With a few more tools at our disposal, we once more pick up the thread of Earth's climate history, starting with a more detailed look at certain aspects of the Proterozoic - the geological eon that extends from 2.5 billion years ago to 543 million years ago, and is subdivided into the Paleo- Meso-and Neo- Proterozoic, in order of decreasing age. From now on, we will have increasing need to refer to the various subdivisions of the geological time scale by name, so these are summarized in Fig. 1.7.
So far we have not said much about continents and continental drift. Continents consist of light material that has floated to the outer portions of the solid Earth and been incorporated into the crust. This material is too buoyant to be subducted into the interior of the planet to any significant degree, so once continental material had completely segregated, it remained at the surface as a kind of scum resting atop the churning cauldron of Earth's interior fluid motions. The continental material is constantly being pushed around, broken up and re-arranged in a process known as continental drift. The Earth is the only planet in the Solar System that has continents in this sense, but it is likely that extrasolar rocky planets of a sufficient size to retain the heat that drives internal motions, and having a surface temperature similar to Earth, would also exhibit the dichotomy of drifting continents vs. areas of rapidly recycled mantle material (analogous to Earth's sea floors). It is at present unresolved whether a water ocean is necessary to maintaining this state of affairs. That is indeed one of the Big Questions of planetary science, but it is not one which we will take up in this book.
Continents are important to climate for three main reasons: They are a platform upon which polar glaciers can form; They are the primary sites of the silicate weathering reaction that governs atmospheric CO2, and the amount of weathering is strongly affected by the continental configuration; They affect the geometry of ocean basins, and hence the ability of oceans to transport heat from one latitude to another. In addition, they provide distinct habitat for novel forms of life, though they were not colonized to any great degree (if at all) until land plants evolved in the late Ordovician and early Silurian.
There is not enough preserved continental crust to get a clear idea of the distribution of the continents until the very end of the Proterozoic. Geophysical modelling and indirect geochemical evidence has lead to a prevailing belief that the total volume of continental material was similar to the present volume throughout the Proterozoic, but this is not a very well settled area of geophysics. By the close of the Proterozoic, however, the picture of the distribution of continents begins to clarify; we will begin showing maps of paleogeography when we come to discuss that time.
The Proterozoic was the Age of Microbes. Indeed, in terms of the functioning of the biogeo-chemical cycles needed to sustain life, we are all still basically the guests of the prokaryotes, but up until the very end of the Proterozoic single-celled organisms had the world to themselves, with the more complex eukaryotic form of microbial life making its appearance roughly midway through the Proterozoic. The Proterozoic was above all a time of adjustment of biosphere and climate to the massive changes wrought by oxygenation of the atmosphere and ocean. It is a matter of current debate as to whether the oxygenation began in this eon because oxygenic photosynthesis only evolved at this time, or because other factors kept photosynthetic oxygen from accumulating earlier; be that as it may, the oxygenation occurred in fits and starts throughout the Proterozoic, and atmospheric oxygen levels probably reached values comparable to modern ones by the close of the eon. One effect of oxygenation we have already noted is that it is likely to have reduced the importance of CH4 as a greenhouse gas, with CO2 (aided by water vapor) becoming increasingly dominant. More speculatively, oxygenation could have changed the abundance of other greenhouse gases that could conceivably have been significant in the anoxic atmosphere, such as N2O and SO2. The nature of this greenhouse turnover, and the extent to which it constituted a habitability crisis for our planet, is another of the Big Questions. In Chapter 4 we'll learn how to evaluate the relative importance of the various greenhouse gases. Oxygenation would have also sharply limited the possibility for H2 to accumulate in the atmosphere, which would have consequences for methanogenic ecosystems that used H2 and CO2 as a feedstock.
Oxygenic photosynthesis turns CO2 into O2 and organic carbon, represented schematically as CH2 O. In order for the oxygen to accumulate, the organic carbon produced by photosynthesis must be buried before it can be re-oxidized by other bacteria, which would just take the free oxygen back out of the system and turn it back into CO2. For this reason, the evolution of oxygen is intimately tied in with the carbon cycle. Since organic carbon is isotopically light compared to the carbon in CO2 outgassed from the interior of the Earth, the long term evolution of 513C in carbonates gives us a window into the carbon cycle, and a good overview of what is going on in the Proterozoic. As noted earlier, when the carbon cycle is approximately in a steady state, the 513C of carbonate is driven more positive as the proportion of organic carbon burial to total carbon burial increases. One must be cautious about this interpretation, however, since the carbon cycle is thrown wildly out of equilibrium in the course of the Snowball episodes that constitute the most dramatic features of Proterozoic climate.
With regard to oxygenation, however, organic carbon burial is not the whole story. Various compounds of sulfur also play a key role in the cycling of oxygen through the Earth system; there is evidence that the role of the sulfur cycle is much more prominent in the Proterozoic than during later times. Bacteria are clever, and have ways of oxidizing organic matter that don't use up free O2. In a process called sulfate reduction, certain bacteria can react the sulfate ion (SO,- ) with organic carbon to produce bicarbonate (HCO—) and the stinky rotten-egg gas hydrogen sulfide (H2S), which reacts with iron oxides and a little bit of free oxygen to produce water and the mineral pyrite (FeS2). If the pyrite is then buried without being oxidized further, the net process turns organic matter into mineral carbonate while leaving much of the O2 liberated by oxygenic photosynthesis in the atmosphere/ocean system. Precisely how much is left in the atmosphere and ocean depends on where the oxygen in sulfate and iron oxides comes from, but the net result is that pyrite burial liberates oxygen in a way that doesn't show up in the carbon isotope record.
The participation of sulfur in a variety of reactions involving oxygen makes the stable isotopes of sulfur - 32S, 33S, 34S and 36S, of which the first is dominant - a key source of information about past behavior of oxygen. By comparing the degree of fractionation for the three minor isotopes, one can get additional information. One form of fractionation is mass-independent fractionation (MIF), which is nearly the same for all three minor isotopes. This kind of fractionation is believed to be produced only in photochemical reactions involving high energy ultraviolet light, and photochemical models of the atmosphere indicate that mass-independent sulfur fractionation can't be preserved in the sedimentary record unless the atmospheric oxygen concentration is extremely low - 10-5 of the present concentration or less, according to current estimates. It is the sulfur MIF proxy that tells us that Archaean oxygen levels are nearly zero; other oxygen proxies only require that Archaean oxygen be below 1% of present atmospheric concentration 7.
The more conventional mass-dependent sulfur fractionation is mediated by sulfate-reducing bacteria. The fractionation nearly disappears when sulfate concentration in ocean water is low, so a strong mass-dependent sulfur fractionation indicates both high sulfate concentration and strong productivity of sulfate-reducing bacteria. An increase in sulfate concentration, in turn, is generally taken as indicative of a rise in atmospheric oxygen, since that permits more oxidation of pyrite into sulfate on land. Once oxygen builds up to the point that at least near-shore bottom waters become oxygenated, a host of additional bacterially-mediated sulfur reactions, called disproportionation reactions become possible, and these provide additional means of producing sedimentary sulfides (e.g. pyrite) that are isotopically light in sulfur. The interpretation of mass-dependent sulfur isotope fractionation is an exceedingly complex subject, which is likely to remain in a considerable state of flux for some time to come.
The proxy record shows a great deal of activity towards the beginning of the Proterozoic (during the Paleoproterozoic), and also towards the end of the Proterozoic (in the later parts of the Neoproterozoic). In between lies a billion-year period that has sometimes been called "the most boring period in Earth history." During this Big Yawn, which stretched from about 1.8 billion years ago to 800 million years ago, carbonate S13C held steady near Woo, indicating steady organic carbon burial. Mass dependent sulfur isotope fractionation suggests oxygen levels of around 10% modern concentrations during this period, though the upper bound on oxygen during this period is not well constrained. There is no indication of any significant glaciation. Eukaryotes arose towards the beginning of the Big Yawn, but appeared to have little effect on biogeochemical cycling or climate
7The atmospheric chemistry models upon which this interpretation of the sulfur MIF is based rest on somewhat shaky assumptions, however.
evolution - unless perhaps they were somehow the cause of the long period of climate stability. It would not be surprising if closer study eventually revealed more features of interest in this period, but at this point we'll turn our attention to the more manifestly dramatic doings at the beginning and end of the Proterozoic.
All lines of evidence point to a Great Oxidation Event at the beginning of the Proterozoic. Preservation of mass-independent sulfur fractionation in sediments ceases abruptly between 2.45 and 2.3 billion years ago, and is never seen again throughout the rest of Earth history. A hiatus in banded iron formations begins at about this time. It is estimated that the oxygen content of the atmosphere soared to values well in excess of 1% of the present level but crashed back to lower levels afterwards, as witnessed by a transient reappearance of banded iron formations; the peak value in the event is at present not well constrained. A subsequent increase in mass-dependent sulfur isotope fractionation preserved in the sediments is indicative of an increase in sulfate concentration in the ocean, most likely associated with an increase of oxidation of pyrite on land. Based on this evidence and disappearance of banded iron formations, it is estimated that oxygen levels in the atmosphere recovered to somewhere around 10% of present atmospheric concentrations around 1.7 billion years ago, and stayed there until 700 million years ago when there was a further oxygenation event.
The Paleoproterozoic is characterized by wild swings in the 613C of carbonates. Around the time of the Great Oxygenation event, 613C has a major positive excursion, reaching values as high as 10 before eventuallly subsiding to the lower values characteristic of the middle Proterozoic. This indicates a major transition in the carbon cycle, most likely an increase in the proportion of organic carbon burial. It is very suggestive of a take-off of oxygenic photosynthesis, but whether the cause is evolutionary, ecological or a matter of factors that allow better burial of carbon is a matter of dispute. There are several major glaciations within the Paleoproterozoic, of which one - the Makganyene alluded to earlier - was a Snowball event in which ice reached tropical latitudes. The Makganyene Snowball occurred within the interval between 2.32 and 2.22 billion years ago, and fine-scale examination of arguably synchronous glacial deposits in the Duitschland formation (Transvaal, S. Africa) indicates extreme carbon isotope excursions associated with major Paleoproterozoic glaciations: Carbonate 613C was around 5 %obefore the glaciation, then dropped to zero or even negative values as the glaciation progressed, recovering slowly afterwards. We'll be able to probe similar features in more detail in connection with the Neoproterozoic Snowballs. The Paleoproterozoic presents us with a puzzle whose pieces include oxygen, the effect of oxygen on greenhouse gases, the carbon cycle, and glaciation. Figuring out how these pieces fit together is one of the Big Questions.
The Neoproterozoic has many features in common with the Paleoproterozoic. The extreme carbonate carbon isotope excursions which had been dormant for so long resume in the Neopro-terozoic. There are several major glaciations during the Neoproterozoic, and two of these were Snowball events in which ice reached tropical latitudes. The more recent of the two Snowballs is the Marinoan event, which occurred about 640 million years ago; the older is the Sturtian, centered on 710 million years ago. Neoproterozoic Snowball-related geological formations exhibit a distinctive sequence of events. The scene starts with high carbonate 613C, up to 5%o, which is in fact higher than the modern value and indicative of a greater proportion of organic carbon burial than is the case at present. Then, the 613C drops, falling to zero or even negative values. At some point in this decline, one sees diamictites and other glacial deposits. The 613C continues to drop, and above the glacial deposits one finds cap carbonates - very unusual carbonate features that are believed to require very high deposition rates from waters highly supersaturated in carbonate. In the carbonates overlying the glacial deposit, the 613C becomes negative, typically around -5%owhich is about the value for abiotic carbon outgassing from the Earth's interior. The 613C gradually recovers to positive values over a long (but somewhat unconstrained) period of time afterwards. A particularly clear depiction of this sequence of events is given in the review by Hoffman and Schrag listed in the Further Readings for this chapter.
However, not all major carbon isotope excursions are associated with Snowball events. In fact, the greatest carbon isotope excursion in Earth history - the Shuram excursion - sets in gradually after a conventional glaciation which is thought to reach only to midlatitudes (the Gaskiers glaciation). The Shuram excursion brings the carbonate 513C all the way down to -12 %o. There is no known process that could bring the carbonate 513C so far below the mantle outgassing value if the carbon cycle is in equilibrium. Indeed, the 513C is so implausibly low in the Shuram that it was long thought to be an artifact of diagenetic alteration. The Shuram is an enigmatic event - indeed one of the Big Questions. Current thinking has it that the Shuram is associated with a transient reorganization of the carbon cycle, in which a large isotopically light pool of suspended organic carbon in the ocean is oxidized and deposited as carbonate.
In fact, a lot is going on with oxygen across the Neoproterozoic, though it is a bit hard to determine what, where and when. What is certain is that oxygen must have been high - even near present levels - right down to the ocean bottom by the end of the Neoproterozoic, since bottom-dwelling animals appear in the fossil record by this time, and it is unquestionable that such creatures require a great deal of oxygen. At the other side of the Neoproterozoic, around 700 million years ago, there is further evidence of oxygenation, in that a sharp rise in mass dependent fractionation in sedimentary sulfides indicates the expression of sulfur disproportionation reactions, which indicates an oxygenation of at least some of the bottom waters. Another important clue as to what is going on is the reappearance of banded iron formations in connection with the Neoproterozoic Snowball events, suggesting that the ocean once more went anoxic, most plausibly as a result of global ice cover shutting down photosynthesis.
A very Big Question is why all this excitement suddenly resumed after nearly a billion years of stasis.
The Snowball events of the early and late Proterozoic are some of the most dramatic events of Earth history. We have used the term "Snowball" to refer to any glaciation where there is evidence of glaciation at tropical latitudes, but it is a matter of considerable debate whether the oceans were indeed nearly completely frozen over all the way to the equator during these events. Sometimes the term Hard Snowball is used to refer specifically to a state with near-total ice cover. A Big Question of climate physics is whether it is possible to cool down the planet enough to yield land-based ice sheets discharging into the tropics, without also freezing over the tropical ocean completely. This requres an understanding of ice-albedo feedback, which will be developed at several places throughout the book. However, it also involves ocean heat transports (which are good at melting ice) and glacier dynamics, which for the most part are subjects that will be left for another time and place.
The Snowball phenomenon is pregnant with Big Questions, the most obvious of which are: How do you get in? And how do you get out? And if your planet does succumb to a global Snowball, how long does it take to get out again? Is it a matter of centuries, millions of years or billions of years? On Earth, the upper limit set by the geological record for the duration of Neoproterozoic snowballs is about 20 million years, and the duration could well have been shorter. However, without a clear understanding of the nature of the event, it is hard to determine whether we just got lucky or whether the event could have lasted much longer.
Most theories for the entry into a Snowball involve the drawdown of whatever greenhouse gas had previously been maintaining the planet's warmth - usually CO2 and CH4 in some combination. Various hypotheses include methane destruction by oxygen, weathering enhancement triggered by catastrophic methane release from sediments, weathering enhancement due to continental configuration or production of weatherable rock by massive volcanic eruptions, and (more speculatively) drawdown of CO2 through runaway photosynthesis and oxygenation. Whatever the mechanism, a key requirement is that the mechanism be compatible with the observed reduction in ó13C before the onset of glaciation. Not all of the relevant biogeochemistry will be treated in this book, but in order to evaluate the hypotheses, it is certainly necessary that one have the tools to assess how low any given greenhouse gas has to go in order to trigger a global glaciation. These tools will be provided in subsequent chapters.
Assuming for the moment that the cooling process caused a Hard Snowball, the next question is how to deglaciate the planet. We'll see in Chapter 3 that one would have to wait a billion years or more to exit from a Snowball if the exit were due to increase in solar output alone. Based on rather simple reasoning of the sort that will be covered in the remainder of this book, Kirschvink proposed that once the Earth freezes over, the weathering of silicate to carbonate (which requires liquid water washing over weatherable rocks) ceases, so that CO2 outgassed from the Earth's interior accumulates in the atmosphere until it reaches concentrations sufficient to cause a deglaciation. This is another illustration of the principle that Big Ideas come from Simple Models. A Big Question (treated in subsequent chapters) is: how high does CO2 have to go in order to trigger deglaciation of a globally ice-covered planet? Much hangs on the answer.
Both cap carbonates and the persistent negative carbon isotope excursion following the Snowball events are consistent with a massive buildup of CO2 in the atmosphere during the frozen-over period. Once the planet gets warm enough to deglaciate, the powerful precipitation in the ensuing hothouse world would wash great quantities of land carbonates into the ocean, where they would precipitate to form cap carbonates. Further, if photosynthesis nearly shuts off during the glaciated phase, the inorganic carbon that accumulates in the ocean-atmosphere reservoir would have ó13C comparable to the mantle outgassing value of about -6%o. As this reservoir is gradually transformed by silicate weathering into carbonate sediments, the isotopically light carbon works its way into the carbonates. If the reservoir is big enough at the termination of the snowball, it can keep the sedimentary carbonate ó13C light even in the face of a resumption of photosynthesis. When interpreting the carbon isotope excursions in the course of a Snowball, it is essential to keep in mind that the carbon cycle is likely to be far out of equilibrium in the course of these events. A full understanding of the connection between the carbon isotope evolution and the sequence of events surrounding the Snowball requires a detailed accounting of flows of carbon between the many different carbon reservoirs in the Earth system - land carbonate rocks, marine carbonate sediments, atmospheric carbon dioxide, various species of dissolved inorganic carbon, and organic carbon.
Assuming that the the exit from a Snowballl state does indeed proceed from accumulation of a great deal of CO2 in the atmosphere, several Big Questions arise in connection with the post-glacial climate. Is there a risk of triggering a runaway greenhouse? If not, how hot does the climate get in the tropics and polar regions? Would it be hot enough to sterilize the planet to any great degree? How long is the post-snowball recovery? In other words, how long does it take for weathering processes to draw the CO2 back down to more normal levels?
Other Big Questions include: Why were there no Snowball event during the Big Yawn period of the middlle Proterozoic? Why did Snowballl events cease at the beginning of the Phanerozoic? Could they happen again, or is this particular threat behind us?
But let us not forget that another Big Question is whether there is a climate state with significant amounts of open water in the Tropics, which is nevertheless consistent with the full range of geological data accompanying the Marinoan and Sturtian events. One could pose the same question for the Makganyene event, but there is less data to constrain the answer. The early and late Proterozoic manifests a lot of climate "weirdness" that is not seen elsewhere in Earth history, and such striking signatures would seem to call for an equally dramatic cause, rather than just a minor variation on the theme of ice ages. The global Snowball seems to fit the bill, but it remains to be seen whether other explanations are possible. The climate physics developed throughout this book will give the reader the underpinning needed to assess hypotheses as they develop, and even perhaps to formulate new ones.
Regardless of whether a true global Snowball glaciation ever happened on Earth, the Snowball certainly represents a state a water-covered planet could fall into if the right stellar and atmospheric conditions are encountered. Once a planet falls into a Snowball state,it is likely to stay there for a long time, and the consequences for existing life and evolution of new forms of life are profound. As such, Snowball states are a potential habitability crisis that extrasolar planets need to avoid or surmount. It is therefore worth understanding, in general terms, the physics of entry into, exit from, and duration of Snowball states, as well as the nature of the climate at various stages of the sequence and the effect of the sequence on life. This constitutes another Big Question, about which will have much to say.
Was this article helpful?