The Solar system was not always as we see it today. It formed from a nebula of material collapsing under the influence of its own gravitation, and once the nebula began to collapse, things happened very quickly. The initial stage of formation of the Solar system was complete by about 4.6 billion years ago. By this time, the Sun had begun producing energy by thermonuclear fusion; the formation of the outer gas giant planets and their icy moons by condensation, and the formation of the inner planets by collision of smaller rocky planetesimals, were essentially complete. The last major event in the formation of the Earth was collision with a Mars-sized body 4.5 billion years ago, which formed the Moon and may have melted the Earth's primitive crust in the process. All these collisions left behind a great deal of heat that had to be gotten rid of before the crust could stabilize. To determine how long it takes to get rid of this heat, we must learn about the mechanisms by which planets lose energy, and about how the rate of energy loss depends on temperature and atmospheric composition; this will happen in Chapters 3 and 4. It turns out that a planet loses energy almost exclusively by radiation of infrared light to space. While the precise rate of loss depends on the nature of the atmosphere, all estimates show that the surface of the Earth quickly cools to 2000K, at which point molten rock solidifies; in the absence of an atmosphere, this process takes a thousand years or less, while with a thick atmosphere it could take as long as two or three million years.
Once a solid crust forms, the flow of heat from the interior of the Earth to the surface is sharply curtailed, because heat diffuses very slowly through solid rock. In this situation, supply of heat from the interior becomes insignificant in comparison with the energy received from the Sun, and the Earth has settled into a state where the climate is determined by much the same processes that determine today's climate: a competition between the rate at which energy is received from the Sun against the rate at which energy is lost to space by radiation of infrared light. This is very likely to have been the case by 4.4 billion years ago, if not earlier. There are no actual rocks as old as this, but there are individual zircon crystals embedded in the Jack Hills formation of Western Australia which are 4.4 billion years old. Zircons of a similar age are also found within the 3.7 billion year old crustal rocks of the neighboring Narryer Gneiss Complex. These crystals provide indisputable evidence for the existence of at least some continental crust of a sort very like that we see today; they also provide convincing though less certain evidence of the existence of liquid water in contact with the early continental crust. The existence of liquid water does not in itself put much constraint on temperature, since water can be maintained in a liquid state even at temperatures in excess of 500 degrees Kelvin, provided the pressure exerted by water vapor in the atmosphere is high enough. The thermodynamics needed to address this issue will be introduced in Chapter 2. Certain aspects of the chemical composition of the zircons, however, suggest that they interacted with near-surface water having a temperature of 100C or less. By 4.4 billion years ago, it would appear, the Earth was no longer a molten volcanic inferno.
The precise nature of the climate evolution between 4.5 and 3.8 billion years ago is obscure at present. Depending on the composition of the atmosphere, the surface temperature could have been as high as 200C or low enough to cause the ocean (if any) to freeze over completely, and the climate could well have swung wildly between the two extremes. In addition, the dates of Lunar craters indicate that the Earth very likely underwent a period of heavy bombardment by interplanetary debris between 4.1 and 3.8 billion years ago; it is generally supposed that this late heavy bombardment affected the rest of the inner Solar system as well, though that is far from certain. The energy brought in by impacts during this period could easily have been sufficient to bring surface temperatures episodically to values well in excess of 100C, sterilizing any nascent ecosystems. Life, if any, may have waged and won a battle for survival in deep ocean refugia.
By 3.8 billion years ago, the veil begins to lift. This is the age of the oldest intact rocks, found today in the Isua Greenstone Belt of Greenland. The appearance of these rocks marks the end of the Hadean eon, and the dawn of the Archaean eon. Remnants of 3.7 billion year old shales in the Isua formation show the unmistakable signs of deposition of sediments in open water. More intriguingly, these shales are rich in organic carbon, and this carbon preserves a chemical signature generally associated with microbial activity - life. The Barberton formation of South Africa and the Warrawoona formation of Australia, both about 3.5 billion years old, contain layered carbonate sedimentary structures known as stromatolites, which in later times are known to be laid down by microbial mats. This is not an unambiguous sign of life, since inorganic processes can also produce stromatolite-like features. Be that as it may, the early stromatolites certainly require ponds of open water evaporating into air. The Barberton and Warrawoona formations also contain microscopic features that are suggestive of bacterial fossils, though not unambiguously so.
The record of surface conditions during the subsequent billion years is hardly continuous, but preserved rocks dating to this period very commonly show a sedimentary character of a type most easily explained by deposition in an open, unfrozen ocean. The first truly unmistakable microbial fossils date to 2.6 billion years ago, where they are found in the Campbell formation of Cape Province, South Africa, and argue for open water conditions having a moderate temperature. At about this time, we bid farewell to the Archean eon, and enter the Proterozoic eon, which extends to the appearance of animal life 544 million years ago. Certain fine-grained silica based sedimentary rocks known as cherts preserve information about past temperatures, as well as a wealth of fossils. Very ancient cherts contain no unambiguous microbial fossils, but certain aspects of their chemical composition point to temperatures as high as 70C at 3.5 billion years ago, declining to 60C at 2 billion years ago, and declining further to 30C at 1 billion years ago. Well-preserved ancient cherts are rare, however, so this data by no means implies that temperatures were uniformly warm on the young Earth. It only indicates that the Earth attained high surface temperatures at least part of the time; there is ample room to hide lengthy cold periods within the gaps in the chert record, as we shall soon see.
The earliest geological indication of the presence of glaciers on Earth occurs in the upper part of the Pongola formation of South Africa, and dates to 2.9 billion years ago. The evidence consists of glacial sedimentary deposits called diamictites, material of a sort usually transported by floating ice, and even glacier-scratched rocks. This does not mean that there were no earlier glaciations, but in light of the chert record and widespread occurence of marine sedimentary rock it seems fairly certain that the Earth did not spend the bulk of its earlier history locked in a deep-freeze. Still, the Pongola glaciation seems to mark the beginning of Earth's long flirtation with ice. The Makganyene glaciation begining around 2.3 billion years ago, recorded in rocks of the Transvaal group of Southern Africa, was a big one, and may well have been global. We know this because a record of the Earth's magnetic field is preserved in the rocks, and this can be used to infer the latitude at which the rocks were located when the glacial deposits were laid down. This paleomagnetic data shows that there was ice within 12 degrees of the Equator, strongly suggestive of a global glaciation.
The first unambiguous bacterial microfossils (found in the Campbell group of South Africa) date to 2.6 billion years ago, shortly before the Makganyene glaciation. While earlier fossil and geochemical evidence is very strongly suggestive of life, the Campbell group fossils are the ocular proof that biology was well underway. These fossils mark a watershed in another important way, in that they are identifiable as cyanobacteria - the type of organisms that produce oxygen by photosynthesis. The issue of when cyanobacteria evolved is hotly debated, with some lines of indirect evidence putting their appearance early in the Archaean and others dating their onset to the time of the Campbell Group microfossils. Be that as it may, the appearance of these fossils speaks for a fairly benign environment, with open water and temperatures no more than about 40C. After the Makganyene glaciation, microbial fossils become quite abundant. The two billion year old Gunflint Chert of Canada is one of many such marine sedimentary formations in which cyanobacterial microfossils are preserved.
So far, no glaciations have been reported in the period between two billion years ago and 800 million years ago, though there are abundant sedimentary rocks dating to this time. The record is far from continuous and the lack of glaciations in this period may be an artifact of preservation, but the evidence certainly indicates that icy climates were not dominant at this time, and were probably quite rare. The long hiatus in ice is terminated by the massive - and possibly global
- Snowball Earth glaciations of the Neoproterozoic, about 700 million years ago. Thereafter, the climate alternated between fairly lengthy periods when the Earth was ice-free or nearly so, and periods when there was at least some ice in polar regions. The ice never again, however, reached the nearly global proportions it attained during the Neoproterozoic suggesting that the Earth passed some new threshold of climate stability in the Neoproterozoic. What might that be? This is one of the central questions of climate science.
Our overall picture of Earth history is that liquid water and moderate temperatures appeared at least episodically very shortly after the Moon-forming collision, and that the next three billion years had widespread open water with temperatures probably not exceeding 70C and generally much less. These conditions were punctuated by occasional glaciations, only a very few of which may have been global in extent. It was an environment that could support the evolution and survival of life, including (by 2.6 billion years ago, if not before) photosynthetic life requiring moderate temperature conditions. Let's keep this picture of relative stability in mind as we go on to discuss long-term changes in the atmosphere and the Sun - the two principle ingredients determining the Earth's climate, or indeed that of any planet.
There are many processes at play that cause the composition of a planet's atmosphere to evolve over time. In the earliest times, bombardments can help supply atmosphere-forming volatiles such as water, nitrogen and carbon dioxide. Equally, however, sufficiently energetic bombardments can cause loss of atmosphere through literally splashing it into orbit. On a volcanically active planet with a hot interior, such as the Earth or Venus, or the younger Mars, new atmosphere is continually being supplied by outgassing of volatile substances from the interior of the planet. The heat needed to keep the interior of the planet churning so it can recycle minerals formed at the surface and cook out volatile gases in the hot interior is supplied by leftover heat from formation of the planet and by radioactive decay. How long this process can continue before the planet freezes out becomes tectonically inactive depends on the size of the planet and the stuff it is made of; the nature of the gases which come out of volcanoes and other types of vents depends on the chemistry of the planet. For example, the early segregation of iron in the Earth's core made it harder to bind up oxygen in minerals, and therefore resulted in fairly oxidized gases like carbon dioxide (CO2), and sulfur dioxide (SO2) being released in preference to gases like methane (CH4) and hydrogen (H2)
- though some of the latter two do nonetheless escape. The interior Earth also outgasses water vapor (H2O), which is cooked out of hydrated minerals; the volume of the oceans appears to have been in a steady state for a long time, though, indicating that the rate of release is balanced by the rate of formation and subduction of new hydrated minerals at the surface. Nitrogen (N2) is fundamentally different from other current and past constituents of the Earth's atmosphere as it doesn't readily get incorporated into the minerals that form the bulk of Earth's crust and interior. Unlike, say, CO2, nitrogen does not cycle through the Earth's interior. The bulk of the Earth's N2 is in its atmosphere and stays there, where it has probably been for almost all of our planet's history. This is likely to be the case as well for any other rocky planet made of stuff similar to the Earth - Iron, oxygen (mostly bound up in minerals), silicon, magnesium and sulfur.
While atmosphere is being supplied by outgassing from the interior, other processes cause material to be lost from the atmosphere. Parts of a planet's atmosphere extend far out from the surface, where hot, fast-moving molecules can reach escape velocity and escape to space. Besides escape from random molecular motions, the solar-heated tenuous outer atmosphere can sustain fluid flows which cause atmospheric mass to fountain systematically into the void. In addition, the solar wind can literally blow way the outer portions of an atmosphere; the extent to which this happens is affected by the intensity of the planet's magnetic field, which shields the atmosphere from the solar wind. As outer parts of the atmosphere are eroded, new gases from lower altitudes well up to replace the lost material, sustaining the gradual loss of atmosphere. All three mass loss processes preferentially remove lighter molecules, either because lighter molecules move more swiftly for a given temperature, or because the outer atmosphere is enriched in gases having a lower molecular weight. For a given density, a smaller planet has lower surface gravity, and so binds its atmosphere less tightly; in consequence, escape of atmosphere to space proceeds more rapidly on a small planet. Impacts by large, swift bodies can impart sufficient energy to part of the atmosphere to blast it into space. This mechanism of atmosphere loss does not discriminate as to molecular weight, but as with the other mechanisms, it is easier for a small planet to lose atmosphere this way. Overall, the Moon or Mars is more prone to lose atmosphere than more massive bodies such as the Earth or Venus, to say nothing of Jupiter or Saturn. For Earth and Venus, escape to space is significant only for H2 and He, and of these the latter is important only as an indicator of planetary history rather than as a physically or chemically active component of the atmosphere. Saturn's satellite Titan is an interesting case, as it maintains a mostly N2 atmosphere more massive than that of Earth (per unit surface area) despite having a surface gravity lower than that of the Moon. The very cold temperature of Titan helps it retain its atmosphere, but it is nontheless likely that the persistence of the atmosphere requires a substantial rate of resupply from the interior of the planet.
Some components of the atmosphere can also be lost through chemical reactions with rocks at the Earth's surface. A particularly important example of this is the class of reactions commonly known today as Urey reactions 3 , which remove CO2 from the atmosphere. When CO2 dissolves in water, it forms a weak acid (carbonic acid), which reacts with silicate minerals (e.g. CaSiO3) to form carbonate minerals (e.g. CaCO3, or "limestone"). The reactions that form carbonate take place only in the presence of liquid water, so if a planet becomes so hot that liquid rain never reaches the surface, or if it somehow loses its water altogether, then CO2 outgassed from the interior of the planet will accumulate in the atmosphere until the interior source is depleted or the rate of supply is balanced by loss to space. On Earth, all of the CO2 presently in the atmosphere
3The reactions are named after the University of Chicago geochemist Harold Urey, who discussed them in a 1952 book called The Planets: Their Origin and Development. Although modern science was made aware of the importance of these reactions through Urey's work, the reactions were first introduced by the French chemist and metallurgist J.J. Ebelman more than a century earlier. Ebelman also introduced the notion that the silicate/carbonate reactions play an important role in determining atmospheric CO2 and hence Earth's climate. Similar ideas were independently rediscovered by the Swedish geochemist A.G Hogbom in 1894, and then finally by Urey. For more details of the history see Berner (1996) Geochim Cosmochim Acta 60.
could be removed by the Urey reactions within 5000 years, forming a layer of limestone a mere 5 millimeters thick; if all the carbon stored in ocean water were to outgas as CO2 and react to form limestone, the process would take a half million years and form a layer a half meter thick.
Life itself, once it appears, has a profound effect on atmospheric composition. While little methane escapes directly from the Earth's interior, bacteria known as methanogens can synthesize it from H2 and CO2 or from organic material produced by other organisms. Methanogens may well have dominated the ecosystems of the Earth's first two billion years, potentially allowing a methane-dominated atmosphere to build up. The advent of life also had a profound effect on nitrogen cycling. The bonds holding together N2 are so strong that in the abiotic world only rare energetic events such as lightning strikes can form nitrogen compounds. In fact, though nitrogen is an essential ingredient of all living material, higher forms of life - including all plants and animals - are incapable of performing the chemical magic that makes N2 available to organisms. This trick is accomplished by nitrogen-fixing bacteria, which can efficiently transform atmospheric N2 into ammonium (NH4), in turn transformed into nitrate (NO3) which can be used by other organisms in the synthesis of living matter. Other bacteria, in oxygen-starved conditions, can make a living combining the oxygen in nitrate with carbon, returning N2 to the atmosphere in the process. It was only in 1910, with the invention of the Haber Process for turning atmospheric N2 into ammonia (NH3) that humans caught up with the bacteria. This innovation has become essential to the human population, as the demands of industrial-style agriculture have far outstripped the ability of the natural bacterial ecosystem to supply nitrate (which is not to deny that other forms of agriculture might be able to live within the means provided by our bacterial friends). Nitrogen-fixing bacteria are still way ahead of industry in terms of chemical sophistication, though, since the Haber Process requires molecular hydrogen (made from fossil methane), an iron catalyst, temperatures exceeding 400C, and operates at a pressure over two hundred times that of air at the Earth's surface; Nitrogen-fixing bacteria can do the same trick, in contrast, in ambient temperature/pressure conditions and using materials found readily in their immediate environment.
In the absence of oxygen-producing photosynthetic life, only minuscule amounts of oxygen would be present in the atmosphere, since only a trickle could be produced through the breakdown of H2O by exposure to sunlight. That there was very little oxygen in the early atmosphere (under 0.2%, compared to 20% today) is confirmed by the widespread presence of striking rock structures known as banded iron formations until 2.4 billion years ago. Banded iron formations can be laid down only when iron is very soluble in the ocean and can be transported long distances. This requires low oxygen, since in the presence of oxygen iron forms compounds that are not very soluble in water. Additional evidence stemming from the chemical composition of certain sulfur-containing minerals indicates that during at least some periods earlier than 2.6 billion years ago the atmospheric oxygen content might in fact have been orders of magnitude lower than 0.2%.
Note that the appearance of oxygen-producing photosynthesis is not synonymous with the rise of oxygen in the atmosphere. For oxygen to accumulate, a sufficient proportion of the organic matter must be buried before it is oxidized by other bacteria, taking the synthesized oxygen right back out of the atmosphere. Further, if the Earth had accumulated a great stock of available organic matter in the ocean during its anoxic phase, this backlog would have to be worked off before oxygen could begin to accumulate to any significant degree. Be that as it may, banded iron formations begin to falter somewhat after the time of the Campbell Group cyanobacterial microfossils, and disapear entirely by around 2 billion years ago. By this time, oxygen may have made up at least 3% of the atmosphere. Once oxygen made its appearance in the atmosphere at significant concentration, it changed all the rules of atmospheric chemistry, since it is so powerfully reactive. In particular, it made it much harder for CH4 and H2 to accumulate in the atmosphere, since the former oxidizes readily to CO2 and the later to H2 O. The rise of oxygen may also have fostered another great biological innovation - the eukaryotic cell, which has a complex internal organization with specialized structures, including a nucleus within which genetic material is segregated. We are made of eukaryotic cells, as are all animals and plants. Eukaryotic cells make their first unambiguous appearance in the fossil record about 1.5 billion years ago, in the Roper Group shales of Australia, though more indirect evidence suggests that eukaryotic life may have evolved much earlier. However the answer to that issue may shake out, it is certain that eukaryotic life - even of the single-celled variety - did not proliferate and diversify until much later in the Proterozoic.
There was a sporadic reappearince of banded iron formations at the time of the Neopro-terozoic Snowball events (600-700 million years ago), but by 600 million years ago oxygen was approaching its present concentration and banded iron formations disappeared for good (at least so far). This second pulse of oxygenation made the rise of multicellular animals possible, which happened in a great flurry of biological innovation known as the Cambrian Explosion, occupying a remarkably short span of years around 543 million years ago. It is even possible that the rise of animals helped to stabilize atmospheric oxygen levels, by providing a reliable means of transporting organic carbon to the sea floor where it can be buried and preserved. Simple multicellular ocean-dwelling plants appeared much earlier, as might be expected from the fact that photosynthetic organisms are not dependent on oxygen.
All of this atmospheric evolution takes place against a backdrop of a gradually brightening Sun. The energy produced within a star leaves the star almost exclusively in the form of electromagnetic radiation - loosely speaking, light of all wavelengths. The net power output is called the luminosity, and is measured in Watts (a measure of energy per unit time), just as if the star were a light bulb. Stars like the Sun get their energy by fusing hydrogen into helium, and as time goes on the proportion of helium in the Sun increases, thus increasing the mean molecular weight of the Sun. This in turn means that the core of the Sun needs to contract and heat up in order to maintain the pressure required to balance gravity. The increased density and temperature increases the rate of fusion more than the reduced availability of hydrogen reduces it, so the rate of production of energy - and hence the Solar luminosity - increases with time. The resulting evolution of luminosity over time rests on fundamental aspects of solar physics that are not seriously in question, and does not depend greatly on the finer points of solar modeling. The results of the standard solar evolution model can be fit by the expression
Was this article helpful?