The details of how this particular sub-cycle fits into the Earth system picture lies outside the scope of this review, however. 304 A. Ridgwell, R.E. Zeebe / Earth and Planetary Science Letters 234 (2005) 299–315 A powerful regulatory mechanism of the Earth system arises because weathering rates respond to surface temperature and atmospheric CO2 while simultaneously silicate weathering rates control the rate of transformation CO2YHCO3 and thus rate of loss of carbon through CaCO3 burial. This is a negative feedback system [14] and acts to regulate the concentration of CO2 in the atmosphere over hundreds of thousands of years [1]. Buried carbonate is eventually recycled back from the geologic reservoir. This can occur if carbonates laid down in shallow seas such as limestones or chalks are subsequently uplifted and exposed to weathering as a result of mountain building episodes. However, carbonates deposited in open ocean sediments are only infrequently exposed at the Earth’s surface, as ophiolite complexes—portions of the oceanic crust and overlying sediments that have been trapped between colliding cratonic blocks and uplifted. Instead, the primary recycling of deep-sea CaCO3 occurs through subduction into the upper mantle and decarbonation (see Fig. 1b; #6). At this point it is important to recognize that carbonate burial represents the principal geologic mechanism of CO2 removal from the ocean and atmosphere. However, the act of precipitating CaCO3 has the effect of re-partitioning dissolved carbon in the surface ocean into CO2(aq), raising ambient pCO2 and pH (see Box 1). Thus, precipitation and deposition of CaCO3 have the short-term effect of increasing the concentration of CO2 in the atmosphere at the expense of the ocean carbon inventory, but at the same time represents the ultimate long-term sink for CO2. 2. The role of the global carbonate cycle in the Earth system Over millions of years the silicate rock weathering feedback controls the concentration of CO2 in the atmosphere [1,14]. On time-scales shorter than ca. 100 ky, however, the weathering feedback is ineffective and the marine carbonate cycle plays an important role in determining atmospheric CO2. We illustrate this by considering some of the global changes that marked the end of the last ice age 18 thousand years ago (18 ka), when CO2 rose from a glacial minimum of 189 ppm to 265 ppm at the beginning of the Holocene [15]. The demise of the great Northern Hemisphere ice sheets was marked by a rise in sea-level of about 120 m [16]. With the flooding of the continental shelves came a 4-fold increase in the area of shallow water environments available for coral growth [17]. Because an increase in the rate of CaCO3 deposition will drive more CO2 into the atmosphere, this mechanism was once proposed as an explanation for the 70–80 ppm deglacial rise in atmospheric CO2—known as the dcoral reefT hypothesis [18]. Subsequent ice core measurements made it apparent that the main increase in CO2 occurred prior to the rise in sea-level [19]. However, one cannot reject a role for corals out of hand because reconstructions of the time-history of reef building episodes are unambiguous in demonstrating a profound increase in CaCO3 deposition following the end of the last glacial [20–22]. A priori geochemical reasoning argues that this must translate into a net re-partitioning of CO2 from the ocean to atmosphere. The solution to this is that the dcoral reefT mechanism is essentially a Holocene phenomenon, with postglacial coral re-colonization and reefal buildup potentially explaining much of the 20-ppm increase in CO2 observed in ice cores that starts at around 8 ka [23]. The global carbonate cycle plays other interesting biogeochemical games. Since the last glacial, the expansion of ecosystems to higher latitudes and stimulation of productivity by rising atmospheric CO2 resulted in an increase in the amount of carbon contained in the terrestrial biosphere (vegetation plus soils). The estimates for this increase vary—from around 600 Gt C based on deep-ocean 13C changes [24] (but see [25] for a new re-assessment), ~ 850 Gt C according to global vegetation models [26], to 1300 Gt C (and higher) in some paleo vegetation reconstructions [27]. A transfer of carbon into the terrestrial biosphere of just 500 Gt C should have driven atmospheric CO2 downwards by some 40 ppm [28], yet ice cores show an increase between glacial and early Holocene of 70–80 ppm [15]. However, as CO2 is sucked out of the atmosphere and ocean, oceanic CO32 concentrations (and pH) increase (Fig. 3) enhancing the stability of CaCO3 in deep-sea sediments. Increased carbonate burial drives more CO2 A. Ridgwell, R.E. Zeebe / Earth and Planetary Science Letters 234 (2005) 299–315 305 Log10 [concentration (mol kg-1)] H modern sea-water -1.5 + -2.0 – CO2(aq) -2.5 HCO3 -3.0 OH – 2- CO3 -3.5 -4.0 -4.5 -5.0 0 2 4 6 8 10 12 14 pH Fig. 3. The concentrations of the dissolved carbonate species as a function of pH (referred to as the Bjerrum plot, cf. [44]): Dissolved carbon dioxide (CO2(aq)), bicarbonate (HCO3 ), carbonate ion (CO23 ), hydrogen ion (H+), and hydroxyl ion (OH ). At modern seawater pH, most of the dissolved inorganic carbon is in the form of bicarbonate. Note that in seawater, the relative proportions of CO2, HCO3 , and CO23 control the pH and not vice versa. into the atmosphere, countering about 60% of the impact of re-growth in the terrestrial biosphere to leave a net CO2 fall of just 17 (rather than 40) ppm [28]. This amelioration of a perturbation of atmospheric CO2 by changes induced in the preservation of CaCO3 in deepsea sediments is known as dcarbonate compensationT [29] and represents a critical regulatory mechanism in the modern global carbon cycle on time-scales of 5–10 ky. (Carbonate compensation also represents an additional way of helping to explain the 20 ppm late Holocene rise in atmospheric CO2 [30], to which a rise in sea-surface temperature (SST) [31] and a reduction in terrestrial carbon storage could also have contributed [32,33].) Carbonate compensation on its own does not explain why atmospheric CO2 should have risen during deglaciation at about the same time as the terrestrial biosphere was accumulating carbon. Clearly, there must be additional carbon cycle mechanisms operating at this time to explain the ice core CO2 record, the main candidates being: higher SSTs, reduced sea-ice cover, a more restricted iron supply to the ocean biota, and increased ventilation of the deep ocean [23,28]. Yet another possible way of explaining an increase in atmospheric CO2 arises because the saturation state of the deep-sea sedimentary pore- waters where CaCO3 dissolution takes place is determined not only by X of the overlying waters but also by the amount of metabolic CO2 released by the in situ respiration of particulate organic carbon (POC) [34,35]. Any change in the POC flux to the sediments will therefore alter the fraction of CaCO3 that dissolves. (Strictly, it is the ratio between CaCO3 and POC fluxes, the CaCO3/POC drain ratioT that is the critical parameter rather than the absolute POC or CaCO3 flux, per se). Models predict an atmospheric CO2 sensitivity of about 1.6 ppm per percent reduction in CaCO3/POC [36,37]. A 67% increase in pelagic POC production (or 40% decrease in CaCO3) could therefore theoretically account for the entire deglacial CO2 rise. Thus, although the responsiveness of deep-sea sedimentary CaCO3 preservation offers a means of stabilizing ocean chemistry through carbonate compensation, the atmospheric CO2 control setting on this carbonate regulator can be adjusted by changing surface ocean productivity and CaCO3/POC rain ratio. However, despite its potential for explaining the ice core CO2 record, a primary role for the rain ratio mechanism does not appear consistent with reconstructed shifts in the CCD and lysocline and model analysis [38,39]. Recent interpretations of sediment 306 A. Ridgwell, R.E. Zeebe / Earth and Planetary Science Letters 234 (2005) 299–315 trap data also question whether changes in the CaCO3/POC rain ratio at the surface would be transmitted to the abyssal sediments [40,41] (see Section 4). A different facet of the carbonate cycle is in its role as regulator of the saturation state of the ocean. Understanding past changes in surface saturation (X) provides the environmental context for the geological interpretation of primary carbonate mineralogy, particularly the occurrence of abundant environmentally controlled carbonates such as marine cements and ooids [42]. The occurrence of extremes in X may also be important in understanding the evolutionary driving force behind the advent of biomineralizing species [43]. Furthermore, given the partial pressure of CO2 ( pCO2), knowledge of X (plus temperature and major cation composition) uniquely determines the state of the entire aqueous carbonate system [44]. Thus, as proxy-based reconstructions of paleo atmospheric CO2 for the Phanerozoic improve [45] an understanding of how ocean X has also changed through time would enable all the properties of the aqueous carbonate state to be deduced, providing critical information in the interpretation of Earth history [46]. 3. Evolution of the global carbonate cycle through Earth history We review the history of global carbonate cycling in two parts; the Precambrian (up to 542 Ma), when inorganic geochemical processes tended to dominate the nature and location of carbonate deposition, and the Phanerozoic (542 Ma to present), when life became the single most important factor. 3.1. Carbonate cycling in the Precambrian—when geochemistry ruled the roost The requirements for carbonate cycling to begin on the early Earth are fairly minimal—the contact of basaltic rock with water and dissolved CO2 to initiate chemical weathering [47]. With the weathering of silicate rocks comes the delivery of solutes to the ocean, making an over-saturated surface and the eventual precipitation of carbonates inevitable. The early start to this biogeochemical cycle is reflected in the dated carbonate record which extends back to at least 3800 Ma [48] and deposition of the first facies would have occurred well before this. Early Precambrian carbonates are characterized by sea-floor encrustations, crystal fans, and thick cement beds, all indicative of a relatively rapid and dabioticT mechanism of CaCO3 precipitation. Progressively younger Precambrian rocks show a decreasing abundance of such inorganically precipitated carbonates [49]. This secular trend in carbonate fabric most likely reflects a progressive decline in the degree of ocean over-saturation. However, the reasons for this are not entirely clear. One possibility is because as the atmosphere and surface ocean become more oxygenated towards the end of the Precambrian, sea-water concentrations of Mn2+ and Fe2+ would have declined [49,50]. Since these cations inhibit the precipitation of CaCO3, a reduction in their concentration would mean that a lower degree of over-saturation is required to achieve the same global carbonate deposition rate. Alternatively, the gradual accretion of continental crust and associated increase in area of shallow water depositional environments means that a lower precipitation rate per unit area (and thus X) would be required to balance the same global weathering flux [51]. Whatever the reasons, the Precambrian inorganic geochemical age was brought to a relatively abrupt end as life stepped on the evolutionary accelerator and drove the Earth system through a succession of new modes of carbonate cycling. 3.2. Carbonate cycling in the Phanerozoic—enter the biota The advent of carbonate biomineralization occurred around the time of the Cambrian–Precambrian boundary [52] when evolutionary innovation conferred on organisms the ability to precipitate carbonate structures (skeletons). Prior to this there could have been no significant biologically driven production of CaCO3, and carbonate deposition would have been primarily restricted to heterogeneous nucleation and crystal growth on organic and inorganic surfaces in warm shallow water environments [49]. Because biomineralization enabled the more efficient removal of weathering products from the ocean by the expenditure of metabolic energy, a lower thermodynamic driving A. Ridgwell, R.E. Zeebe / Earth and Planetary Science Letters 234 (2005) 299–315 force for carbonate precipitation would have been required in the ambient marine environment. The result would have been a reduction in ocean saturation (X) as the Phanerozoic got under way. A second major development took place several hundred million years (My) later, with the Mesozoic proliferation of planktic calcifiers [53] and the establishment of the modern mode of carbonate cycling in a dMid-Mesozoic RevolutionT [46]. We illustrate the profound importance of this by considering the response of the marine carbonate cycle to two environmental forcings; (i) sea-level, which varies over hundreds of million years by up to 300 m (Fig. 4a), and (ii) the cation chemistry of the ocean; particularly Magnesium (Mg2+) and Calcium (Ca2+) ion concentrations (Fig. 4c). Although global temperatures and continental paleo-latitude also affect global carbonate deposition by determining the latitudinal extent of carbonate production by warm-water corals [43,54], we will restrict our analysis to just two factors. Times of high sea-level such as the Mid Paleozoic produced flooding extents in excess of 50% on some cratons [55] and the creation of extensive inland (epeiric) seas. This in turn facilitated widespread carbonate platform development and shallow water carbonate accumulation [43,54] which would have driven a lower X. Conversely, times of low sea-level and restricted depositional area would produce a tendency towards high ocean X [56]. Superimposed on this is a variation in the oceanic ratio of Mg2+ to Ca2+ by a factor of three [57–59]. In order to maintain the same global rate of carbonate production, higher ambient Mg2+/Ca2+ requires a more over-saturated ocean because Mg2+ inhibits calcite precipitation [60]. (The inhibition is predominantly a result of the Mg2+ interaction with the solid calcite phase, rather than a solution effect involving Mg2+–CO32 complexation. The latter effect reduces CO32 activity but is independent of the CaCO3 polymorph present in solution—see [61] and references therein.) At higher X, aragonite becomes more common in abiotic cements and hyper-calcifying organisms, which we observe in the geological record as distinctive times of relatively abundant shallow water carbonate aragonite [62,63]—periods dubbed daragonite seasT [64] (Fig. 4d). The coincidence of times of low sea-level and low Ca2+ concentrations (Fig. 4) should have given rise to a 307 highly over-saturated ocean. This is consistent with the widespread occurrence of abundant environmentally controlled carbonates such as cements, calcified cyanobacteria, and thick precipitated beds during parts of the Permian and Triassic [42,65,66], all indicative of comparatively rapid and dabioticT modes of carbonate precipitation. However, despite similar sea-level and cation chemistry, environmentally controlled carbonates are rare in the modern ocean. The difference is a direct consequence of the proliferation of calcareous plankton during the Mesozoic and creation of a new and substantive sink for CaCO3 [46,65,66]. Although benthic foraminifera and other bottomdwelling calcifiers evolved early in the Phanerozoic, it is not until the Mesozoic that a marked proliferation in coccolithophore and planktic foraminiferal diversity and abundance is observed [53,67] (Fig. 4b). Only then would a substantive deep-sea sedimentary carbonate sink have been possible. This supposition is supported by the observed composition of Phanerozoic ophiolite suites which indicate that pelagic carbonate accumulation was comparatively rare in Paleozoic ocean sediments [68] (Fig. 4e). Conversely, the mean area of platform carbonates during the Mesozoic and Cenozoic is much reduced compared to the Paleozoic (Fig. 4f). One might speculate whether the ca. 200 My gap between the first appearance of calcifying planktic foraminifera and coccolithophorids and their rise to relative dominance in global pelagic ecosystems [53] is related to extreme ocean over-saturation in the late Permian and early Triassic, a potential environmental driving force favoring calcifiers. A similar thesis can be advanced to help explain the timing of the advent of metazoan biomineralization following the inferred occurrence of extreme oceanic saturation events during the late Precambrian [69]. The establishment of a substantive deep-sea carbonate sink is important because it introduced a new stabilizing mechanism to the Earth system— dcarbonate compensationT (see Section 2). Indeed, the absence of a responsive deep-sea carbonate sink in the Precambrian would have made the carbon–climate system much more sensitive to perturbation. Ice ages of near-global extent and multi million-year duration deduced for the end of the Precambrian [70] could have been facilitated by the weak dbufferingT of the Precambrian carbon cycle and atmospheric CO2 [71,72]. This view is also consistent with the wide- A. Ridgwell, R.E. Zeebe / Earth and Planetary Science Letters 234 (2005) 299–315 spread occurrence of strange dcapT carbonate facies deposited during postglacial flooding of the shelves. Other explanations for the genesis of cap carbonates in the aftermath of extreme late Precambrian glaciation have been proposed, such as the removal of the solutes derived from rapid rock weathering under a CENOZOIC N Pg high CO2 atmosphere [73,74] and the overturning of a stagnant ocean [42]. However, all hypotheses recognize extreme changes taking place in global carbonate cycling at this time. As well as adding new mechanisms for stabilizing atmospheric CO2, the establishment of a substantive MESOZOIC C PRE-CAMBRIAN PALEOZOIC J Tr Pr C D S O ∈ sea-level (m) 300 E a 200 100 0 Major changes in plankton assemblage -100 Acritarchs Dinoflagellates Diatoms Coccolithophorids Planktic foraminifers Radiolaria b [Ca2+] (mmol kg-1) 60 40 20 c some of the forcings of the global carbonate cycle 308 calcite dominant (low Mg2+/Ca2+) area of CaCO3 % occurrence accumulation (106 km2) 100 aragonite dominant (high Mg2+/Ca2+) calcite dominant (low Mg2+/Ca2+) d deep-sea deposition dominant 80 60 40 e 20 0 neritic CaCO3 deposition dominant 40 30 20 f 10 0 0 100 200 300 Age (Ma) 400 500 600 response of global carbonate cycling as recorded in the geological record 0 A. Ridgwell, R.E. Zeebe / Earth and Planetary Science Letters 234 (2005) 299–315 309 deep-sea sedimentary CaCO3 sink would have also had a destabilizing effect. For instance, episodes of high weathering rates and sequestration of carbon in pelagic carbonates could subsequently lead to periods of enhanced metamorphic CO2 out-gassing to the atmosphere as the sea-floor CaCO3 is subducted and undergoes decarbonation [75]. The subduction of carbonates as ocean basins close and are destroyed would also result in episodic enhanced CO2 release [76,77]. Both mechanisms predict secular oscillation in metamorphic CO2 out-gassing rates on tectonic time-scales, and both would not have been possible before the MidMesozoic Revolution in carbonate deposition. changes in the population or ecological success of shallow-water calcifiers. The Mesozoic shift towards widespread pelagic biomineralization finally led to a significant stabilization of the marine CaCO3 saturation state, termed the dCretanT ocean. Large and rapid shifts between, e.g. the Neritic- and Cretanocean mode have likely occurred in the aftermaths of catastrophic events such as the Cretaceous–Tertiary bolide impact [78]. 3.3. Synthesis The ocean is capable of absorbing about 70% of all the CO2 released by fossil fuel combustion [79]. For a 4000 Gt C dburnT, this means that the equivalent of ~ 600 ppm CO2 will remain in the atmosphere after hundreds of years [80]. An atmospheric CO2 concentration of ~ 1000 ppm is about three times the presentday (year 2003) value of 376 ppm [81] and represents a very significant long-term radiativ

Do you have a similar assignment and would want someone to complete it for you? Click on the ORDER NOW option to get instant services at We assure you of a well written and plagiarism free papers delivered within your specified deadline.