||The concept of a natural hazard involves the risk of a natural event which could damage people or property. The term is usually used for sudden events, which can vary from very slight effects to the catastrophic impacts of a major flood or earthquake. It is estimated that 2.8 million people were killed by natural events between 1968 and 1988, with the worst cases, an earthquake at Tangshan city in China and two coastal floods in Bangladesh, each killing more than a quarter of a million people. The puzzling thing about natural hazards is that millions of people continue to expose themselves to risk from natural events which are to some degree predictable. The nature of the risks and the possibilities for reducing them vary from hazard to hazard.
Earthquakes occur where movements of the Earth\'s crust build up tension which is suddenly released as a fault moves. They occur mainly near plate boundaries, especially where continental plates collide (especially from the Mediterranean to the Himalayas) or move past each other (as in California). Their magnitude is measured on the Richter scale, ranging from just perceptible at force 2 to massive shaking at force 9. Each step on this scale involves a tenfold increase in amplitude and a 31 fold increase in energy released. Damage depends on magnitude, distance from the epicentre, the nature of the underlying rock and the nature of buildings. Casualties tend to be greater in less-developed areas, where planning regulations and building quality are lower. For example, the 1988 Armenian earthquake was magnitude 6.9 and killed 25,000 while the San Francisco event in 1989 was magnitude 7.1 and killed 30. Although the areas most subject to earthquakes are well known, they can occur in other areas and the location and timing of future events are unpredictable. Many millions of people live in areas subject to earthquakes, including the populations of Los Angeles, San Francisco and Tokyo, so it is certain that disasters will occur in future and possible that some could have worldwide economic impact.
Earthquakes under the sea can trigger tsunami, incorrectly known as tidal waves. The worst on record occurred in Japan in 1896, when a wave 24 metres high killed 27,000 people. The wave travels very quicklyâ€”crossing the Pacific in less than 24 hoursâ€”and is almost imperceptible until it approaches land. Fortunately, the areas at risk are well known and an international warning system allows them to be evacuated, so casualties are much reduced.
Although volcanoes are spectacular, they are responsible for fewer recent casualties than floods or earthquakes. They also tend to occur near plate boundaries, indeed, most occur on the sea floor along ocean ridges where they affect people only on islands, such as Iceland and Hawaii. More damaging are the active volcanoes which build steep cones and sometimes explode. Krakatoa, in 1883, was such an explosion and killed 30,000 people, as was Mount St Helens in 1980, which killed only 50 because it had been predicted and evacuation of the already sparse population had nearly been completed.
Smaller explosions can be dangerous if they trigger landslips. At worst, mudflows can descend river valleys at over 50 km/h and overwhelm settlements, as they did at Nevado del Ruiz, Colombia in 1985, killing 22,000. Lava flows are much slower and can sometimes be impeded or diverted by barriers or pumped water. In many parts of the world people continue to live on volcanoes because they create extremely fertile soils. They may persuade themselves that the volcano is extinct, though vulcanologists now prefer to consider them dormant if not recently active. The prospects for reducing casualties look good as almost all volcanoes give warning signs before erupting.
Although eruptions are a modest hazard in recent history, there are now studies which suggest that some 25 eruptions in the last 10,000 years have had global impacts. Examples are Hekla in 1159 Â BCE, which blew some 12 cubic kilometres of rock into the atmosphere, and the even larger Thera in 1627 Â BCE. Deposits of ash and sulphur compounds plus historical records suggest that these eruptions put so much dust into the stratosphere that solar radiation could hardly reach the surface and massive crop failures occurred in one or more seasons. The result was famine and depopulation on the scale predicted for the â€˜nuclear winterâ€™.
The most serious causes of coastal flooding are tropical cyclones or hurricanes. These occur over the oceans between 8 and 15 degrees N&S. Cyclones are intense areas of low pressure causing torrential rain and winds which reach some 200 km/h. They quickly lose energy when they reach land, but islands and low-lying coasts can be subject to damage from wind and floods, which can raise sea levels by several metres. Areas frequently affected include the bay of Bengal and the Caribbean. In developed countries, weather forecasts offer enough warning to allow evacuation of coastal areas, though the cost of damage to buildings may be high. In poor countries like Guatemala or Bangladesh mass evacuation is less feasible, and many of the islands of Bengal are so low lying that concrete cyclone shelters offer the only hope of survival, but the 200 shelters can accommodate only 3% of the population at risk. Yet such is the poverty and pressure on land that people still choose to live in the coastal zone.
Storm surges can occur on coasts far from the tropics. In 1953 a deep depression in the North Sea brought gales which lifted sea level two to three metres above normal and flooded large areas of Holland and East Anglia. Subsequently, very large sums have been spent in an attempt to prevent a recurrence of this flooding, and the Thames Barrier has been built to guard against the possibility of a flood in central London.
Floods are the most common and among the most damaging of natural hazards because the processes that produce the hazard are also the creators of fertile flood plains, which are rich agricultural areas. Major rivers are also the sites of cities, so hundreds of millions of people live on flood plains in the Third World, as do 10% of the population of the US. In principle, river floods seem amenable to rational precautions because past flood records can be used to calculate the probability of future floods of various depths. It should, therefore, be possible to identify and avoid areas at risk. Unfortunately, this has not proved to be true in practice. The US is an extreme case: in spite of spending $10 billion on flood protection, financial losses grew fourfold between the 1920s and the 1980s in real terms. The range of adjustments to floods includes building dams, levees and channels to prevent flooding reaching populated areas, strengthening buildings and evacuation schemes to reduce losses, land use controls to restrict building on flood plains and insurance or disaster relief to compensate victims. In practice, most of these precautions can create further problems: dams have failed, engineering structures and warning schemes encourage the belief that the problem is solved and encourage building on flood plains. Controlling such building is apparently the most cost-effective policy but few governments have been able to enforce it for long. The attractions of flood plains to farmers and developers are usually sufficient to overcome individual and institutional fear of flooding, especially because the risks are usually underestimated.
It seems certain that natural disasters will continue to extract a toll of human life and economic damage, and likely that hazards will do increasing damage. Many Third World citizens face a choice of starvation or farming in areas exposed to floods. Even people with a choice of where to live seem repeatedly to choose hazardous areasâ€”mass immigration into California in the last 90 years or so is an example. In part this is due to lack of public knowledge of the risks, but also it seems that most people deliberately underestimate the risks to themselves. Public authorities are willing to invest in protective works, but few seem able to impose controls on exposure to hazard for long enough periods to have an effect. The best prospect seems to be for more effective systems of evacuation and disaster relief.
If these kinds of response to hazard seem irrational, it is worth remembering that even in California the probability of being killed in an earthquake is only 1 in 500,000 per annum, whereas the probability of being killed in a car crash is 1 in 20,000. PS
Further reading K. Smith, Natural Hazards: Assessing Risk and Reducing Disaster.