Report

Late Precambrian Oxygenation; Inception of the Clay Mineral Factory

See allHide authors and affiliations

Science  10 Mar 2006:
Vol. 311, Issue 5766, pp. 1446-1449
DOI: 10.1126/science.1118929

Abstract

An enigmatic stepwise increase in oxygen in the late Precambrian is widely considered a prerequisite for the expansion of animal life. Accumulation of oxygen requires organic matter burial in sediments, which is largely controlled by the sheltering or preservational effects of detrital clay minerals in modern marine continental margin depocenters. Here, we show mineralogical and geochemical evidence for an increase in clay mineral deposition in the Neoproterozoic that immediately predated the first metazoans. Today most clay minerals originate in biologically active soils, so initial expansion of a primitive land biota would greatly enhance production of pedogenic clay minerals (the “clay mineral factory”), leading to increased marine burial of organic carbon via mineral surface preservation.

Geochemical and physical evidence suggests that a stepwise increase in oxygen occurred around 1.1 to 0.54 billion years ago (Ga) (15) and was a necessary precondition to support the physiological needs of large metazoans (68). Whereas the dominant supply of free oxygen to the atmosphere is oxygenic photosynthesis, which began by at least 2.8 Ga (9), the changes in the earth system facilitating a rise in oxygen in the latest Precambrian remain controversial. Photosynthetic oxygen stays in the atmosphere and ocean unless it is consumed by oxidation of organic matter or other reducing agents at the Earth's surface. The stepwise pattern of oxygen increase implies the modification or activation of a mechanism that became a permanent part of the oxygen and carbon cycles, one that should be important and apparent in modern organic carbon burial but absent in the Precambrian. Although many hypotheses have been offered to explain the late Precambrian rise of oxygen (1, 4, 8), the irreversible and stepwise pattern argues against processes that are episodic, reversible, or equally active before and after the rise of oxygen, such as tectonics. Here, we offer a paradigm for organic carbon burial in Precambrian marine sediments based on clay minerals, the major cause of carbon preservation and burial in the modern system that we believe was initiated during the latest Precambrian.

Studies of modern marine systems over the past 10 years give a new understanding of how organic carbon enters the sedimentary record. Rather than the conventional model of particles of organic matter rapidly buried in sediment and preserved, the bulk of organic carbon deposited in continental margin sediments is dispersed among or bound to clay minerals (phyllosilicates), as indicated by strong correlations with surface area, data from physical separations, and microscopy (1014). By contrast, discrete, unbound particulate organic matter comprises <10% of total organic carbon (TOC) (11, 13, 15). The spatial correlation of TOC with clay mineral concentration is likely a result of a sheltering and preservative effect that phyllosilicate surfaces provide to organic matter, and laboratory experiments show that mineral association powerfully enhances preservation of labile carbon compounds (15). Whatever the exact mechanism of protection, studies of both modern and ancient (16) sediments indicate the first order importance of clays as a means of concentrating, accompanying, or preserving organic carbon (11, 13, 14).

Elemental and mineralogical trends indicate long-term changes in shale composition from the Precambrian to the Phanerozoic that could have important implications for an increase in clay production. There is a progressive decline in the K2O/Al2O3 of shale (17, 18) in the Russian and the North American platforms. Loss of K relative to Al is expected with a shift from tektosilicate-dominated (feldspar and quartz) shales and siltstones, with minor amounts of illite-mica-chlorite accessory clays, to smectite- and kaolinite-dominated claystones [summarized in (18, 19); this trend is also influenced by burial diagenesis]. This chemical shift is consistent with a mineralogic shift from mica-illite and chlorite toward smectite, mixed layer smectite-illite, and kaolinite assemblages (19) typical of pedogenic clay minerals (PCM). Simultaneously, incompletely weathered arkose and greywacke become less abundant (18, 20), giving way to chemically mature orthoquartzite (21) (Fig. 1B).

Fig. 1.

(A) Solid line represents generalized 87Sr/86Sr seawater evolution from carbonate sediments [from (26)]. Minimum 87Sr/86Sr for a given age show a stepwise increase in 87Sr/86Sr, suggested here to result from an increase in chemical weathering resulting from the advent of biogenic soils. Dashed line represents the rise in concentration of dissolved marine sulfate [from (2)], which coincides with general rise in atmospheric oxygen (2). (B) Solid lines and left axis record the increase in expandable clays and kaolinite (PCM) into the Phanerozoic relative to illite, mica, and chlorite within shale from (19), reflecting illitization as well as enhanced chemical weathering and PCM production associated with biotic soils. The right axis and dashed line records K2O/Al2O3 of shale from the North American and the Russian platforms compiled by (18), showing a shift toward PCM production from mechanically produced lithologies (mica, illite, and feldspar). Also see (17). Elemental data from our sample suite (table S1) show a positive relation between K and Al content (r = 0.614), suggesting postdepositional illitization is the dominant control on K/Al, but not necessarily on total clay, in the samples we studied.

On the basis of these data, we hypothesize secular change in the mineralogy of fine-grained marine sediments (shale and mudstone) from mechanically weathered micas and tektosilicates of high-temperature origin to a more phyllosilicate-rich pedogenic assemblage that should be associated with an enhanced preservation of organic carbon in the sedimentary reservoir in the mode of modern sediments discussed above. The absence of a working Precambrian “clay factory” should be evident in the rock record, because PCM comprise an important sink of ions produced by weathering and, with recycled clays, make up >60% of Phanerozoic shale, the most common rock type at the Earth's surface (19).

To determine whether a secular change in clay abundance occurred during the latest Precambrian, we investigated a prominent passive-margin succession spanning the 850 to 530 million years ago (Ma) transition in Australia. This succession was selected because it exhibited (i) a thick interval (>10 km) of passive-margin, shelfal, fine-grained, marine sediments, a majority of which are shales and mudstones (22) typical of depositional environments in which modern mudstones accumulate, and (ii) sufficient exposure to sample the finest-grained lithologies. Although the Australian section provides one of the most complete successions globally (it houses the stratotype section and point defining the Ediacaran Period) and is an ideal test of our hypothesis, we also studied successions in south China and Baltica. To avoid pitfalls of averaging, we sampled the finest-grained intervals within each succession (15 formations total) (table S1) to test the hypothesis.

We found a striking increase in the x-ray diffraction (XRD) peak ratio of total phyllosilicates (clay minerals and micas) to quartz through the Neoproterozoic (Fig. 2) (23), supporting the hypothesis that clay formation increased radically at the end of the Proterozoic. We suggest that this trend most likely represents a secular change in PCM formation, delivery, and abundance rather than local, effects, diagenesis, or tectonically influenced trends, because it (i) occurs in multiple margins with different geologic histories, (ii) spans a broader range of time than any single process capable of influencing grain-size trends (i.e., sea level change, deltaic deposition, glaciation, etc.), and (iii) is robust across multiple depositional sequences (cycles of deepening) capable of concentrating the finest-grained fraction available. Additionally, there are no shale diagenetic mechanisms we are aware of that could account for such a strong shift from tektosilicates to octahedrally bound Al in phyllosilicates in (pH neutral) marine sediments.

Fig. 2.

Secular changes in the relative amounts of phyllosilicate and quartz determined by using the Schultz ratio (16, 23, 25, 38) for Neoproterozoic to Cambrian mudstones. Samples represent the finest-grained units that were at least 5 m thick from three locations: South Australia (triangles), Baltica (circles), and south China (squares). These data show an increase in the proportion of phyllosilicates relative to quartz with decreasing age. Because of the stability of quartz, its abundance serves as a crude standard indicating the amount of tectosilicates either present (Schultz ratio low) or once present, now clay minerals (Schultz ratio high) in a shale of average chemical composition. Gray symbols are the range of values from a specific unit (all samples analyzed), whereas the solid figure is the average value for a given unit. Units studied include the following: 1, Oraparinna Shale; 2, Doushantuo Formation; 3, Birri Formation; 4, Ekre Shale; 5, Innerelva Member and Stappogiedde Formation; 6, Bunyeroo Formation; 7, Brachina Formation; 8, Nyborg Formation; 9, Reynella Member and Elatina Formation; 10, Enorama Shale; 11, Tarcowie Siltstone; 12, Tapley Hill Formation; 13, Mintaro Shale; 14, Saddleworth Formation; and 15, Woolshed Flat Shale. Age assignments are approximate, and correlations between basins have their basis in the assumed age of 630 Ma for the Marinoan (lower Varanger) glacial and 740 Ma for the Sturtian glacial. Sample locations, data, methods (including elemental data), and stratigraphic references can be found in table S1.

Because clay minerals deposited in marine sediments are sensitive records of continental paleoenvironments (24), a secular change in clay mineral abundance at the end of the Precambrian may have important implications for terrestrial soils. Unlike other detrital grains such as quartz or feldspar, most clay minerals are not produced by simple mechanical reduction of parent rock and should not be confused with mechanical reduction in grain size of any silicate to clay size (<2 μm). PCM rarely form by solid-state transformation of another mineral but rather precipitate as phyllosilicate crystals from cation-rich soil solutions that are in equilibrium with the ambient environment. Likewise, there is little evidence for large-scale conversion of tektosilicates (dominantly feldspars) to clay minerals in fine-grained sediments, and hydrothermal clay formation produces different clay mineral suites than those found in shales (19, 24, 25). The latitudinal mineralogical zoning of detrital clay minerals in modern marine sediments is consistent with increasing chemical weathering intensity with decreasing latitude, indicating the importance of this terrestrial contribution to marine clay deposition (24).

The lower clay mineral abundance in the earlier sections indicates a reduced continental chemical weathering intensity. This dilution does not, however, preclude the formation of clay minerals and their concentration in local depositional environments throughout the Proterozoic; such clay formation commonly occurs by weathering of volcanic sediments, and clay can be concentrated by sedimentary processes. However, it is not representative of the dominant mode of shale formation, and so these deposits should be rare. The increased phyllosilicates with time identify a new or enhanced source of PCM and increased clay sedimentation (apparent in Cambrian mudstones and through the Phanerozoic), whereas the increasing range of values indicates the effects on the growing clay flux by varying amounts of silt dilution. TOC (without considering sedimentation rate) does not record carbon burial flux; however, TOC values (table S1) are at least consistent with our hypothesis, showing a commensurate rise with PCM even after thermal alteration.

Enhanced continental chemical weathering implied by PCM formation is also recorded in the step to more radiogenic 87Sr/86Sr values within Neoproterozoic seawater (26) (Fig. 1). Because the marine 87Sr/86Sr record provides a globally integrated measure of chemical weathering, it corroborates the secular origin for the clay mineral trend (Fig. 2). Seawater Sr isotopic composition derives from continental weathering and hydrothermal or mantle sources, whose contributions can be estimated because continental runoff is much richer in 87Sr than hydrothermal sources. The Sr isotope record offers an important corroboration of the hypothesis because the seawater values can only be influenced by chemically weathered ions and will not change with variations in fluxes of mechanically weathered material (assuming constant hydrothermal contribution). Further, because minerals such as feldspar and mica derived from mechanical weathering of igneous and metamorphic rocks are strongly enriched in 87Sr as well as K, an enhanced rate of chemical weathering and ion liberation to solution (and PCM formation) should increase 87Sr/86Sr in marine carbonates.

Two patterns of change mark the marine Sr isotopic record (Fig. 1): (i) a long-term (billion year) plateau with a major step at 0.7 Ga, establishing markedly more radiogenic marine values, and (ii) a shorter-term (∼ 100 My) rise and fall in values coincident with the Cambrian Pan-African and the Tertiary Himalayan orogenic events. Here we focus on the long-term, step-like rise of marine 87Sr/86Sr from 0.705 during the early Neoporoterozoic to 0.707, which begins in the late Neoproterozic and subsequently comprises the lowest value through the Phanerozoic. We suggest that this new plateau results from an irreversible change in chemical weathering regime in parallel with a secular rise in clay mineral abundance.

What might have caused such a change in continental weathering? In the Phanerozoic, biotic soils promote production of PCM by providing organic matter that mechanically stabilizes soil profiles, retains cations and water, and increases fluid residence time (24, 27). Whereas minor amounts of clay formation are possible directly on rock surfaces, the difficulties of retaining and concentrating leached ions are substantial, so that rock surfaces do not produce clay minerals as well as biotic soils. The biotic soil–forming environment for clay minerals can be likened to the carbonate “factory” within the carbonate system. In the absence of a clay mineral–forming mechanism before the late Precambrian, clay minerals deposited in marine sediments were probably not substantial until the evolution and colonization of the terrestrial environment by some form of a primitive land biota.

The transience of terrestrial environments (especially soils) prevents the geological record from preserving reliable evidence of the transformation from an abiotic to a biotic surface. However, physical and geochemical data suggest a primitive land biota by at least 1 Ga (27, 28, 29). Molecular evidence suggests mosses, fungi, and liverworts by ∼700 Ma (30) and a fossil marine fungi/lichen record by 600 Ma (31) [fungi markedly enhance weathering (32)]. Carbon isotopes of Neoproterozoic karst suggest that soil biomass was sufficient to impart a Phanerozoic-like meteoric isotopic signal to diagenetic carbonate (31). Thus although we cannot specifically identify the transitional states of a land biota, there is ample evidence for these transitions during the Proterozoic. Furthermore, modern microbial crusts stabilize soil profiles, retain water, enhance weathering rates, drive chemical differentiation, and provide chemically adsorptive organic matter (3234). Thus, the advent of soils sufficiently biotic for clay formation likely predated complex terrestrial ecosystems (33, 35).

One hypothesis for late Precambrian rise of oxygen calls on enhanced burial of organic matter associated with increased tectonic activity between 1.1 and 0.8 Ga (4). This hypothesis uses the positive relation between sedimentation rate and organic matter burial flux observed in modern sedimentary systems, which likely results from the scaling of clay mineral flux with total sedimentation rate (13). Our hypothesis changes the emphasis from temporal changes in quantity of deposition to changes in its quality, that is, its clay mineral abundance. The Neogene Bengal fan illustrates a variation on this theme. Onset of the monsoon in the Miocene drove an increase in chemical weathering in the Ganges-Brahmaprutra watersheds, resulting in a shift from mechanical weathering (tektosilicates, plus mica, illite, and chlorite) to chemical weathering (secondary detrital PCM; smectite, kaolinite) (36). The later PCM mineral assemblage resulted in an ∼fourfold increase in organic loading and substantially increased the organic carbon burial flux (36).

The global change that promoted the rise of animals by 0.6 Ga (37) remains one of the most important yet least understood events in the geobiologic record. Here we adopt a nonuniformitarian approach by identifying an important component of the modern system that was largely absent or ineffective in the Precambrian and changed in an irreversible manner sometime during the Neoproterozoic. The modern ocean buries 1.6 × 1014 g C/year, of which 86% (1.38 × 1014 g C/year) is buried in ocean margins (11), where OC concentration and hence burial flux is linearly proportional to clay content. Applying this burial rate throughout the Phanerozoic and assuming that the peak ratios in Fig. 2 are proportional to clay content, then the clay-driven increase in OC burial that developed during 730 to 500 Ma (Fig. 2) ramped from 0.21 × 1014 to 1.38 × 1014 g C/year. The resulting sixfold increase in oxygen retention would have greatly influenced biogeochemical cycling of redox sensitive elements such as Fe+2 and S2– (2, 3, 5) and ultimately increased the oxygen concentration of the atmosphere. The evolutionary innovation and expansion of land biota could permanently increase weathering intensity and PCM formation, establishing a new level of organic carbon burial and oxygen accumulation.

Supporting Online Material

www.sciencemag.org/cgi/content/full/1118929/DC1

Materials and Methods

Table S1

References and Notes

References and Notes

View Abstract

Navigate This Article