14.7.1 Basic carbon–silicate weathering cycle
How did a thick carbon dioxide atmosphere dwindle away over time? Carbon dioxide is taken out of the atmosphere by weath- ering of rock, and by plants and bacteria during photosynthe- sis. Much of the biologically trapped carbon continues to cycle through living organisms at the surface. However, some of the carbon dioxide ends up in shell-forming organisms, which die and drop to the ocean floor, effectively removing the carbon from circulation in the biosphere. The chemistry of the weath- ering process, which depends critically on rainwater, goes as follows.
The breakdown of silicate rocks by the weathering action of rainwater is very efficient, because CO2 gas dissolves in the rainwater to make a weak acid that can attack the rock chemi- cally. This yields, among other products, silicon dioxide (SiO2, the basic molecule of which quartz is made), hydrogen carbon- ate ions (HCO3−, where the negative sign indicates that the ion is negatively charged), and doubly charged ions of calcium (Ca2+). These ions are quite reactive, and are used by shell- forming organisms to make calcium carbonate (CaCO3) shells.
The shells, along with silica (opal) shells made by other organ- isms, are preserved as thick layers of sediment on the floors of lakes, seas, and the ocean. The coral reefs are perhaps the most spectacular example of structures formed by deposition of calcium carbonate.
How long would it take this process to remove all carbon dioxide from the present atmosphere? Calculations show that, at current weathering rates and with the present mass of biota in the oceans, the removal time is less than one million years – only 0.02% of the whole history of Earth. With most of the carbon dioxide gone, the oceans would freeze over very quickly.
Something else must happen, both today and in the past, to release the carbon dioxide from the calcium carbonate and return it to the atmosphere.
That “something” is plate tectonics. The ocean floor is contin- ually being recycled back into the mantle at subduction zones, and the carbonate-laden sediments are carried with it. Much of the ocean-floor material subducted at trenches is melted at tem- peratures well over 1,000 K. Carbon-bearing materials, such as the calcium carbonates, react with silicates at these high temper- atures to make calcium silicates (CaSiO3) and CO2gas. The gas makes its way back to the surface not at the midocean ridges, but right at the subduction zones where volcanism is occurring.
Mount St. Helens, Mount Pinatubo, and other active volcanoes
Table 14.1Earth’s carbon reservoirs (adapted from Falkowski et al.2000)
Reservoir Amount in gigatons
Atmosphere 720
Ocean 38,000
Carbonates in sediments ≥60,000,000
Biomass, alive, in biosphere 600
Dead biomass 1,200
Fossil fuels (oil, goal, gas) 4,100
Kerogens 15,000,000
Inorganic carbon 38,000
belch CO2gas from deep within the subducted plates, resupply- ing the atmosphere.
Carbon is cycled through the atmosphere into the ocean and onto the seafloor, only to be subducted and returned to the atmo- sphere. This cycle, shown in Figure 14.7, is limited by the time it takes ocean floor (once formed at mid-ocean ridges) to be subducted. At current plate-tectonics spreading rates, the cycle takes about 60 million years; in other words, any given carbon atom in atmospheric carbon dioxide typically will form carbon- ate, be subducted, and then released again as carbon dioxide gas in a time of order 60 million years.
Table 14.1 lists the known reservoirs of carbon on the Earth today with measured or estimated abundances in one type of unit often used for the carbon cycle: billions of tons, or giga- tons. The atmospheric reservoir is mostly carbon dioxide, and the oceanic reservoir is primarily hydrogen carbonates, both of which are dwarfed by the carbonate sediments on the ocean floor.
Biomass – carbon in molecules that are part of the biosphere today, either in living or dead organisms, is much less than the
carbon in the oceans. Buried carbon exists in several forms.
The amount of carbon buried and processed into fossil fuels (Chapter 23) is much smaller than what is believed to have been converted by heat and pressure of deep burial into so-called kerogens – very carbon-rich, hydrogen-poor organic molecules.
However, the abundance of kerogens is highly uncertain. Finally, so-called inorganic carbon is that defined to be present in miner- als such as limestone, and not containing hydrogen or fluorine.
This reservoir is small compared to the carbonate sediments, most of which will become mineralized, with a fraction con- verted back to carbon dioxide in subduction zones.
14.7.2 Negative feedbacks in the carbon–silicate cycle
Michigan geophysicist J. C. G. Walker proposed that this carbon–silicate weathering cycle might well act as a stabiliz- ing influence on Earth’s climate. Because the cycle requires liquid water to dissolve carbon dioxide gas and to effectively weather rock, a fully frozen Earth would have lost the erosive portion of the process. Carbon dioxide gas would not be lost to ocean floor, while more carbon – previously cycled into the crust, or derived from deeper mantle rocks – would continue to accumulate in the atmosphere. This would raise the temperature through the greenhouse effect, until the oceans could melt and liquid precipitation became possible again. (These are two sepa- rate conditions; rainfall requires somewhat higher temperatures than does melting the oceans, the freezing point of which is lowered by salts.)
Conversely, higher atmospheric temperatures increase the evaporation from oceans, the amount of cloud, and hence rain- fall rates. Higher temperatures also favor rainfall over snow at high latitudes and elevations. These have the net effect of increasing the rate of weathering by rainfall, and hence removal
Figure 14.7Weathering cycle of carbon, silicate, and water.
4.5
CO2 partial pressure, bar
Time before present, Gyr late Precambrian
glaciation
Cretaceous simulations Huronian glaciation
early outgassing, no continents
Permo-Carboniferous glaciation 10
1
10−1
10−2
10−3
10−4
.5 1.5
2.5 3.5
Figure 14.8History of the CO2partial pressure in Earth’s atmosphere based on various lines of evidence regarding Earth’s surface temperature.
Changes in slope at 2.5 and 1 billion years are notional only, based on occurrence of ice age episodes. “Cretaceous simulations” are described in Chapter 19. Adapted from Kasting and Ackerman (1986).
of carbon dioxide from the atmosphere. Sedimentation rates are higher, but the rates of subduction and carbon dioxide outgassing are not affected. As a result, carbon dioxide abundance in the atmosphere decreases, weakening the greenhouse effect and lowering temperature.
The carbon–silicate cycle possesses what is called anegative feedbackloop in which changing the conditions in one direction tends to cause the system to move back in the other direction.
This is characteristic of physical systems that are stable and provides at least a partial explanation as to why Earth’s climate has remained in the temperature range allowing liquid oceans over geologic time. Absent such a feedback, changes in the Sun’s luminosity, in the rate of spreading of plates, and in the amount of volcanism, as well as disasters such as giant impacts (Chapter 18) could have moved Earth’s climate out of the range in which life is sustainable.
Life itself has played a role in altering carbon loss rates:
the development of soil-forming microorganisms some 3 billion years ago accelerated the trapping of carbon dioxide in soils and hence may have led to a net decrease in carbon dioxide levels in the late Archean atmosphere. The development of calcareous plankton shifted most of the deposition of carbonate to deep ocean rather than to shallow, continental-shelf environments, hastening the transport of carbonates to subduction zones and increasing the rate of reintroduction of carbon dioxide to the atmosphere. These and other evolutionary changes in Earth’s biosphere have thus caused shifts in climate to which life has had to adjust through the formation of new species (Chapter 18). The British scientist James Lovelock, and others, have even
proposed that life acts to control the environmental feedback pro- cesses to maximize the habitability of Earth. This controversial
“Gaia” hypothesis is thought provoking but not required neces- sarily to explain the stability of Earth’s climate. Perhaps instead, life is a somewhat meddlesome passenger largely along for the ride.
14.7.3 The carbon–silicate cycle during the Archean Although the carbon–silicate cycle seems to be a good candidate for explaining how Earth’s climate has been stabilized over time, it is necessary to think carefully about how it operated during the Archean, when conditions were different from today. Smaller amounts of continental mass exposed above sea level could have reduced the efficacy of silicate weathering, which is the step that determines the rate of carbon dioxide sequestration. More rapid recycling of crust at the time would have led to a faster return of carbon dioxide to the atmosphere, with less stored in sediments on the seafloor. Both of these somewhat speculative differences between the Archean and more modern times mitigate in favor of leaving the available carbon dioxide in the atmosphere, rather than locked in sediments, during the Archean.
The high carbon dioxide abundances required to sustain Archean temperatures above that of the water freezing point led to a potential instability in the early Archean: the Sun’s lumi- nosity was low enough that the feedbacks in the carbon–silicate cycle might not have worked to bring Earth out of an ice-covered state, if it fell into one during that time. The reason for this lies
in a phenomenon that we discuss for Mars in Chapter 15 – cold temperatures and high carbon dioxide abundances would have caused carbon dioxide clouds to form, with the cooling effect of these clouds short-circuiting the gas’s ability to warm the surface. This is a problem unique to the Archean because, for higher solar luminosities, less carbon dioxide is required to drive Earth out of a global ice age, so that carbon dioxide cloud formation is not an issue. However, some models suggest that carbon dioxide clouds might, under certain conditions, actually warm the surface (Chapter 15).