How to Guard Against Earthquakes

The recent earthquake in Soviet Armenia, in which thousands died, and the earthquake that came later in Soviet Tadzhikistan, forces us to think about our trembling Earth again.

The Earth does tremble and this can’t be helped. It is geologically alive. Its crust is made up of half a dozen large plates and a number of smaller ones. These are constantly (if very slowly) moving against each other and building up terrific strains. At some point, the strain overcomes the enormous friction and a plate slips slightly. That is enough to set up the fearsome vibrations that represent an earthquake, sometimes a monstrous one.

It is not, however, the earthquake itself that kills people; falling buildings do. If someone is caught in the open in an earthquake, the shaking of the Earth may be frightening indeed, but it will be over in five minutes or so and you will be safe. If, however, you are within a building at the time, the building may collapse and you may be killed. The worst earthquake we know of killed an estimated 830,000 people in China in 1556. We cannot know how accurate that estimate is, but 242,000 were killed in an earthquake in a city in China in 1976.

How can we prevent this?

--We might argue that we should not build cities too close to known plate junctions, where most of the major earthquakes are known to occur. That is a lost cause, however. Cities are built where climate and communications are easy--on coastlines, on rivers, in valleys--and it so happens that some two-fifths of the major cities on Earth happen to be built near plate junctions. There’s no way of shifting them.

Furthermore, the situation grows steadily worse, for the cities grow larger and will continue to do so for at least a while. The largest city in the world now is Mexico City, which has at least 15 million people within reach of a major earthquake and which lost 5,000 people to one that was not very close by in 1985. By the year 2000, the number of people at risk in Mexico City will be perhaps 25 million.

This can be repeated for any of a hundred cities, with a total population of perhaps 290 million at risk in the year 2000. The mind boggles at what might happen.

--If we can’t help having cities in dangerous areas, can we construct them so as to be earthquake-resistant? The havoc in Soviet Armenia came about in part because of the shoddy construction of the buildings. We can feel superior about this, but when we say shoddy construction we mean cheap construction and when we talk about earthquake resistance, we mean expensive construction. The fact is that it is difficult to invest in expensive construction in an economy that has little surplus wealth and with many problems more immediate than an earthquake that may or may not come for a long time.

About 80% of the 290 million people at risk in the year 2000 will live in developing nations that are simply not going to have the means to keep all their millions in safe structures. Even the United States may not be able to keep pace with structural deterioration, if our deficit problems continue and grow worse.

--Perhaps we can predict earthquakes and help matters in that way. But if we learn to tell that an earthquake is going to take place near a certain city next year, that does not give us time to reinforce the city and make it safe. All we can do is evacuate the city. Getting millions of people out of a city and keeping them out until the city falls down is an impossible task, and, even if it could be done, the resulting dislocation might be as traumatic as the actual earthquake.

What we need to do, then, is to learn how to make long-range predictions of earthquakes. Suppose we could tell, with good accuracy, that a certain city will suffer a major earthquake in 55 years, give or take a year or so. That would give time to strengthen structures, to demolish and rebuild, with perhaps the entire world contributing. Little by little all cities would become safer and the potential for earthquake damage, while probably never eliminated altogether, would be minimized.

How do we make long-range predictions? Techniques have already been developed to measure the tiny motions of the geologic plates that cause earthquakes, making use of fancy interferometers (instruments that measure wavelengths of light) and lasers. Using these measurements, scientists can measure the strains that build up, but it is difficult to make use of these devices everywhere on Earth.

It may well be possible, however, to measure the slight changes in the shape of the Earth and the slight movements of its surface from satellites--a kind of “global geodesy” made possible by the view from space. We could get a general picture of the motions of all the plates, the building up of strains everywhere. A computerized analysis may tell us just when the strains will give at certain points. Then we can build a real picture of future earthquake distribution in both time and space and take the best action to avoid major damage.

To those who think that space projects are a waste of money that could be spent more usefully on “down-to-Earth” problems, we can ask: What can be more down-to-Earth than what I have just described?