HAZARDS

Hazards are low-probability, high-magnitude phenomena that have the potential to cause large negative impacts on people. While this definition is unavoidably imprecise (what counts as a "phenomenon"? what probabilities qualify as "low"? and what impacts qualify as "large" or even "negative"?), in general hazards can be understood as acting outside of daily human expectations to adversely affect the quality of life of those exposed to them. Hazards refer to a prospect or risk of an occurrence; a particular occurrence of a hazard is more typically termed a "disaster" or sometimes an "extreme event"; when they are technological in origin they may be termed "accidents."

Some types of phenomena—such as hurricanes, earthquakes, landslides, and reactor meltdowns—are unambiguously classified as hazards, whereas others, especially those that are less temporally or spatially discrete, such as droughts, famines, and epidemics, may or may not be included under the term, depending on who does the classifying. Wars and other types of human conflict are generally not categorized as hazards.

A related use of the word "hazard" refers to existing conditions of the environment that may pose a risk to humans, such as a toxic waste site or even the edge of a cliff. Similarly, hazardous materials are those that may create a risk to human or environmental health if exposure to them is not regulated and controlled. This entry, however, focuses on hazards as dynamic phenomena, not as static conditions or material properties.

In the ten-year period 1992 to 2001, hazardous events, or disasters, worldwide were responsible for more than 620,000 deaths. Drought caused almost 45 percent of these deaths; floods, earthquakes, and windstorms caused most of the remainder. An additional 2 billion people required immediate assistance (60% as a result of floods), and the direct costs due to the destruction of infrastructure, crops, homes, and so on was more than $600 billion (with earthquakes, floods, and windstorms making up about 90% of this total). To put these numbers in some perspective, every year hazards seriously disrupt the lives of as many people as the entire population of Brazil or Indonesia, and cost about as much as the entire economic output of Pakistan or Peru (World Health Organization, United Nations Development Programme 2002).

Hazards Are Not Natural

Hazards are commonly divided into two types: natural and technological. Technological hazards are those arising from the failure of technological devices or systems to behave as intended. Natural hazards arise from nonhuman forces and can be subdivided into geophysical hazards, such as volcanoes, earthquakes, and tsunamis, and hydrometeorological hazards, such as hurricanes, floods, and tornadoes. Natural and technological hazards, however, are often related to each other, in that natural disasters may trigger technological failures, for example of power grids or dams. Moreover, natural hazards must be understood not simply as the result of natural phenomena, but as arising from the socioeconomic context within which such phenomena occur.

Human exposure to hazards results from humans living in areas where hazards are present; human vulnerability to hazards arises from the types of development exposed to hazards. The consequences of hazards are determined as much or more by the extent of exposure and level of vulnerability than by the characteristics of the hazard itself. Thus, for example, when a magnitude 6.9 earthquake struck a densely populated region in Armenia in December 1988, more than 25,000 people died and 1.6 million were directly affected. When, ten months later, a similar magnitude earthquake struck a highly populated region of California (the October 1989 Loma Prieta event near Santa Cruz), sixty-three people died and fewer than 10,000 were affected. This stark difference in impacts was largely a reflection of poor design and construction standards for buildings in Armenia compared to those in California. Moreover, despite Armenia's much lower level of economic development, its economic losses from the 1988 event, estimated at about $14 billion, were greater than the estimated $6-to-$10 billion price tag of Loma Prieta.

The inseparability of hazards from their social context is clearly illustrated by historical trends in disasters, which show a continual and rapid increase in the number of disasters, rising from a worldwide average of about 100 per year in the early 1960s to between 300 and 500 per year by the early 2000s. ("Disasters" here is defined by the World Health Organization's Collaborating Centre for Research on the Epidemiology of Disasters [CRED] as events that kill at least ten people, affect at least 100, result in a call for international assistance, or result in a declaration of emergency.) While some of this increase reflects changes in reporting, most of it arises from increased exposure and vulnerability throughout the world because of growing population, expanding economies, migrations to coasts and other vulnerable regions, increasing urbanization, and related factors. These changes are especially reflected in the costs of major disasters, which according to the German insurance company Munich Re rose more than tenfold in the second half of the twentieth century, from an average—in real (2002) U.S. dollars—of about $4 billion per year in the 1950s to more than $65 billion in the 1990s.

It is important to emphasize that these increases are best explained by changes in social context, not changes in the occurrence or type of hazardous events. For example, it has been well documented that rapidly increasing economic losses from hurricanes striking the U.S. eastern seaboard are caused by growing population and wealth, not by increased frequency or magnitude of storms. The great Miami hurricane of 1926 caused about $76 million in damage (in inflation-adjusted dollars); when Hurricane Andrew, of similar force, struck south Florida in 1992, it caused more than $30 billion in damage (Pielke and Landsea 1998).

Complexity of Hazards

Because hazards are socially embedded, their impacts arise from the complex interaction of many variables. In Armenia, steel that had originally been produced to reinforce buildings was diverted to weapons construction instead, thus revealing cold war geopolitics as one source of vulnerability to the 1988 earthquake (Mileti 1999).

Hurricane Mitch, which in October and November of 1998 killed more than 10,000 people and caused severe economic and social disruption in Central America, was responsible for triggering a mudslide in Nicaragua that killed about 2,000 people (Olson et al. 2001). The mudslide, however, wascreated notjustby the torrentialrains brought by Mitch, but also by land-use patterns that led to deforestation of a steep mountain slope, which collapsed when it became saturated with water. Eighteen months later, a debris flow in Manila, Philippines, triggered by normal monsoon rains, killed about 200 people. But in this case the disaster occurred on the flank of a huge landfill where thousands of people scavenged garbage for a living.

In Chicago, a heat wave in the summer of 1995 led to the death of more than 700 people. The temperatures in Chicago were no higher than those regularly experienced in many places; the huge number of casualties was instead caused by a combination of failed social services (for example, insufficient number of emergency vehicles and workers) and the large number of people, mostly poor and elderly, living alone, without resort to social networks (Klinenberg 2002).

Such examples also show that a preliminary event may trigger additional hazards that may themselves be damaging or that may combine with the principal hazard to multiply damages. For example, the Chicago heat wave led to technological failures in the form of power outages and water service interruptions that made it more difficult for people to cope. Major disasters may also trigger disease outbreaks, especially when water supplies are cut off or contaminated. The 1906 San Francisco earthquake is often called the San Francisco Fire because of the disastrous conflagrations it caused throughout the city. These sorts of complexities also underscore the futility of making a clear distinction between natural and technological hazards.

Uneven Distribution of Hazard Impacts

The impacts of hazards are disproportionately borne by poor people living in poor regions and countries; thus, hazards are a manifestation of socioeconomic inequality and an issue of social justice. While the poorest thirty-five countries account for only about 10 percent of the world population, they suffered more than half of the disaster-caused deaths between 1992 and 2001. Of those directly affected by disasters during that decade, almost 90 percent lived in Asia, where dense populations combine with high vulnerability and widespread poverty in nations such as India, China, and Indonesia. As the contrast between the Armenia and Loma Prieta earthquakes starkly shows, the benefits of affluence include a capacity to protect against the most direct and devastating effects of hazards, and a significant component of this capacity is the scientific and technological infrastructure that typically accompanies (and fuels) the growth of affluence.

Not surprisingly, affluent nations suffer the greatest absolute economic losses from hazards. The disproportionately large sizes of their economies create the potential for much greater economic damage from the impacts of hazards. For the decade 1992 to 2001, the forty-five richest countries (making up about 18 percent of global population and accounting for 82 percent of global wealth) experienced about 62 percent of total global economic damage from hazards. As a percentage of gross national product (GNP), however, the economic effects of hazards on poor countries are about 100 times greater than for rich countries. Damages from Hurricane Mitch, for example, were estimated at between $5 billion and $7 billion, which was about the same as the annual combined total GNP of the two most affected nations, Honduras and Nicaragua. The magnitude 6.7 Northridge, California, earthquake of 1994 was the most costly disaster in U.S. history, causing between $20 billion and $40 billion in losses; the total, however, was equivalent to only between about 2 and 4 percent of California's economic activity for that year.

Disparities between rich and poor will compound over time. Global population growth is mostly concentrated in poor countries and leads to rapid urbanization, usually in vulnerable coastal zones, as well as dense rural populations. Unregulated land use translates into widespread environmental degradation, especially deforestation, which in turn exacerbates flooding and related phenomena such as mudflows, debris flows, and landslides. Design and construction standards are typically low, and even when adequate building codes exist, corruption, lack of enforcement, and insufficient resources result in an unsafe built environment. Emergency response capabilities are often inadequate, and hazard insurance is usually unavailable, slowing the recovery process. Technological infrastructure, such as communication and transportation systems, is typically fragile, and capacity to repair damaged systems is limited. Such factors reinforce one another to magnify the vulnerability of poor people and nations to hazards, and they act as a brake on development.

Mitigation

In the affluent world, numerous approaches have been adopted to mitigate the effects of hazards, including building codes that are appropriate to known risks; landuse regulations for floodplains, coastal zones, and seismic zones; and dams, levees, and other engineering interventions for floodplain management. There is little question that such measures, combined with early warning systems for hurricanes, tornadoes, and floods, and coordinated emergency response plans, have limited the human and economic toll of hazards in the developed world. Nevertheless, while the number of people killed and injured has declined for some hazards, and stayed relatively stable for others, the economic costs of hazards appear to be rising at an exponential rate. Absent mitigation efforts, they would be rising more rapidly still.

Despite aggressive mitigation efforts, affluent nations are not exempt from major disasters. The magnitude 7.2 earthquake that struck Kobe, Japan, in January 1995 killed 6,000 people and led to an estimated $100 billion in damages, yet Japan is justifiably considered to have the world's most sophisticated and effective earthquake hazard mitigation practices. In the U.S. Midwest, decades of flood control engineering preceded the 1993 Mississippi River basin floods that caused $18 billion in damages and that arguably constituted, in the aggregate, the worst flood in U.S. history (Changnon 1996).

Such events point to the complexity of mitigating hazards. While mitigation efforts may protect against anticipated or typical hazards, they may also have the effect of attracting more people to live and work in hazardous areas, thus increasing exposure over the long term to even larger events. (This trend is reinforced by the apparent security provided by hazard insurance and disaster relief programs.)

Mitigation of hydrological hazards in particular can alter the function of natural systems in ways that are not sustainable over the long term, both because such altered systems may behave in unanticipated ways and because "unprecedented," and thus unplanned-for, events will inevitably occur at some point, in some areas. Mitigation efforts, it seems—especially those focused on trying to control the behavior of the environment through engineered structures—may have the affect of trading a number of smaller, more manageable events in the short-to-medium term for much greater disasters in the more distant future. This can become a self-perpetuating and self-amplifying process, because after a disaster occurs political pressure inevitably focuses on allowing people to return to their homes and businesses to reopen, which in turn requires increased commitment to environmental control via structural hazard mitigation.

Policy Assessment

While societies have an obligation to limit the negative effects of hazards on people and economies, such action should be informed by the inevitability of hazards, rather than a vain quest to eliminate their impacts or occurrence. Such a perspective focuses on the characteristics of human development, rather than the control of nature, as the cornerstone of effective mitigation. For example, environmental degradation invariably exacerbates hazard damages by altering or destroying natural features that buffer the impacts of hazards—such as forests that stabilize steep slopes, floodplains that allow for dispersion of floodwaters, and coastal lagoons that absorb storm surges. Mitigation policies that keep such features intact, and govern land use in ways that protect them over the long term, are likely to be successful both because they preserve natural function and because they thereby limit human development in particularly hazard-prone areas. In acknowledgement of these realities, after the 1993 floods in the Midwest, the U.S. government increased efforts to remove floodplain structures—thus returning some of the natural function of the river—and relocate flood-prone communities to higher ground.

Yet it remains to be seen if it is possible to actually stabilize or reduce the costs of natural hazards in developed countries characterized by continual growth of wealth, infrastructure, urban centers, coastal and wildland development, and overall interconnectedness. Hazards may simply be an unavoidable overhead cost on the growth of affluence.

Outside of the developed world, however, the path to reducing the toll of hazards is clear, if difficult to follow. Poverty and the conditions associated with it—poorly constructed and maintained housing and infrastructure, degraded environmental conditions, rapidly increasing populations, insufficient or ineffective social and emergency services, lack of technical capacity—are the nutrients of hazards. At the global scale, reducing poverty, and the environmental degradation and failures of governance that accompany it, will continue to be the most effective strategy for hazard mitigation.

DANIEL SAREWITZ

SEE ALSO Building Destruction and Collapse; Safety Engineering: Practices.

BIBLIOGRAPHY

Changnon, Stanley A., ed. (1996). The Great Flood of 1993: Causes, Impacts, and Responses. Boulder, CO: Westview. Scientific and policy perspectives on this historical event.

Klinenberg, Eric. (2002). Heat Wave: A Social Autopsy of Disaster in Chicago. Chicago: University of Chicago Press. An analysis of the social imbeddedness of a hazard and its impacts.

Mileti, Dennis S. (1999). Disasters by Design: A Reassessment of Natural Hazards in the United States. Washington, DC: Joseph Henry Press. A comprehensive discussion of the social context of hazards.

Olson, Richard Stuart, et al. (2001). The Storms of '98: Hurricanes Georges and Mitch; Impacts, Institutional Response, and Disaster Politics in Three Countries. Boulder: University of Colorado, Natural Hazards Research and Applications Information Center. Basic facts and political interpretation of these catastrophic storms.

Pielke, Roger A., Jr., and Christopher W. Landsea. (1998). "Normalized Hurricane Damages in the United States: 1925–95." Weather and Forecasting 13(3): 621–631. Analysis of the causes of increasing coastal hazard losses.

United Nations Development Programme. (2002). Human Development Report. New York: Oxford University Press. Source for national economic data.

INTERNET RESOURCES

Munich Re. "Annual Review of Natural Catastrophes, 2002." Available from http://www.munichre.com/2/publications_db_e.asp. Data on costs of hazard losses worldwide.

World Health Organization. Collaborating Centre for Research on the Epidemiology of Disasters (CRED). "EM-DAT: The OFDA/CRED International Disasters Data Base." Available from http://www.em-dat.net/. Unless otherwise noted, all data on hazard losses in the entry are derived from this unique database.