by Isaac Asimov
The reverse situation takes place at the height of an ice age. So much water is tied up in the form of land-based icecaps (up to three or four times the present amount) that the sea-level mark is as much as 440 feet lower than it now is. When this is so, the continental shelves are exposed.
The continental shelves are relatively shallow portions of the ocean adjoining the continents. The sea floor slopes more or less gradually to a depth of about 130 meters. After this, the slope is much steeper, and considerably greater depths are achieved rapidly. The continental shelves are, structurally, part of the continents they adjoin: it is the edge of the shelf that is the true boundary of the continent. At the present moment, there is enough water in the ocean basins to flood the borders of the continent.
Nor is the continental shelf small in area. It is much broader in some places than others; there is considerable shelf area off the east coast of the United States, but little off the west coast (which is at the edge of a crustal plate). On the whole, though, the continental shelf is some 50 miles wide on the average and makes up a total area of 10 million square miles. In other words, a potential continental area rather greater than the Soviet Union in size is drowned under the ocean waters.
It is this area that is exposed during periods of maximum glaciation and was indeed exposed in the last great ice ages. Fossils of land animals (such as the teeth of elephants) have been dredged up from the continental shelves, miles from land and under yards of water. What’s more, with the northern continental sections ice-covered, rain was more common than now, farther south, so that the Sahara Desert was then grassland. The drying of the Sahara as the icecaps receded took place not long before the beginning of historic times.
There is thus a pendulum of habitability. As the sea level drops, large continental areas become deserts of ice, but the continental shelves become habitable, as do present-day deserts. As the sea level rises, there is further flooding of the lowlands, but the polar regions become habitable, and again deserts retreat.
You can see, then, that the periods of glaciation were not necessarily times of desolation and catastrophe. All the ice in all the ice sheets at the time of the maximum extent of glaciation makes up only about 0.35 percent of the total water in the ocean. Hence, the ocean is scarcely affected by the oscillations in ice. To be sure, the shallow areas are greatly decreased in area, and those areas are rich in life. On the other hand, the tropic ocean waters are anywhere from 2 to 5 degrees cooler than they are now, which means more oxygen in solution and more life.
Then, too, the advance and retreat of the ice is exceedingly slow, and animal life in general can adapt, migrating slowly north and south. There is even time for evolutionary adaptation to take place, so that during the ice ages, the woolly mammoth flourished.
Finally, the oscillations are not as wild as they might seem, for the ice never entirely melts. The Antarctica icecap has been in existence, relatively unchanged, for some 20 million years and limits the fluctuation in sea level and in temperature.
And yet I do not mean to say that the future gives us no cause for worry. There is no reason to think that a fifth glaciation may not eventually come—with its own problems. In the previous glaciation, the few human beings were hunters who could easily drift southward and northward on the tracks of the game they hooted. In the next glaciation, human beings will undoubtedly be (as they are today) great in numbers and relatively fixed to the ground by virtue of their cities and other structures. Furthermore, it is possible that various facets of human technology may hasten the advance or retreat of the glaciers.
CAUSES OF ICE AGES
The major question regarding the ice ages involves their cause. What makes the ice advance and retreat, and why have the glaciations been relatively brief, the present one having occupied only 1 million of the last 100 million years?
It takes only a small change in temperature to bring on or to terminate an ice age—just enough fall in temperature to accumulate a little more snow in the winter than melts in the summer, or enough rise to melt a little more snow in the summer than falls in the winter. It is estimated that a drop in the earth’s average annual temperature of only 3.5° C is sufficient to make glaciers grow, whereas a rise of the same amount would melt Antarctica and Greenland to bare rock in a matter of centuries.
A small drop in temperature sufficient to increase the ice cover slightly over a few years serves to make the process continue. Ice reflects light more efficiently than bare rock or soil does; ice reflects 90 percent of the light that falls on it, while bare soil reflects less than. 10 percent. A slight increase in ice cover reflects more sunlight and absorbs less, so that the average temperature of the earth would drop a little farther, and the growth of the ice cover would accelerate.
Similarly, if the earth’s temperature went up slightly—just enough to force a small retreat in the ice—less sunlight would be reflected and more absorbed, accelerating the retreat.
What, then, is the process that triggers the action either way?
One possibility is that the earth’s orbit is not entirely fixed and does not repeat itself exactly over the years. For instance, the time of perihelion is not fixed. Right now, perihelion, the time when the sun is closest to Earth, comes shortly after the winter solstice. However, the position of the perihelion shifts steadily and makes a complete circuit of the orbit in 21,310 years. Then, too, the direction of the axis changes and marks out a circle in the sky (the precession of the equinoxes) in 25,780 days. Then, too, the actual amount of the tilt changes very slightly, growing a tiny bit more, then a tiny bit less, and in a slow oscillation.
All these changes have a small effect on Earth’s average temperature—not great, but enough at certain times to pull the trigger for either the advance of the glaciers or their retreat.
In 1920, a Yugoslavian physicist, Milutin Milankovich, suggested a cycle of this sort that was 40,000 years in length, with a “Great Spring,” a “Great Summer,” a “Great Fall,” and a “Great Winter,” each 10,000 years long. The earth would, by this theory, be particularly susceptible to glaciation in the time of the “Great Winter” and would actually undergo it when other factors were favorable as well. Once glaciated, the earth would undergo deglaciation most likely in the “Great Summer” if other factors were favorable.
Milankovich’s suggestion did not meet with much favor when it was advanced; but in 1976, the problem was tackled by J. D. Hays and John Imbrie of the United States and by N. 1. Shackleton of Great Britain. They worked on long cores of sediment dredged up from two different places in the Indian Ocean—relatively shallow places far from land, so that no contaminating material would be brought down from nearby coastal areas or shallower sea bottom.
These cores were made up of material laid down steadily over a period of 450,000 years. The farther down the core one observed, the farther back the year. It was possible to study the skeletons of tiny one-celled animals, which come in different species that flourish at different temperatures. From the nature of the skeleton, the temperature could be determined.
Then, too, oxygen atoms chiefly come in two different varieties, and the ratio of these varieties vary with the temperature. By measuring the ratio at different places in the core, one could determine the ocean temperature at different times.
Both methods of measuring temperature agreed, and both seemed to indicate something much like the Milankovich cycle. It may be, then, that the earth has a glaciated Great Winter at long intervals, just as it has a snow-covered winter every year.
But then why should the Milankovich cycle have worked during the course of the Pleistocene but not for a couple of hundred million years before that when there was no glaciation at all?
In 1953, Maurice Ewing and William L. Donn suggested the reason might lie in the peculiar geography of the Northern Hemisphere. The Arctic region is almost entirely oceanic, but it is a landlocked ocean with large continental masses hemming it in on all sides.
Imagine t
he Arctic Ocean a trifle warmer than it is today, with little or no sea ice upon it and offering an unbroken stretch of liquid surface. The Arctic Ocean would then serve as a source of water vapor, which, cooling in the upper atmosphere, would fall as snow. The snow that fell back into the ocean would melt, but the snow that fell on the surrounding continental masses would accumulate, and trigger the glaciation: the temperature would drop, and the Arctic Ocean would freeze over.
Ice does not liberate as much water vapor as does liquid water at the same temperature. Once the Arctic Ocean freezes over, then, there is less water vapor in the air and less snowfall. The glaciers start retreating, and if they then trigger deglaciation, the retreat is accelerated.
It may be, then, that the Milankovich cycle sets off periods of glaciation only when there is a landlocked ocean at one or both poles. There may be some hundreds of millions of years when no such landlocked ocean exists and there is no glaciation; then the shifting of the tectonic plates creates such a situation, and there begins a million years or more during which the glaciers advance and retreat regularly. This interesting suggestion is not as yet totally accepted.
There are, to be sure, less regular changes in Earth’s temperature and more erratic producers of cooling and warming trends. The American chemist Jacob Bigeleisen, working with H. C. Urey, measured the ratio of the two varieties of oxygen atom in the ancient fossils of sea animals in order to measure the temperature of the water in which the animals lived. By 1950, Urey and his group had developed the technique to so fine a point that, by analyzing the shell layers of a millions-of-years-old fossil (an extinct form of squid), they could determine that the creature had been born during a summer, lived four years, and died in the spring.
This “thermometer” has established that 100 million years ago the average world-wide ocean temperature was about 70° F. It cooled slowly to 61° F 10 million years later and then rose to 70° F again after another 10 million years. Since then, the ocean temperature has declined steadily. Whatever triggered this decline may also be a factor in the extinction of the dinosaurs (which were probably adapted to mild and equable climates) and put a premium on the warm-blooded birds and mammals, which can maintain a constant internal temperature.
Cesare Emiliani, using the Urey technique, studied the shells of foraminifera brought up in cores from the ocean floor, He found that the overall ocean temperature was about 50° F 30 million years ago and 43° F 20 million years ago and is now 35° F (figure 4.8).
Figure 4.8. The record of the ocean temperatures during the last 100 million years.
What caused these long-term changes in temperature? One possible explanation is the so-called greenhouse effect of carbon dioxide. Carbon dioxide absorbs infrared radiation rather strongly. Thus, when there are appreciable amounts of it in the atmosphere, it tends to block the escape of heat at night from the sun-warmed earth. The result is that heat accumulates. On the other hand, when the carbon dioxide content of the atmosphere falls, the earth steadily cools.
If the current concentration of carbon dioxide in the air should double (from 0.03 percent of the air to 0.06 percent), that small change would suffice to raise the earth’s overall temperature by 3 degrees and would bring about the complete and quick melting of the continental glaciers. If the carbon dioxide dropped to half the present amount, the temperature would drop sufficiently to bring the glaciers down to the area of New York City again.
Volcanoes discharge large amounts of carbon dioxide into the air; the t weathering of rocks absorbs carbon dioxide (thus forming limestone). Here, then, is a possible pair of mechanisms for long-term climatic changes. A period of greater-than-normal volcanic action might release a large amount of carbon dioxide into the air and initiate a warming of the earth. Contrariwise, an era of mountain building, exposing large areas of new and unweathered rock to the air, could lower the carbon-dioxide concentration in the atmosphere. The latter process may have happened at the close of the Mesozoic (the age of reptiles) some 80 million years ago, when the long decline in the earth’s temperature began.
Whatever the cause of the ice ages may have been, it seems now that human beings themselves may be changing our future climate. The American physicist Gilbert N. Plass has suggested that we may be seeing the last of the ice ages, because the furnaces of civilization are loading the atmosphere with carbon dioxide. A hundred million chimneys are ceaselessly pouring carbon dioxide into the air; the total amount is about 6 billions tons a year—200 times the quantity coming from volcanoes. Plass pointed out that, since 1900, the carbon-dioxide content of our atmosphere has increased about 10 percent and may increase as much again by the year 2000. This addition to the earth’s “greenhouse” shield against the escape of heat, he calculated, should raise the average temperature by about 1.1° C per century. During the first half of the twentieth century, the average temperature has indeed risen at this rate, according to the available records (mostly in North America and Europe). If the warming continues at the same rate, the continental glaciers may disappear in a century or two.
Investigations during the IGY seemed to show that the glaciers are indeed receding almost everywhere. One of the large glaciers in the Himalayas was reported in 1959 to have receded 700 feet since 1935. Others had retreated 1,000 or even 2,000 feet. Fish adapted to frigid waters are migrating northward, and warm-climate trees are advancing in the same direction. The sea level is rising slightly each year, as would be expected if the glaciers are melting. The sea level is already so high that, at times of violent storms at high tide, the ocean is not far from threatening to flood the New York subway system.
And yet there seems to be a slight downturn in temperature since the early 1940s, so that half the temperature increase between 1880 and 1940 has been wiped out. This change may be due to increasing dust and smog in the air since 1940: particles that cut off sunlight and, in a sense, shade the earth. It would seem that two different types of human atmospheric pollution are currently canceling each other’s effect, at least in this respect and at least temporarily.
Chapter 5
* * *
The Atmosphere
The Shells of Air
Aristotle supposed the world to be made up of four shells, constituting the four elements of matter: earth (the solid ball), water (the ocean), air (the atmosphere), and fire (an invisible outer shell that occasionally became visible in the flashes of lightning). The universe beyond these shells, he said, was composed of an unearthly, perfect fifth element that he called ether (from a Latin derivative, the name became quintessence, which means “fifth element”).
There was no room in this scheme for emptiness: where earth ended, water began; where both ended, air began; where air ended, fire began; and where fire ended, ether began and continued to the end of the universe. “Nature,” said the ancients, “abhors a vacuum” (Latin for “emptiness”).
MEASURING AIR
The suction pump, an early invention to lift water out of wells, seemed admirably to illustrate this abhorrence of a vacuum (figure 5.1). A piston is fitted tightly within a cylinder. When the pump handle is pushed down, the piston is pulled upward, leaving a vacuum in the lower part of the cylinder. But since nature abhors a vacuum, the surrounding water opens a one-way valve at the bottom of the cylinder and rushes into the vacuum. Repeated pumping lifts the water higher and higher in the cylinder, until it pours out of the pump spout.
Figure 5.1. Principle of the water pump. When the handle raises the piston, a partial vacuum is created in the cylinder, and water rises into it through a one-way valve. After repeated pumping, the water level is high enough for the water to flow out of the spout.
According to Aristotelian theory, it should have been possible in this way to raise water to any height. But miners who had to pump water out of the bottoms of mines found that, no matter how hard and long they pumped, they could never lift the water higher than 33 feet above its natural level.
Galileo grew interested in this puzz
le toward the end of his long and inquisitive life. He could come to no conclusion except that apparently nature abhorred a vacuum only up to certain limits. He wondered whether the limit would be lower if he used a liquid denser than water, but he died before he could try this experiment.
Galileo’s students Evangelista Torricelli and Vincenzo Viviani did perform it in 1644. Selecting mercury (which is 13½ times as dense as water), they filled a yard-long glass tube with mercury, stoppered the open end, upended the tube in a dish of mercury, and removed the stopper. The mercury began to run out of the tube into the dish; but, when its level had dropped to 30 inches above the level in the dish, it stopped pouring out of the tube and held at that level.
Thus was constructed the first barometer. Modern mercury barometers are not essentially different. It did not take long to discover that the height of the mercury column was not always the same. The English scientist Robert Hooke pointed out, in the 1660s, that the height of the mercury column decreased before a storm, thus pointing the way to the beginning of scientific weather forecasting or meteorology.
What was holding the mercury up? Viviani suggested that it was the weight of the atmosphere, pressing down on the liquid in the dish. This was a revolutionary thought, for the Aristotelian notion had been that air had no weight, being drawn only to its proper sphere above the earth. Now it became plain that a 3-foot column of water, or a 30-inch column of mercury, measured the weight of the atmosphere—that is, the weight of a column of air of the same cross section from sea level up to as far as the air went.
