Risks of natural and technological hazards

Historically, the perception of risk, and thus society's motivation for reducing it, is dependent to some extent on exposure to the risk - i.e. its probability of occurrence. Generally, however, the 'dread factor' which is related to the scale of the potential catastrophe has a greater impact. There is little doubt that disasters are formidable and newsworthy events because of the number of deaths that can occur even though higher levels of risk may be represented by losses on a smaller scale. For example, leukemia or diabetes may be the greater killers in aggregate, but the fact that they kill individually and without drama makes them less "newsworthy" and less alarming.

Disasters kill numbers of people at once - an airplane crash causes more horror than an equivalent number of the more frequent car crashes because larger numbers of passengers are killed in a single event. The size of a disaster then, represented by its number of fatalities, is almost as important for its perception as its frequency of occurrence.

Data on the size and frequency of disaster occurrences for a particular country can be described as f:N curves plotting the frequency of events causing greater than a certain Number of fatalities. The f:N curves for several types of disaster for the world as a whole are presented in figure 4.¹¹

It is clear from these that natural disasters greatly exceed technological disasters caused by industry or transportation in their capability to cause massive loss of life. Indeed the scale of energy release that is possible in nature - in a cyclone, flood, volcano or large earthquake (which may be equivalent to hundreds of atomic bombs) - still far outstrips any human-made source of energy. Drought and famine have been the greatest killers this century, though precise numbers killed are difficult to estimate. Among the so-called rapid onset disasters, floods and earthquakes are the world's severest hazards, both in frequency and lethality. Storms, including cyclones and tornados are only slightly less severe.

The largest single life losses from a rapid-onset disaster to have occurred this century were from floods in China; an estimated 2 million people were killed in Northern China in flooding in 1956 and 1.4 million people were reportedly killed in a flood in 1931 on the Yangtze-Kiang river in China. The worst casualty rate from an earthquake this century was also in China, in Tangshan in 1976, when a quarter of a million people died.

These individual events represent extreme causes of severe hazards, dense populations and vulnerable communities: risks which may be extremely rare. Less severe situations resulting in lower death tolls happen more commonly. The f:N curves in figures 4 show the risk per year of disasters of this less severe type happening. Judging by the record of disasters this century, an earthquake killing at least 100,000 people can be expected on average every 15 years. These diagrams tell us something about the levels and scale of risks of disasters faced but risk analysis and vulnerability assessment can also help to structure effective disaster mitigation to reduce risk levels.

Figure 4 - f:N curves for various disaster types (worldwide)

Figure

Figure

Figure

It is clear that natural disasters greatly exceed technological disasters caused by industry or transportation in their capability to cause massive loss of life