by Carl Sagan
CHAPTER 14
THE CLIMATES OF PLANETS
Is it not the height of silent humour
To cause an unknown change
in the earth’s climate?
ROBERT GRAVES,
The Meeting
BETWEEN 30 and 10 million years ago, it is thought, temperatures on Earth slowly declined, by just a few Centigrade degrees. But many plants and animals have their life cycles sensitively attuned to the temperature, and vast forests receded toward more tropical latitudes. The retreat of the forests slowly removed the habitats of small furry binocular creatures, weighing only a few pounds, which had lived out their days brachiating from branch to branch. With the forests gone, only those furry creatures able to survive on the grassy savannas were to be found. Some tens of millions of years later, those creatures left two groups of descendants: one which includes the baboons and the other called humans. We may owe our very existence to climatic changes that on the average amount to only a few degrees. Such changes have brought some species into being and extinguished others. The character of life on our planet has been powerfully influenced by such variations, and it is becoming increasingly clear that the climate is continuing to change today.
There are many indications of past climatic changes. Some methods reach far into the past, others have only a limited applicability. The reliability of the methods also differs. One approach, which may be valid for a million years back in time, is based on the ratio of the isotopes oxygen 18 to oxygen 16 in the carbonates of shells of fossil foraminifera. These shells, belonging to species very similar to some that can be studied today, vary the oxygen 16/oxygen 18 ratio according to the temperature of the water in which they grew. Somewhat similar to the oxygen-isotope method is one based upon the ratio of the isotopes sulfur 34 to sulfur 32. There are other, more direct fossil indicators; for example, the widespread presence of corals, figs and palms denotes high temperatures, and the abundant remains of large hairy beasts, such as mammoths, indicate cold temperatures. The geological record is replete with extensive evidence of glaciation-great moving sheets of ice that leave characteristic boulders and erosional traces. There is also clear geological evidence for beds of evaporites-regions where briny water has evaporated leaving behind the salts. Such evaporation occurs preferentially in warm climates.
When this range of climatic information is put together, a complex pattern of temperature variation emerges. At no time, for example, is the average temperature of the Earth below the freezing point of water, and at no time does it even approach the normal boiling point of water. But variations of several degrees are common, and even variations of twenty or thirty degrees may have occurred at least locally. Fluctuations of a few degrees Centigrade happen over characteristic times of tens of thousands of years, and the recent succession of glacial and interglacial periods has this timing and temperature amplitude. But there are climatic fluctuations over much longer periods, the longest being on the order of a few hundred million years. Warm periods appear to have occurred about 650 million years ago and 270 million years ago. By the standards of past climatic fluctuations, we are now in the midst of an ice age. For most of the Earth’s history, there were no “permanent” ice caps, as in the Arctic and Antarctic today. We have, over the past few hundred years, made a partial emergence from our ice age caused by some as yet unexplained minor climatic variation; and there are certain signs that we may plunge back into the global cold temperatures characteristic of our epoch as seen from the perspective of the immense vistas of geological time. It is a sobering fact that 2 million years ago the site of the city of Chicago was buried under a mile of ice.
What determines the temperature of Earth? As seen from space, it is a rotating blue ball streaked with varying cloud patches, reddish-brown deserts and brilliant white polar caps. The energy for heating the Earth comes almost exclusively from sunlight, the energy conducted up from the hot interior of the Earth amounting to less than one thousandth of one percent of that arriving in the form of visible light from the Sun. But not all the sunlight is absorbed by the Earth. Some is reflected back to space by polar ice, clouds, and the rocks and water on the surface of the Earth. The average reflectivity, or albedo, of the Earth, as measured directly from satellites and indirectly from Earthshine reflected off the dark side of the Moon, is about 35 percent. The 65 percent of sunlight that is absorbed by the Earth heats it to a temperature which can readily be calculated. This temperature is about −18°C, below the freezing point of seawater and some 30°C colder than the measured average temperature of the Earth.
The discrepancy is due to the fact that this calculation neglects the so-called greenhouse effect. Visible light from the Sun enters the Earth’s clear atmosphere and is transmitted through to the surface. The surface, however, in attempting to radiate back into space, is constrained by the laws of physics to do so in the infrared. The atmosphere is not so transparent in the infrared, and at some wavelengths of infrared radiation-such as 6.2 microns or 15 microns-radiation would travel only a few centimeters before being absorbed by atmospheric gases. Since the Earth’s atmosphere is murky and absorbing at many wavelengths in the infrared, the thermal radiation given off by the surface of the Earth is impeded in escaping to space. In order to have a close equality between the radiation received by the Earth from the Sun and the radiation emitted by the Earth to space, the surface temperature of the Earth must then rise. The greenhouse effect is due not to the major atmospheric constituents of the Earth, such as oxygen and nitrogen, but almost exclusively to the minor constituents, especially carbon dioxide and water vapor.
As we have seen, the planet Venus is probably a case where the massive injection of carbon dioxide and smaller amounts of water vapor into a planetary atmosphere has led to such a large greenhouse effect that water cannot be maintained on the surface in the liquid state; hence, the planetary temperature runs away to some extremely high value-in the case of Venus, 480°C.
We have so far been talking about average temperatures. The temperature of the Earth varies from place to place. It is colder at the poles than at the equator because, in general, sunlight falls directly on the equator and obliquely on the poles. The tendency for the temperatures to be very different between equator and poles on Earth is moderated by atmospheric circulation. Hot air rises at the equator and moves at high altitudes to the poles, where it settles and returns to the surface; it then retraces its path, but at low altitudes, from pole back to equator. This general motion-complicated by the rotation of the Earth, its topography and the phase changes of water-is responsible for weather.
The observed average temperature of about 15°C on the Earth today can be explained quite well by the observed intensity of sunlight, global albedo, the tilt of the rotational axis and the greenhouse effect. But all of these parameters can, in principle, vary; and past or future climatic change can be attributed to changes in any of them. In fact, there have been almost a hundred different theories of climatic change on Earth, and even today the subject is hardly marked by unanimity of opinion. This is not because climatologists are by nature ignorant or contentious, but rather because the subject is exceedingly complex.
Both negative and positive feedback mechanisms probably exist. Suppose, for example, there were a decrease of a few degrees in the Earth’s temperature. The amount of water vapor in the atmosphere is determined almost entirely by temperature and declines by snowing out as the temperature declines. Less water in the atmosphere implies a smaller greenhouse effect and a further lowering of the temperature, which may result in even less atmospheric water vapor, and so on. Likewise, a decline in temperature may increase the amount of polar ice, increasing the albedo of the Earth and decreasing the temperature still further. On the other hand, a decline in temperature may decrease the amount of cloudiness, which will decrease the average albedo of the Earth and increase the temperature-perhaps enough to undo the initial temperature decrease. And it has been proposed recently that the biology
of the planet Earth acts as a kind of thermostat to prevent too extreme excursions in temperature which might have deleterious global biological consequences. For example, a decline in temperature may cause an increase of a species of hardy plants that has extensive ground cover and low albedo.
Three of the more fashionable and more interesting theories of climatic change should be mentioned. The first involves a change in celestial mechanical variables: the shape of the Earth’s orbit, the tilt of its axis of rotation, and the precession of that axis all vary over long periods of time because of the interaction of the Earth with other nearby celestial objects. Detailed calculations of the extent of such variations show that they can be responsible for at least a few degrees of temperature variation, and with the possibility of positive feedbacks this might, by itself, be adequate to explain major climatic variations.
A second class of theories involves albedo variations. One of the more striking causes for such variations is the injection into the Earth’s atmosphere of massive amounts of dust-for example, from a volcanic explosion such as Kiakatoa’s in 1883. While there has been some debate on whether such dust heats or cools the Earth, the bulk of present calculations shows that the fine particulates, very slowly falling out of Earth’s stratosphere, increase the Earth’s albedo and therefore cool it. There is recent sedimentological evidence that past epochs of extensive production of volcanic particulates correspond in time to past epochs of glaciation and low temperatures. In addition, episodes of mountain building and the creation of land surface on the Earth increase the global albedo because the land is brighter than the water.
Finally, there is the possibility of variations in the brightness of the Sun. We know-from theories of solar evolution-that over many billions of years the Sun has been getting steadily brighter. This immediately poses a problem for the most ancient climatology of the Earth, because the Sun should have been 30 or 40 percent dimmer some 3 or 4 billion years ago; and this is enough, even with the greenhouse effect, to have resulted in global temperatures well below the freezing point of seawater. Yet there is extensive geological evidence-for example, underwater ripple marks, pillow lavas produced by the quenching of magma in the ocean, and fossil stromatolites produced by oceanic algae-that there was ample water then available. One proposed way out of this quandary is the possibility that there were additional greenhouse gases in the early atmosphere of the Earth-especially ammonia-which produced the required temperature increment. But apart from this very slow evolution of the brightness of the Sun, is it possible that shorter-term fluctuations occur? This is an important and unsolved problem, but recent difficulties in finding neutrinos-which should, according to current theories, be emitted from the interior of the Sun-have led to the suggestion that the Sun is today in an anomalously dim period.
The inability to distinguish between the various alternative models of climatic change might appear to be nothing more than an unusually annoying intellectual problem-except for the fact that there appear to be certain practical and immediate consequences of climatic change. Some evidence on the trend of global temperature seems to show a very slow increase from the beginning of the industrial revolution to about 1940, and an alarmingly steep decline in global temperature thereafter. This pattern has been attributed to the burning of fossil fuels, which has two consequences-the liberation of carbon dioxide, a greenhouse gas, into the atmosphere, and the simultaneous injection into the atmosphere of fine particles, from the incomplete burning of the fuel. The carbon dioxide heats the Earth; the fine particles, through their higher albedo, cool it. It may be that until 1940 the greenhouse effect was winning, and since then the increased albedo is winning.
The ominous possibility that human activities may cause inadvertent climate modification makes the interest in planetary climatology rather important. There are worrisome positive feedback possibilities on a planet with declining temperatures. For example, an increased burning of fossil fuels in a short-term attempt to stay warm can result in more rapid long-term cooling. We live on a planet in which agricultural technology is responsible for the food of more than a billion people. The crops have not been bred for hardiness against climatic variations. Human beings can no longer undertake great migrations in response to climatic change, or at least it is more difficult on a planet controlled by nation-states. It is becoming imperative to understand the causes of climatic variations and to develop the possibility of performing climatic re-engineering of the Earth.
Oddly enough, some of the most interesting hints on the nature of such climatic changes appear to be coming from studies not of the Earth at all, but of Mars. Mariner 9 was injected into Martian orbit on November 14, 1971. It had a useful scientific lifetime of a full terrestrial year and procured 7,200 photographs, covering the planet from pole to pole, as well as tens of thousands of spectra and other scientific information. As we saw earlier, when Mariner 9 arrived at Mars there was virtually no detail whatever to be seen on the surface because the planet was in the throes of a great global dust storm. It was readily observed that the atmospheric temperatures increased, but the surface temperatures decreased during the dust storm, and this simple observation immediately provides at least one clear case of the cooling of a planet by the massive injection of dust into its atmosphere. Calculations have been performed that use precisely the same physics for both the Earth and Mars and treat them as two different examples of the general problem of the climatic effects of massive dust injection into a planetary atmosphere.
There was another and entirely unexpected climatological finding by Mariner 9-the discovery of numerous sinuous channels, replete with tributaries, covering the equatorial and mid-latitudes of Mars. In all cases where relevant data exist, the channels are going in the proper direction-downhill. Some of them show braided patterns, sand bars, slumping of the banks, streamlined teardrop-shaped interior “islands” and other characteristic morphological signs of terrestrial river valleys.
But there is a great problem with the interpretation of the Martian channels as dry riverbeds, or arroyos: liquid water apparently cannot exist on Mars today. The pressures are simply too low. Carbon dioxide on Earth is known as both a solid and a gas, but never as a liquid (except in high-pressure storage tanks). In the same way, water on Mars can exist as a solid (ice or snow) or as vapor, but not as a liquid. For this reason some geologists are reluctant to accept the theory that at one time the channels contained liquid water. Yet they are dead ringers for terrestrial rivers, and at least many of them have forms inconsistent with other possible structures such as collapsed lava tubes, which may be responsible for sinuous valleys on the Moon.
Furthermore, there is an apparent concentration of such channels toward the Martian equator. The one striking fact about the equatorial regions of Mars is that they are the only places on the planet where the average daytime temperature is above the freezing point of water. And no other liquid is simultaneously cosmically abundant, of low viscosity, and with a freezing point below Martian equatorial temperatures.
If, then, the channels were made by running water on Mars, that water apparently must have run at a time when the Martian environment was significantly different from what it is today. Today Mars has a thin atmosphere, low temperatures and no liquid water. At some time in the past, it may have had higher pressures, perhaps somewhat higher temperatures and extensive running water. Such an environment appears to be more hospitable to forms of life based on familiar terrestrial biochemical principles than the present Martian environment.
A detailed study of the possible causes of such major climatic changes on Mars has laid stress on a feedback mechanism known as advective instability. The Martian atmosphere is composed primarily of carbon dioxide. There seem to be large repositories of frozen CO2 in at least one of the two polar caps. The pressure of CO2 in the Martian atmosphere is quite close to the pressure of CO2 expected in equilibrium with frozen carbon dioxide at the temperature of the cold Martian pole. This is a situation quite
similar to the pressure in a laboratory vacuum system determined by the temperature of a “cold finger” in the system. At the present time the Martian atmosphere is so thin that hot air, rising from the equator and settling at the poles, plays a very small role in heating the high latitudes. But let us imagine that the temperature in the polar regions is somehow slightly increased. The total atmospheric pressure increases, the efficiency of heat transport by advection from equator to pole also increases, polar temperatures increase still further, and we see the possibility of a runaway to high temperatures. Likewise a decrease in temperature, from whatever cause, could bring about a runaway toward a lower temperature. The physics of this Martian situation is easier to work out than the comparable case on Earth, because on Earth the major atmospheric constituents, oxygen and nitrogen, are not condensable at the poles.
For a major increase in pressure to occur on Mars, the amount of heat absorbed in the polar regions of the planet must be increased by some 15 or 20 percent for a period of at least a century. Three possible sources of variation in the heating of the cap have been identified, and they are, interestingly enough, very similar to the three fashionable models of terrestrial climatic change discussed above. In the first, variations of the tilt of the Martian rotational axis toward the Sun are invoked. Such variations are much more striking than for the Earth, because Mars is close to Jupiter, the most massive planet in the solar system, and the gravitational perturbations by Jupiter are pronounced. Here variations in global pressure and temperature will occur on hundred thousand to million year time scales.
Secondly, a variation in the albedo of the polar regions can cause major climatic variations. We can already see substantial sand and dust storms on Mars, because of which the polar caps seasonally darken and brighten. There has been one suggestion that the climate of Mars may be made more hospitable if a hardy species of polar plant can be developed that will lower the albedo of the Martian polar regions.