the long-term, because, just as with seismic energy and rising magma, the material from landslides and rockslides along with poison gas must be released at some point. Barriers will rarely block these hazards completely since gas pressure would build up and rocks and earth would pile up behind the barrier. Barriers and landscape engineering, as with lava, would serve to redirect or control the release of these hazards.
5.2.4 Preventing Hydrometeorlogical Hazards
Engineers have been at the centre of attempts at preventing floods through projects such as dams, levees, floodways, and reconstructing the courses of waterways. Sometimes these efforts have been successful, such as the floodway built in 1968 to divert Red River waters around Winnipeg, Manitoba. This floodway was used 18 times in the thirty years following its construction, most notably during the 1997 Red River flood in which Winnipeg sustained minimal damage while other communities along the Red River were inundated causing CAN$ several hundred million (1997 dollars) of damage (IJC, 1997). At other times, poor planning and poor use of technology augmented a flood disaster, as occurred along the Mississippi River, U.S.A. in 1993 (Changnon, 1996) and along the Saguenay River, Québec in 1996 (Environment Canada et al., 1996).
Cotton and Pielke (1995), Kahan et al. (1995), and Kriege (1984) describe cloud seeding experiments around the world for suppressing hail, enhancing precipitation (to prevent water shortages and droughts), and affecting the path and severity of hurricanes. No experiments have yielded a success which has been generally accepted, and debates continue with respect to the effectiveness of cloud seeding. Intensive efforts at cloud seeding have dwindled recently, partly due to the contemporary environmental ethic of minimizing potential environmental impacts during experimentation, and partly due to the fear of litigation if weather is apparently modified and causes damage. One of the first such cases occurred in 1947 when General Electric Corporation seeded a hurricane which subsequently changed direction and made landfall on Georgia, prompting lawsuits for hurricane damage. Programs have also been impeded by a lack of understanding of the physics behind cloud seeding. Scientists and funding agencies are reluctant to support large-scale experiments without being able to compare theoretical results with experimental results.
Anthropogenic climate change could influence hydrometeorological hazards, potentially preventing some of the hazards. Contemporary experience with rapid climate change (which might be either anthropogenic or natural), however, illustrates the disadvantages of likely scenarios. Most literature on climate change and natural hazards (see for example Burton et al., 1993; Environment Canada et al., 1996; Etkin and Brun, 1997; Vellinga and Tol, 1993) warns of large uncertainties, but indicates the strong likelihood of increases in both the frequency and severity of extreme events due to global warming, particularly with respect to storms, floods, and heat-related disasters, but not necessarily with respect to cold-related disasters. The biological hazards of vector-borne diseases--including malaria, schistosomiasis, onchocerciasis, dengue fever, and yellow fever--are expected to affect larger spatial ranges due to global warming (Environment Canada et al., 1996). These examples mainly relate to the augmentation of natural hazards, thus current trends in climate change will be unlikely to assist in natural hazard prevention.
5.2.5 Challenges in Preventing Natural Hazards
Sections 5.2.1 through 5.2.4 have identified significant problems in implementing natural hazard prevention. In any case, preventing natural hazards is not a particularly appropriate approach to managing environmental processes, because many natural hazards processes are extremely important to the environment, and thus to society. For example, section 2.6 discussed how tectonic activities are forces of both construction and destruction; preventing tectonic activity to avoid the destructive aspects will also prevent the constructive aspects. As well, volcanoes contribute significantly to biogeochemical cycles of metals, including aluminum, arsenic, cadmium, copper, iron, lead, mercury, and zinc; assist the sulphur, chlorine, and nitrogen cycles through gas emissions; and contributed enormously to the atmosphere’s evolution (Butcher et al., 1992; Schlesinger, 1991). Decker and Decker (1998) estimate that about “25 percent [by mass] of the water, chlorine, and nitrogen in the Earth’s atmosphere was reworked by subduction and volcanic eruption and about 75 percent [by mass] of the carbon was recycled” (p. 204). Tectonic subduction and spreading processes also cycle minerals and elements by eliminating old rock and creating new rock.
Hydrometeorological hazards involve the transfer of mass, heat, and linear momentum through the atmosphere and hydrosphere, and so are responsible for air flows, ocean currents, weather, and climate. The water cycle, incorporating precipitation, heavily influences most biogeochemical cycles (Butcher et al., 1992; Schlesinger, 1991). Hydrometeorological hazards also contribute to non-chemical ecological and geological processes. Floods often deposit fertile soil layers or scour out stagnant waterways while winds knock down dead trees permitting the wood to be decomposed (recycled) and leaving spaces for new growth. Glacial movement has been responsible for the good farming till, along with much of the topography, in places such as southern Ontario and Orkney, Scotland.
Some advantages of biological hazards were described in section 5.2.2. Astronomical hazards also have advantages in bringing materials to Earth, such as the Sudbury Igneous Complex near Sudbury, Ontario which is the world’s largest and richest known nickel ore deposit and which is the remnants of a meteor strike approximately four billion years ago (Erickson, 1994). Natural hazard processes are based on balances amongst, interaction between, and cycles throughout the atmosphere (air), hydrosphere (water), lithosphere (Earth’s surface), and biosphere (life). They are unavoidable--and essential--components of the Earth’s characteristics which make the planet habitable through the provision and recycling of environmental resources. Thus, even if it were possible, significantly preventing natural hazards would severely impact the resources available to society, particularly over the long-term.
Another major concern of preventing natural hazards is effectiveness: the consequences could be a natural disaster far more deleterious than any natural disaster in the absence of such interference. Examples have already been provided with respect to the prevention of earthquakes, rain, hailstorms, and floods, and with respect to anthropogenic climate manipulation. Tradeoffs will always be present, yet they are difficult to model, analyze, and predict.
Since natural hazards are difficult to prevent, and successful prevention can lead to new dangers, preventing natural hazards is not currently appropriate for managing vulnerability to natural disasters. Society at present should therefore not rely on prevention of natural hazards. Understanding the origins, behaviour, and consequences of natural hazards is the most important “root of the problem”. This knowledge can be used for properly applying the tool of technology for managing vulnerability to natural disasters.
5.3 Preventing Vulnerability
Since managing vulnerability generally implies preventing excessive damage from natural disasters due to vulnerability, the prevention or reduction of vulnerability tends to be paramount. Vulnerability is a characteristic of society (section 2.7), so society should be able to use its activities to manage vulnerability. Technology can be used effectively for managing vulnerability, which also includes knowing when not to use technology (or using minimal technology).
5.3.1 Technology Used for Managing Vulnerability
Engineers develop technology for use during each stage of society’s activities, as illustrated in Table 5-1. The development and implementation of many of the technologies are often interdisciplinary and collaborative efforts, hence not all the examples involve practising engineers7. Because this thesis emphasizes the role of technology, and is not a treatise that compiles all technologies, only illustrative examples are provided in Table 5-1. More examples, both specific and general, are used throughout this thesis.
5.3.2 Pre-Disaster and Post-Disaster Engineering
Three points arise from the examples in Table 5-1.
First, the pre-disaster technologies and systems tend to be more event-specific than the post-disaster technologies and systems. Designing buildings for tornadoes is different than designing buildings for earthquakes--and there are even wide variations in designs used for different types of earthquakes (Bolt, 1993). The engineer has the design purpose clearly defined and implements engineering techniques for this purpose, as per the three-stage framework discussed in section 4.3 (Figure 4-1). Conversely, post-disaster technologies and systems are more similar, irrespective of the cause. Rescue operations tend to invoke similar principles for people trapped under collapsed buildings irrespective of the cause of the collapse. Medical technologies and systems respond to the observed injuries rather than the cause(s) of these injuries.
Second, the engineer’s role is much more dominant during the pre-disaster phase than during the post-disaster phase. The engineer works with others to determine how much mitigation/prevention and preparation/planning is desired (the load/response objectives) and how to best achieve the desired level
Table 5-1: Examples of How Technology is Used for Natural Disasters
Monitoring, Predicting, and
Models for and simulations of the behaviour of natural hazards.
Finding a chaotic strange attractor in traits of volcanic tremors (Scholz, 1989).
Rapid communication networks.
Cellular phones and ham radios.
With preparation and planning)
Models, systems and simulations for disaster prediction and response.
Predicting which wind speeds will damage or destroy which buildings (Liu, 1993).
Technologies for monitoring and tracking natural hazards.
Doppler radar for thunderstorms and tornadoes; satellites for hurricanes and volcanic ash clouds.
The Built Environment
Hydrological and landscape engineering.
Barriers and channels for landslides, avalanches, floods, lava, and lahars.
Materials and construction engineering.
Building codes (e.g., NRCC, 1995).
Land-use planning, land zoning, and community design.
Creating and enforcing no-settlement flood zones; engineering lifelines to maintain their integrity during and following an earthquake.
Information and education about
Dispelling false beliefs about natural disasters.
Teaching that rubber footwear will not protect against lightning strikes.
technology and natural disasters.
Generating knowledge and awareness about natural disasters.
CD-ROMS and WWW sites (see section 2.4) along with information pamphlets.
Education how to appropriately develop and use technology.
Equipment and devices for rescue operations.
The “Jaws of Life” for rescuing people trapped in crushed vehicles.
Medical technologies and systems.
Wheelchairs, prosthetic limbs and defibrillators.
Models, systems, and simulations to analyze disaster events.
Mapping earthquake damage to lifelines onto fractals (Nakagawa and Satake, 1991).
(the system design). The engineer is involved in assessing the needs, analyzing potential solutions, and implementing the solutions. Following a disaster event, non-engineers have the dominant roles and the technologies and systems are used as tools when and how the non-engineers choose to use them.
Third, managing vulnerability is achieved more easily during the pre-disaster phase. The post-disaster technology is important to society, but most of it patches up damage which occurred because of vulnerability. For example:
•rescue and medical technologies attempt to prevent casualties from ending up in a worse condition whereas building codes attempt to prevent casualties occurring;
•models and analyses of disaster events occur following a disaster whereas models and analyses of disaster preparations attempt to eschew a disaster event; and
•information and educational material, which is often disseminated with technology, aims to reduce the number of people who understand natural disasters through personal experience.
Pre-disaster engineering, which embodies the preventive principle, achieves more effective management of vulnerability than post-disaster engineering.
5.3.3 Challenges in Using Technology to Prevent Vulnerability
The predominant challenge which arises in using technology to prevent vulnerability to natural disasters is that society’s actions are not always focussed on managing such vulnerability, and society’s activities for other purposes could unintentionally increase vulnerability. For example, many people build large houses near cliffs along California’s Pacific Ocean coastline, because through either ignorance or choice, they have rated a prestigious location and a beautiful view as being more important than preventing vulnerability to landslides induced by earthquakes or storms. Another example is that mobile homes and recreational vehicles are highly vulnerable to destruction by tornadoes, but are attractive--or necessary--for economic reasons. More than half of all tornado deaths in Canada and the U.S.A. occur in these dwellings (“Tornado!”, 1995). Chapter 3 described various non-technological influences on vulnerability.
Even when properly managing vulnerability is the intention of society’s activities, mistakes--such as poor analysis, poor judgement, or ignorance--may make this goal elusive. The eruption of the Nevado del Ruiz volcano in Colombia (summarized from Mileti et al., 1991) illustrates this problem in demonstrating how one broken link in the cycle of society’s activities can create a devastating natural disaster. At Nevado del Ruiz, scientists and civil defence authorities had a reasonable evacuation plan, knew what volcanic hazards were present and the vulnerable locations, were monitoring the volcano, and correctly identified the threat following an eruption on the evening of November 13, 1985. Communication and attitude difficulties at the local level caused the failure of evacuation warnings to be disseminated to the populace, and between 22,000 and 24,000 people died in the town of Armero when it was obliterated by lahars (mudflows of volcanic origin). People who received and acted upon warnings, sometimes minutes before the disaster, tended to survive. The mayor of Armero and Armero Red Cross officials were reportedly swept away as they were discussing the situation with others by radio. Approximately 1,800 more deaths occurred throughout the region, but survivors from Armero numbered around 5,000.
Another relevant case study occurred in California in 1982 when volcanologists predicted that a volcano in the tourist resort of Mammoth Lakes, California would likely erupt. Tourists stayed away and property values fell, but the volcano failed to erupt and the scientists received death threats from angry residents (WGBH, 1992). Because of this incident, any similar warnings in the future would likely be met with heavy scepticism, to the detriment of managing vulnerability by preventing volcanic disasters in the community.
Another concern is that appropriate vulnerability management might not be achievable for all natural hazards in a given location. As mentioned in section 4.4.1, building codes for Kobe, Japan deliberately reduced vulnerability to typhoons but inadvertently increased vulnerability to earthquakes. Where vulnerability tradeoffs occur, innovative and flexible solutions will be needed to determine whether or not vulnerability prevention can be well-managed despite the apparent need for vulnerability tradeoffs, and if not, to decide how to manipulate the tradeoffs. This problem is similar to the issue raised in section 4.4.3, that it is not possible to anticipate every potential scenario, particularly for multiple-event scenarios.
The solution is also similar: as many factors as possible should be taken into account to achieve the best overall solution and to achieve as much vulnerability prevention as possible. Alternatively, there are many occasions when technology for preventing vulnerability coincides with technology used for other goals, such as sustainable resource management. For example, the vulnerability of energy lifelines to natural disasters can be reduced by implementing sustainable energy techniques, such as demand reduction and developing small-scale, renewable sources, because:
•less demand for energy implies fewer problems when energy lifelines are cut;
•decentralization implies fewer people are affected if an energy system, such as a switching station or power plant, is ruined;
•an emphasis on renewable sources implies fewer links in the source-to-consumer chain which could be impacted by a natural disaster (such as transportation networks for bringing coal to a fossil fuel plant); and
•small-scale energy systems permit locals to operate and maintain them so that if a natural disaster causes damage, there will be less need for outside help and supplies to reestablish the energy lifeline.
Engineers can alleviate many of the problems in preventing vulnerability to natural disasters by developing and implementing integrated solutions which impact beneficially on other goals of society.
5.4 Does Managing Risk Prevent Natural Disasters?
Section 5.2 illustrated the challenges and, at times the inappropriateness, of preventing natural hazards. Section 5.3 demonstrated the possibilities for preventing vulnerability. For preventing damage from natural disasters, managing vulnerability through prevention would seem to be more effective than managing natural hazards. Although risk, the combination of vulnerability and natural hazards, feeds into natural disasters (Figure 2-1), society’s activities should be concentrated on preventing vulnerability rather than on preventing risk.
Risk prevention can actually lead to increased vulnerability (Green et al., 1993) if the hazard is prevented but vulnerability is augmented. Wilde (1994) describes how engineering and other safety measures give people such a sense of security that, at times, they will undertake behaviour which increases their vulnerability so much that their risk level is higher than before the safety measure was implemented. Because people do not always estimate their risk correctly, behaviour which appears acceptable with the safety measure in place might actually leave people at a higher risk than they realize. Wilde (1994) presents detailed evidence from the transportation sector.
A natural disaster analogy would be if hydrological engineering reduces the impact of annual spring flooding along a river. Inhabitants downstream will become inured to the absence of floods. Because there are few floods, they will tend have decreased interest in flood-proofing activities, decreased awareness of the potential flood hazard, decreased understanding of how to predict and react to floods, and decreased ability to psychologically cope with any flood events. Monitoring and warning systems may also lapse. The hazard is minimal, but vulnerability is immense. This phenomenon was discussed in section 3.3 with respect to psychological impacts on vulnerability: Torontonians are more familiar with blizzards and thus are less vulnerable to their impact than Dubliners. When a snowstorm hits Dublin, or a flood hits these unprepared communities, the high vulnerability can result in devastating consequences.
The difference between the snowstorm example and the flood example is the human influence. The absence of snowstorms in Dublin is naturally occurring; the absence of floods is anthropogenic and anthropogenic design must always select somewhat arbitrary design criteria, as discussed in section 4.4. When an extreme event occurs which exceeds those design criteria--as must happen eventually--the high level of vulnerability and the psychological unpreparedness create a “natural” disaster with worse consequences than could have occurred without the hydrological engineering. The natural disaster causes more damage because technology was used to prevent risk rather than to prevent vulnerability. As well, it is conceivable that the physical properties of the event are more extreme than would have occurred in the absence of the hydrological engineering. As Green et al. (1993) write, “In some cases, the effect of reducing the risk of some events is to increase the challenge which will be presented by more extreme events”.
Using technology to prevent vulnerability of society to natural disasters helps to avoid such pitfalls. Cases where natural hazard prevention leads to risk reduction but increased vulnerability would be identified and avoided, because vulnerability would be the key characteristic in natural disaster risk management. Green et al. (1993) also create a scenario where, following intervention by society, damage from floods actually increases, but vulnerability decreases and economic output from the land increases. Although this scenario is hypothetical, it again supports the importance of preventing vulnerability and of integrating vulnerability prevention with other goals of society, such as increasing economic wealth and sustainable resource management (which are not necessarily opposite goals).
The achievement of sustainable practices in society is not exclusive of preventive engineering. Using technology to prevent vulnerability is useful for sustainability, even though preventing risk, preventing natural hazards, and preventing damage are less suitable (and section 2.6 also pointed out that damage is not necessarily detrimental). Preventive techniques, in vulnerability management and other areas, form a large component of achieving sustainability, because the preventive approach examines the root of problem thereby enabling a clear understanding of how to define the problem--the key challenge identified in Chapter 4.
Managing risk can prevent natural disasters to some extent, but a more suitable objective is preventing vulnerability. There are elements of managing natural hazards and managing risk, but care is required to ensure that the focus is on vulnerability. Preventing vulnerability prevents natural disasters and prevents subsequent damage--when it is appropriate to do so.
Preventive engineering should focus on managing vulnerability, rather than on managing natural hazards or managing risk. Preventing vulnerability implies implementing plans over a scope which is narrow enough to be dealt with realistically and effectively, but which is broad enough to encompass all necessary areas. Chapter 6 discusses boundaries and scales and suggests those which should be used for using technology to manage vulnerability to natural disasters.