Slow-Onset Versus Rapid-Onset Hazards

The distinction between slow- and rapid-onset hazards is useful because the methods to deal with them often differ. Rapid-onset hazards often occur with violent intensity and have profound effects on the surrounding environment, resulting in measurable numbers of casualties and damage. Slow-onset climatic changes brought on by deforestation, drought, desertification, or environmental pollution change the suitability of different parts of the world for human habitation, and they affect agriculture and flora and fauna. The effects of slow-onset disasters are often insidious. Their impact can be measured only through environmental studies and in terms of reduction in quality of life and productivity for the affected population. As discussed later, slow onset hazards rarely are identified as disasters or emergencies, although their impacts may be greater or similarly widespread. Variables include levels of public attention to the hazards and the ability of the government to deal with them. Typically, threats become mitigated by slow movement away from hazards; for example, many of the 11,000 residents of the Pacific island nation of Tuvalu have left the country because the island is being swallowed by sea level rise.

Conservatively, an estimated 1.5 million to 2 million people have been killed in rapid-onset disasters since 1946—an average annual death toll of 35,000 to 50,000. The primary killers are earthquakes, tropical cyclones, and floods. Most deaths are concentrated in a relatively small number of communities, predominantly in poorer nations of Africa, Asia, Latin America, and Oceania. In comparison, North America, Europe, Japan, and Australia have average annual death tolls that rarely exceed a few hundred persons. In terms of total numbers of disasters worldwide, during the period from 1991 to 2005, flood disasters occurred most frequently, followed by hurricanes or cyclones and other windstorms.

Although comprehensive data for economic losses from rapid-onset hazards are difficult to obtain, a few examples illustrate the scale of the problem. Annual worldwide losses from tropical cyclones are estimated at between $6 billion and $7 billion. For landslides, the comparable figure exceeds $1 billion. These figures only hint at the impact of such disasters on the affected human population. The eruption of Colombia’s Nevado del Ruiz volcano in 1985 killed approximately 22,000 people and left 10,000 more homeless. An earthquake in Bam, Iran, in December of 2003 claimed at least 30,000 lives and destroyed 80% of the city. The 2005 Atlantic hurricane season caused record damages of $100 billion. Hurricane Katrina alone, in August of 2005, killed 1417 people in three U.S. states, displaced 1.5 million, and caused $75 billion in damages.

The relative human, economic, and social impacts of rapid-onset disasters are usually greatest in smaller, poorer nations. The 1985 earthquake in Mexico City caused economic losses equivalent to about 3.5% of Mexico’s gross national product (GNP). Hurricane Allen in 1980 caused losses in St. Lucia equivalent to 89% of this island nation’s GNP and destroyed 90% of its banana crop, which normally accounts for 80% of the country’s agricultural output. One of the strongest storms in recent history, Hurricane Mitch in 1998 devastated the economies and infrastructure of Honduras and Nicaragua.

Economic losses from rapid-onset hazards are increasing at a fast pace. In the United States, damage to buildings from earthquakes, tropical cyclones, and floods was estimated to increase from approximately $6 billion in 1978 to more than $11 billion in 2000 without additional loss reduction measures. At this same time, it was estimated that a major earthquake in Tokyo would probably kill more than 30,000 people, cause the collapse of 60,000 houses, and set fire to more than 400,000 homes.

Slow-onset disasters take an even greater toll, but precise figures are difficult to establish. Drought currently affects more people than does any other disaster; worldwide, it is estimated that droughts affected more than 18 million people each year during the 1960s, more than 24 million people during the 1970s, and 101 million between 1980 and 1989. During the early 1980s, drought affected up to 30 million people in Africa alone. In the United States, drought leads in economic impact, causing losses of $6 billion to 8 billion per year. Worldwide, droughts have led to famines, resulting in large numbers of deaths and displacements. Increasing desertification in arid areas may be contributing to droughts. Desertification, or decline in biologic productivity, extends to 70% of total productive arid lands (3.6 billion acres worldwide) and may adversely affect the quality of life for 10% of the world’s population, including urban dwellers.

Possible global warming is predicted to occur over the next 100 years as a result of increased atmospheric carbon dioxide caused by the burning of fossil fuels, deforestation, and generation of methane. If this occurs, sea levels will rise and coastal cities worldwide will be inundated. A rise of 1 m (3.3 feet) in sea level could flood 15% of arable land in Egypt’s Nile Delta and completely submerge the tiny islands of the Maldives, currently inhabited by 200,000. Hundreds of millions of people will also be affected if increased ultraviolet radiation is delivered to the earth’s surface as a result of stratospheric ozone depletion caused by continued release of chlorofluorocarbons.

Although global warming and ozone depletion are threats that may become more evident in the future, other forms of environmental pollution, such as water and air pollution, have immediate effects on life today. Massive oil spills, such as the 2010 leakages in the Gulf of Mexico, make headlines, and adverse health effects are seen from contamination and smog. Deforestation, particularly in the tropical rainforests, is highly significant. In addition to its contribution to possible global warming, the loss of forested land increases vulnerability to droughts, landslides, and floods.

Assessing Vulnerability and Risk

Not all hazards become disasters. Whether or not a disaster occurs depends on the magnitude, intensity, and duration of the event and the vulnerability of the community. For example, a severe earthquake is not a disaster unless it significantly disrupts a community by creating large numbers of casualties and substantial destruction. Effective disaster risk management requires information about the magnitude of the risk faced and how much importance society places on the reduction of that risk. Risks are often quantified in aggregated ways (e.g., a probability of 1 in 23,000 per year of dying in an earthquake in Iran). The importance placed on the risk for a hazard is likely to be influenced by the nature of the risks faced on a daily basis. For instance, in Pakistan, where communities are regularly affected by floods, earthquakes, and landslides, people use their meager resources to protect against what they perceive to be the greater risks, such as disease and irrigation failure. In California, where risk for disease is low, communities choose to initiate programs against natural disasters.

Vulnerability is often measured as the susceptibility of buildings, infrastructure, economy, and natural resources to damage from hazards. Many aspects of vulnerability, however, cannot be described in monetary terms and should not be overlooked. These include personal loss of family, home, and income, along with related human suffering and psychosocial problems. Although communities in developed nations may be as prone to hazards as those living in poorer nations, wealthier communities are often less vulnerable to damage. For example, although both southern California and Managua, Nicaragua, are prone to earthquakes, California is less vulnerable to damage because of strictly enforced building codes, zoning regulations, earthquake preparedness training, and sophisticated communications systems. In 1971, the San Fernando earthquake in California measured 6.4 on the Richter scale but caused minor damage and 58 deaths, whereas an earthquake of similar magnitude that struck Managua, Nicaragua, 2 years later reduced the center of the city to rubble, killing approximately 6000 people. Similarly, in wealthy countries, drought and resulting loss of food production and groundwater are managed by use of food surpluses and treated water, but drought in poor nations often leads to deaths from famine, as well as sickness and death from contaminated water supplies.

Disaster Mitigation Strategies

Mitigation involves not only saving lives and reducing injury and property losses, but also reducing the adverse consequences of hazards to economic activities and social institutions. Where resources are limited, they should be directed toward protecting the most vulnerable elements. Vulnerability also implies a lack of resources for rapid recovery.

For most risks associated with natural geophysical hazards, such as volcanic eruptions, tsunamis, and tropical cyclones, little or no opportunity is available to reduce the hazard itself. In these cases, the emphasis must be placed on reducing the vulnerability of the elements at risk. However, for technologic and human-made hazards or slow-onset hazards, such as environmental pollution and desertification, reducing the hazard is likely to be the most effective mitigation strategy.

Actions by planning authorities to reduce vulnerability can be active, in which desired actions are promoted through incentives, or passive, in which undesired actions are prevented by use of controls and penalties. Discussion of mitigation options follows.

Nature of Hazards

Some hazards exist naturally, and others are partially rooted in natural systems. Many of these occur infrequently or affect only small populations. One example is the eruption of toxic gases from several volcanic lakes in Cameroon that killed 2000 people in 1984 and 1986. Other rare events, such as meteor impacts, may occur only once every few centuries. Additional widespread but minor phenomena that damage property but do not generally cause loss of life include land subsidence and sinkholes. Some hazards, such as snowstorms, often occur in areas that are prepared to deal with them, and thus they rarely become disasters.

This chapter discusses hazards that affect large populations and that can be categorized as follows:

To plan appropriate responses to implement emergency medical care and other measures to save or restore physical and mental health of affected populations, governments and communities first need to understand the causal phenomena, characteristics, and predictability of the hazards and the factors that contribute to vulnerability. Examination of the hazard’s effects on humans, property, and the environment can promote measures to prevent or lessen casualties and destruction.

Causal Phenomena

The earth’s crust is a rock layer varying in thickness from a depth of about 10 km (6.2 miles) under the oceans to 65 km (40.4 miles) under the continents. The theory of plate tectonics holds that seven major and about six minor crustal plates, varying in size from a few hundred to many thousands of kilometers, “ride” on the earth’s mobile mantle. When the plates contact each other, stresses arise in the crust. Stresses occur along the plate boundaries by pulling away from, sliding alongside, and pushing against one another. All these movements are associated with earthquakes.

Faults are areas of stress at plate boundaries that release accumulated energy by slipping or rupturing. Elastic rebound occurs when the maximum point of supportable strain is reached and a rupture occurs, allowing the rock to rebound until the strain is relieved (Figure 89-1). Usually, the rock rebounds on both sides of the fault in opposite directions. The point of rupture is called the focus and may be located near the surface or deep below it. The point on the surface directly above the focus is termed the epicenter (Figure 89-2).

FIGURE 89-1 Elastic rebound in earthquake. A, Forces build up over time. B, Crust deforms. C, Crust snaps. D, Plates slide.

FIGURE 89-2 Motion of the earth’s plates causes increased pressure at faults where the plates meet. Eventually, the rock structure collapses, and movement occurs along the fault. Energy is propagated to the surface above and radiates outward. Waves of motion in the earth’s crust shake landforms and buildings, causing damage.

(Courtesy Disaster Management Center, University of Wisconsin.)

The energy generated by an earthquake is not always released violently and can be small or gradual. Minor Earth tremors are recorded daily in the United States, but whether these are caused by the same processes that can level a city is not known. Most damaging earthquakes are associated with sudden ruptures of the crust.

Characteristics

The actual rupture process may last from a fraction of a second to a few minutes for a major earthquake. Seismic (from the Greek seismos, meaning “shock” or “earthquake”) waves are generated. These last from less than one-tenth of a second to a few minutes and cause ground shaking. The seismic waves propagate in all directions, causing vibrations that damage vulnerable structures and infrastructure.

There are three types of seismic waves. The body waves (P, or primary, and S, or secondary) penetrate the body of the earth, vibrating quickly (Figure 89-3). P waves travel at about 6 km per second (kps) (3.7 miles per second [mps]) and provide the initial jolt that causes buildings to vibrate up and down. S waves travel about 4 kps (2.5 mps) in a movement similar to the snap of a whip, causing a sharper jolt that vibrates structures from side to side and usually resulting in the most destruction. Surface waves (L waves) vibrate the ground horizontally and vertically and cause swaying of tall buildings, even at great distances from the epicenter.

FIGURE 89-3 A, Propagation of seismic waves in an earthquake. Surface waves vibrate the ground horizontally (B and C) and vertically (D and E).

Earthquake focus depth is an important factor in determining the characteristics of the waves. The focus depth can be deep (from 300 to 700 km [186 to 435 miles]) or shallow (less than 60 km [37 miles]). Shallow-focus earthquakes are extremely damaging because of their proximity to the surface. The earthquake may be preceded by preliminary tremors and followed by aftershocks of decreasing intensity.

Earthquake Scales

Earthquakes can be described by use of two distinctly different scales of measurement demonstrating magnitude and intensity. Earthquake magnitude, or amount of energy released, is determined by use of a seismograph, which records ground vibrations. The Richter scale mathematically adjusts the readings for the distance of the instrument from the epicenter. The Richter scale is logarithmic; an increase of one magnitude signifies a tenfold increase in ground motion, or about 30 times the energy. Thus an earthquake with a Richter magnitude of 7.5 releases 30 times more energy than one with a 6.5 Richter magnitude. The smallest quake to be felt by humans was of magnitude 3. The largest earthquakes that have been recorded under this system are 9.5 (Chile, 1960) and 9.25 (Alaska, 1969).

The moment magnitude scale is a successor to the Richter scale and is most often used to estimate large earthquake magnitudes. Theoretically, all magnitude scales should yield approximately the same value for any given earthquake. However, controversy exists over the measurement of the great Indian Ocean earthquake that occurred on December 26, 2004, generating a tsunami that killed more than 280,000 people. The Pacific Tsunami Warning Centre (PTWC) estimated the magnitude as 8.5 on the Richter scale shortly after the earthquake. The U.S. Geological Survey (USGS), using the moment magnitude scale, increased its estimate from 8.1 to 9.0 Other scientists, using moment magnitude, have revised the estimate to 9.3, and the PTWC has accepted this, but the USGS has so far not changed its estimate of 9.0. The most definitive estimate so far has put the magnitude at 9.15.

The earthquake intensity scale measures the effects of an earthquake where it occurs. The most widely used scale of this type is the modified Mercalli scale, which expresses the intensity of earthquake effects on people, structures, and the earth’s surface in values from I to XII (Table 89-1). Another, more explicit, scale used in Europe is the Medvedev-Sponheuer-Karnik (MSK) scale.

TABLE 89-1 Modified Mercalli Intensity Scale of 1931

Location and Predictability

Most earthquakes (95%) occur in well-defined zones near the boundaries of the tectonic plates. These areas bordering the Pacific Ocean are called the circum-Pacific belt. Areas traversing the East Indies, the Himalayas, Iran, Turkey, and the Balkans are called the Alpide belt. Earthquakes also occur along the ocean trenches, such as those around the Aleutian Islands, Tonga, Japan, and Chile and within the eastern Caribbean. Some earthquakes occur in the middle of the plates, possibly indicating where earlier plate boundaries might have been. These have included the New Madrid earthquake in 1811 and the Charleston earthquake in 1816 in the United States, the Agadir earthquake in 1960 in Morocco, and the Koyna earthquake in 1967 in India.

Earthquake prediction was a constant preoccupation for early astrologers and prophets. Some signs of earthquake noted by observers were buildings gently trembling, animals and birds becoming excited, and well water turning cloudy and smelling bad. Although some modern scientists claim ability to predict earthquakes, the methods are still controversial. For example, the 1995 earthquake in Kobe, Japan, which killed more than 5000 people, was not predicted. However, mechanical observation systems make it possible to issue warnings to nearby populations immediately after detection of an earthquake. Reasonable risk assessments of potential earthquake activity can be made with confidence based on the following:

The island of Hispaniola, shared by Haiti and the Dominican Republic, has a history of destructive earthquakes. In January 2010, a magnitude 7.0 earthquake occurred approximately 25 km (16 miles) west-southwest from the capital city Port-au-Prince at a depth of 13 km (8.1 miles). Two years before this, scientists had detected signs of growing stresses in the fault that forms a boundary between the Gonave microplate and the Caribbean plate to the south, specifically the Enriquillo-Plantain Garden fault system, which includes much of Haiti. They warned Haitian officials that the fault was capable of causing a 7.2 magnitude earthquake, only slightly stronger than the actual 7.0 earthquake that eventually occurred. Unfortunately, 2 years is little time to prepare for such an event in a country like Haiti, which endures widespread poverty and lacks resources for preparedness and mitigation. A legacy of poor building standards has increased vulnerability and cannot be easily remedied.

Earthquake Hazards

Earthquakes produce many direct, and sometimes indirect, effects. Landslides, flooding, and tsunamis are considered secondary hazards and are discussed later in this chapter.

Typical Adverse Effects

Ground shaking can damage human settlements, buildings and infrastructure (particularly bridges), elevated roads, railways, water towers, water treatment facilities, utility lines, pipelines, electricity generating facilities, and transformer stations. Aftershocks can do great damage to already weakened structures. Significant secondary effects include fires, dam failures, and landslides, which may block waterways and cause flooding. Flooding may also be caused by seiches (back-and-forth wave action in bays) or by failures in dams and levees. Damage may occur to facilities that use or manufacture dangerous materials, resulting in chemical spills. Communications facilities may break down. Destruction of property may have a serious impact on shelter needs, economic production, and living standards of the affected community. Depending on their level of vulnerability, many people may be homeless in the aftermath of an earthquake.

The casualty rate is often high, especially when earthquakes occur in areas of high population density, particularly when streets between buildings are narrow, buildings are not earthquake resistant, the ground is sloping and unstable, or adobe or dry stone construction is used, with heavy upper floors and roofs.

Casualty rates may be high when quakes occur at night because the preliminary tremors are not felt during sleep and people are not tuned in to receive media warnings. In the daytime, people are particularly vulnerable in large unsafe structures such as schools and offices. Casualties generally decrease with distance from the epicenter. As a rule of thumb, quakes result in three times as many injured survivors as persons killed. The proportion of dead may be higher with major landslides and other secondary hazards. In areas where houses are of lightweight construction, especially with wood frames, casualties are generally much fewer, and earthquakes may occur regularly with no serious, direct effects on human populations.

The most widespread acute, serious medical problems are broken bones. Other health threats may occur with secondary flooding, when water supplies are disrupted (earthquakes can change levels in the water table) and contaminated water is used or water shortages exist, and when people are living in high-density relief camps, where epidemics may develop or food shortages exist.

In the aftermath of the Colombia earthquake of January 1999, which most heavily affected the city of Armenia, the death toll was 1185, and 160,000 people were left homeless, most in urban areas. In Armenia, where 60% to 70% of homes had been destroyed, movement was restricted by fallen debris and unemployment rose from 12% to 35%. People were living in unsatisfactory shelters made with plastic sheeting. Many migrated from the area to other places that could not absorb them. Although international response to aid Colombia was strong, the overwhelming need continued to pose problems. Five weeks after the earthquake, supplies of food, clean drinking water, and shelter materials were still urgently required. Hygiene and sanitation services and essential medicines were desperately needed. Social services were required to work toward normalizing the lives of victims, especially children. In Gujarat, India, where 30,000 people died in 2001, assistance agencies struggled for years to help rebuild the more than 300,000 houses that were lost. In the Kashmir earthquake of 2005, 3.3 million persons were left homeless in Pakistan, and many of them were at risk of dying from the winter cold and spread of disease.

The January 2010 Haiti earthquake affected an estimated 3 million people. It killed approximately 100,000 persons and injured approximately 300,000, although estimates of casualties widely vary. More than 1 million Haitians were left homeless. Vital infrastructure necessary to respond to the disaster, including air, sea, and land transport facilities and communication systems, was severely damaged or destroyed. Treatment of the injured was hampered by the lack of hospitals and morgue facilities; bodies were left to decay on the streets for many days. International assistance was offered in abundance, but the logistical capabilities in Haiti to receive emergency aid were limited. Doubtless, more lives were lost as a result of this vulnerability (Figures 89-4 and 89-5, online).

FIGURE 89-4 Nurse attending earthquake casualties in Haiti.

(Courtesy Pan America Health Organization.)

FIGURE 89-5 Makeshift hospital in Haiti.

(Courtesy Pan America Health Organization.)

Earthquake Risk Reduction Measures

Earthquake warning systems currently in use warn of an earthquake that has already occurred. Examples include those that notify the high-speed trains in Japan, which if derailed would cause hundreds of deaths. One minute before the 2011 Tohoku earthquake was felt in Tokyo, 1000 seismometers sent out warnings that saved many lives. In California, it is technically feasible to develop a system that could warn Los Angeles up to a minute before the arrival of the seismic waves, allowing certain preventive actions, such as taking cover, to occur. However, because predicting the location, time, and magnitude of earthquakes is still likely many years away, warning systems and earthquake prevention measures are currently not reliable alternatives to preparedness. Preparedness actions include the following:

FIGURE 89-6 Earthquake resistant lintels in Afghanistan.

(Courtesy Sheila B. Reed.)

Causal Phenomena and Characteristics

The geologic movements that cause a tsunami are produced in three major ways (Figure 89-7). The foremost cause is fault movement on the sea floor, accompanied by an earthquake. The second most common cause is a landslide occurring underwater or originating above the sea and then plunging into the water. The highest tsunamis ever reported were produced by a landslide at Lituya Bay, Alaska, in 1958. A massive rock slide produced a wave that reached a high water mark of 530 m (1740 feet) above the shoreline. A third cause of a tsunami is volcanic activity, which may uplift the flank of the volcano or cause an explosion.

FIGURE 89-7 Tsunamis are produced in three ways. A, Fault movement on the sea floor. B, Landslide. C, Submarine explosion from volcanic eruption.

Tsunamis differ from ordinary deep ocean waves, which are produced by wind blowing over water. Normal waves are rarely longer than 300 m (984 feet) from crest to crest. Tsunamis, however, may measure 150 km (90 miles) between successive wave crests. Tsunamis also travel much faster than ordinary waves. Compared with normal wave speed of around 100 km/hr (62 mph), tsunamis in the deep water of the ocean may travel at the speed of a jet airplane—800 km/hr (497 mph). Despite their speed, tsunamis increase the water height only 30 to 45 cm (12 to 18 inches) and often pass unnoticed beneath ships at sea. In 1946, a ship’s captain on a vessel lying offshore near Hilo, Hawaii, claimed he could feel no unusual waves beneath him, although he saw them crashing on the shore.

Contrary to popular belief, a tsunami is not a single giant wave. A tsunami can consist of 10 or more waves, termed a tsunami wave train. The waves follow each other in 5- to 90-minute intervals. As tsunamis approach the shore, they travel progressively slower. The final wave speed depends on the water depth. Waves in 18 m (59 feet) of water travel about 50 km/hr (31 mph). The shape of the near-shore sea floor influences how tsunami waves behave. Where the shore drops off quickly into deep water, the waves are smaller. Areas with long shallow shelves, such as the major Hawaiian Islands, allow formation of very high waves. In the bays and estuaries, seiches, in which the water sloshes back and forth, can amplify waves to some of the greatest heights ever observed.

On shore, the initial sign of a tsunami depends on what part of the wave first reaches land; a wave crest causes a rise in the water level, and a wave trough causes a recession. The rise may not be significant enough to be noticed by the general public. Observers are more likely to notice the withdrawal of water, which may leave fish floundering on the exposed sea floor. A tsunami does not always appear as a vertical wall of water, known as a bore, as typically portrayed in drawings. More often, the effect is that of an incoming tide that floods the land. Normal waves and swells may ride on top of the tsunami wave, or the tsunami may roll across relatively calm inland waters.

The flooding produced by a tsunami may vary greatly from place to place over a short distance, depending on submarine topography, shape of the shoreline, reflected waves, and modification of waves by seiches and tides. The Hilo, Hawaii, tsunami of 1946, originating in the Aleutian Trench, produced 18-m (59-foot) waves in one location and only one-half that height a few miles away. The sequence of the largest wave in the tsunami wave train also varies, and the destructiveness is not always predictable. In 1960 in Hilo, many people returned to their homes after two waves had passed, only to be swallowed up in a giant bore that, in this case, was the third wave.

Predictability

Tsunamis have occurred in all oceans and in the Mediterranean Sea, but the majority of them occur in the Pacific Ocean. The zones stretching from New Zealand through East Asia, the Aleutians, and the western coasts of the Americas all the way to the South Shetland Islands are characterized by deep ocean trenches, explosive volcanic islands, and dynamic mountain ranges.

Prior to 1946, the recorded effects of tsunamis included only local casualties and significant damage. The Tsunami Warning System (TWS) was developed in Hawaii shortly after the 1946 Hilo tsunami and is headquartered in the Pacific Warning Center in Honolulu. There are 26 member countries in the Pacific basin. The TWS works by monitoring seismic activity from a network of seismic stations. A tsunami is almost always generated by an undersea earthquake of magnitude 7 or greater. Therefore special warning alarms sound when a quake measuring 6.5 or more occurs anywhere near the Pacific. A tsunami watch is declared if the epicenter is close enough to the ocean to be of concern. Government and voluntary agencies are then alerted, and local media are activated to broadcast information. The five nearest tide stations monitor their gauges, and trained observers watch the waves. With positive indicators, a tsunami warning is issued.

The TWS met with general success in saving lives during the tsunamis of 1952 and 1957 in Hawaii. In 1960, however, two major earthquakes occurring a day apart rocked the coast of Chile in South America. The first registered 7.5 on the Richter scale and produced a small but noticeable wave in Hilo Bay. The second registered a stunning 8.5, more than 30 times the energy of the first, and authorities predicted generation of a large, destructive tsunami. When the waves hit Hilo, 15 hours after the earthquake, not all the public had taken the warnings seriously, and 61 people were killed. About 7 hours later the tsunami struck Japan, killing 180. By the time that information of conditions in Chile reached TWS, three giant waves had already destroyed villages along an 805-km (500-mile) stretch of coastal South America, arriving only 15 minutes after the earthquake.

Lack of an effective warning system has been blamed for the extensive loss of life from the tsunami generated in the Indian Ocean in December 2004. Over 290,000 are estimated to have died in 11 countries, and thousands more remain missing. Tsunamis have been relatively rare in the Indian Ocean, and the area has no international warning system. The first tsunami-generated wave crashed into Sumatra only 30 minutes after shaking from the earthquake had subsided. The tsunami ultimately traveled nearly 5000 km (3107 miles) to Africa. In contrast to stronger preparedness levels in the Pacific countries, citizens and tourists were not fully aware of the dangers and many watched from the beach with catastrophic results. In Kobe, Japan, the World Conference on Disaster Reduction in January of 2005 laid the groundwork for the first tsunami warning system in the Indian Ocean, much of which was positioned in recent years. Indonesia has set up costly and sophisticated tsunami warning systems and carried out numerous drills. However, 400 people were killed in an October 2010 tsunami on the Mentawai Islands, indicating that those most at risk are not able to receive warnings through communications systems or cannot flee from a tsunami generated close to shore.

Vulnerability

The following major factors contribute to vulnerability to tsunamis:

Typical Adverse Effects

The force of water in a bore, with pressures up to 10,000kg/m², can raze everything in its path. The flooding from a tsunami, however, affects human settlements most, by water damage to homes, businesses, roads, and infrastructure. Withdrawal of the tsunami also causes significant damage. As the water is dragged back toward the sea, bottom sediments are scoured out, collapsing piers and port facilities and sweeping out foundations of buildings. Entire beaches have disappeared, and houses have been carried out to sea. Water levels and currents may change unpredictably, and boats of all sizes may be swamped, sunk, or battered (Figures 89-8 to 89-10; Figures 89-8 and 89-10, online).

FIGURE 89-9 Boat perched on a house near Banda Aceh, Indonesia, following the December 2004 tsunami.

(Courtesy Sheila B. Reed.)

FIGURE 89-8 Ship washed ashore in tsunami, December 2004, in Koh Lanta, Thailand.

(Courtesy Sheila B. Reed.)

FIGURE 89-10 Generator ship washed ashore 2.3 miles inland, near Banda Aceh, Indonesia.

(Courtesy Sheila B. Reed.)

Casualties and Public Health

Deaths occur principally from drowning as water inundates homes or neighborhoods. Many people may be washed out to sea or crushed by the giant waves. Injuries occur from battering by debris. Little evidence exists of tsunami flooding directly causing large-scale health problems. Rapid effective assistance to the Banda Aceh (Indonesia) area in early 2005 prevented widespread outbreaks of disease in displacement camps (Figure 89-11). Malaria mosquitoes may increase because of water trapped in pools. Open wells and other groundwater may be contaminated by salt water and debris or sewage. Normal water supplies may be inaccessible for days because of broken water mains.

FIGURE 89-11 Displaced people in temporary settlements in Aceh Province, Indonesia.

(Courtesy Sheila B. Reed.)

Crops and Food Supplies

Flooding and damage by tsunami waves may result in the following:

1 Harvests may be lost, depending on time of year.

2 Land may be rendered infertile from saltwater incursion from the sea.

3 Food stocks not moved to high ground are damaged.

4 Animals not moved to high ground may perish.

5 Farm implements may be lost, hindering tillage.

6 Boats and fishing nets may be lost.

In July 1998, an earthquake of magnitude 7 occurred close to the northwest coast of Papua New Guinea. Although the tremor was felt over a large area, no earthquake damage was reported. Only 10 minutes after the quake, however, the first of three 7- to 10-m (23- to 33-foot) waves came ashore in Sandaun Province. The tsunamis struck at high speed after dark and penetrated up to 1 km (over half a mile) inland, totally destroying villages and vegetation along 50 km (31 miles) of the coast. Of the 9000 people affected, more than 2000 died, mainly as a result of being battered by debris as they were swept away by the water.

Tsunami Risk Reduction Measures

Strategies to reduce vulnerability to tsunamis include:

Causal Phenomena

The basic ingredients for a volcanic eruption are magma and an accumulation of gases beneath an active volcanic vent, which may be either on land or below the sea. Magma is composed of silicates containing dissolved gases and sometimes crystallized minerals in a liquid-like suspension. Driven by buoyancy and gas pressure, magma, which is lighter than surrounding rock, forces its way upward. As it reaches the surface, the pressures decrease, enabling the dissolved gases to effervesce, pushing the magma through the volcanic vent as the gases are released.

The chemical and physical composition of magma determines the amount of force with which a volcano erupts. Magmas that are less viscous allow gas to be released more easily. More viscous magma, perhaps containing a greater concentration of solid particles, may confine these gases longer, allowing greater pressures to build up. This greater pressure may lead to more violent eruptions.

Types

Volcanic eruptions may be described as follows in descending order of intensity (Figure 89-12).

FIGURE 89-12 Eruption types. A, Pelean. B, Plinean. C, Vesuvian. D, Vulcanian. E, Strombolian. F, Hawaiian. G, Icelandic.

(Modified from United Nations Development Program: Introduction to hazards, ed 3, Disaster Management Training Programme, UN Organization for Coordination of Humanitarian Assistance, New York, 1997.)

Characteristics

No international scale exists to measure the size of volcanic eruptions. The volcanic explosivity index (VEI) estimates the energy released in a volcanic eruption, based on measurements of the ejected matter, height of the eruption cloud, and other observations. The VEI scale ranges from 0 to 8. The largest eruption recorded was in Tambora, Indonesia in 1815, which had a VEI of 7.

The primary volcanic hazards are associated with products of the eruption: pyroclastic flows, air-fall tephra, lava flows, and volcanic gases. The most destructive secondary hazards include lahars, landslides, and tsunamis.

Location

The distribution of volcanoes, as with earthquakes, is determined by the location of geologic forces involving the tectonic or crustal plates. About 80% of the active volcanoes are located near subduction boundaries. Subduction volcanoes occur where denser crustal plates are shoved beneath less dense continental plates, which occurs in most of the Pacific Ocean, especially in the area along the rim, known as the Pacific Ring of Fire. Subduction volcanoes are found in the United States in the Cascade Range of the Pacific Northwest and further north in the Aleutian Islands off Alaska. The ring of subduction volcanoes continues along the Aleutian Trench to Japan, stretching south to the Philippines and Indonesia. Many volcanoes are located beneath the ocean, and submarine eruptions may cause tsunamis and other effects.

Rift volcanoes occur at divergent zones where two distinct plates are slowly being separated, in areas such as Iceland and East Africa; they account for about 15% of active volcanoes. Hot spot volcanoes are located where crustal weaknesses allow molten material to penetrate, but not necessarily on the plate boundaries. These isolated regions of volcanic activity exist in about 100 places in the world. The Hawaiian Islands, in the middle of the Pacific plate, and Yellowstone Park, within the North American plate, are good examples.

Predictability

Systematic surveillance of volcanoes, begun early in the 20th century at the Hawaiian Volcano Observatory, indicates that most eruptions are preceded by measurable geophysical and geochemical changes. Short-term forecasts of future volcanic activity in hours or months may be made through volcano monitoring techniques that include seismic monitoring, ground deformation studies, and observations and recordings of hydrothermal, geochemical, and geoelectric changes. By carefully monitoring these factors, scientists were able to issue a high confidence forecast of the 1991 Mt Pinatubo eruption, allowing a largely successful evacuation. The best basis for long-term forecasting (a year or longer) of a possible eruption is through geologic studies of the past history of each volcano. Each past eruption has left records in the form of lava beds. Deposits and layers of ash and tephra can be studied to determine the extent of the flows and length of time between eruptions.

Vulnerability

Rich volcanic soils and scenic terrains attract people to settle on the flanks of volcanoes. These people are more vulnerable if they live downwind from the volcano, in the path of historical channels for mudflows or lava flows, or close to waterways likely to flood because of silting. Structures with roof designs that do not resist ash accumulation are vulnerable even miles from a volcano. All combustible materials are at risk.

Typical Adverse Effects