Two-billion-year-old evaporites capture Earth’s great oxidation

See allHide authors and affiliations

Science  20 Apr 2018:
Vol. 360, Issue 6386, pp. 320-323
DOI: 10.1126/science.aar2687

A strongly oxidizing Paleoproterozoic era

Two billion years ago, marine sulfate concentrations were around one-third as high as modern ones, constituting an oxidizing capacity equivalent to more than 20% of that of the modern ocean-atmosphere system. Blättler et al. found this by analyzing a remarkable evaporite succession more than 1 billion years older than the oldest comparable deposit discovered to date. These quantitative results, for a time when only more qualitative information was previously available, provide a constraint on the magnitude and timing of early Earth's response to the Great Oxidation Event 2.3 billion years ago.

Science, this issue p. 320


Major changes in atmospheric and ocean chemistry occurred in the Paleoproterozoic era (2.5 to 1.6 billion years ago). Increasing oxidation dramatically changed Earth’s surface, but few quantitative constraints exist on this important transition. This study describes the sedimentology, mineralogy, and geochemistry of a 2-billion-year-old, ~800-meter-thick evaporite succession from the Onega Basin in Russian Karelia. The deposit consists of a basal unit dominated by halite (~100 meters) followed by units dominated by anhydrite-magnesite (~500 meters) and dolomite-magnesite (~200 meters). The evaporite minerals robustly constrain marine sulfate concentrations to at least 10 millimoles per kilogram of water, representing an oxidant reservoir equivalent to more than 20% of the modern ocean-atmosphere oxidizing capacity. These results show that substantial amounts of surface oxidant accumulated during this critical transition in Earth’s oxygenation.

The geological record preserves evidence of Earth’s dynamic surface oxygenation [reviewed in (1, 2)], but quantifying this history remains a challenge. The presence or absence of red beds, banded iron formations, and detrital grains of pyrite and uraninite (1, 3) qualitatively indicate increasing oxidation during the Paleoproterozoic era, and the disappearance of large-magnitude mass-independent fractionation (MIF) of sulfur isotopes at 2.4 to 2.3 billion years ago (4, 5) shows that the atmosphere exceeded a redox threshold of ~1 part per million Po2 (partial pressure of oxygen) (6, 7). However, this limit reflects only a tiny fraction of the potential surface oxidant budget and does not capture subsequent redox changes in the Earth system. Today, marine sulfate ([SO42–(aqueous)] = 28 mmol/kg) constitutes one of the largest surface oxidant reservoirs, equivalent to almost twice the modern atmospheric O2 inventory. Therefore, quantitative bounds on marine sulfate concentrations are essential for constraining the net electron balance and accumulation of oxidants on Earth’s surface.

Sedimentary evaporite minerals are one of the best archives of ancient seawater chemistry [e.g., (8, 9)] and specific isotopic signals [e.g., (10, 11)]. Unfortunately, most Precambrian evaporite deposits consist of pseudomorphic replacements (12), and, until recently, the oldest known preserved evaporitic halite and bedded sulfates dated from ~830 million years ago (13, 14) and ~1.2 billion years ago (15), respectively. This study presents analyses from a remarkably preserved ~2.0-billion-year-old marine evaporite succession bearing carbonates, sulfates, halites, and bittern salts. This succession was discovered during the 2007–2009 drilling of the Onega Parametric Hole (OPH), which intersected 2.9 km of Paleoproterozoic sedimentary and volcanic rocks and 600 m of Archean gneiss in the Onega Basin, Karelia, Russia (16, 17). By extending the record of extensive marine evaporites by almost a billion years, core samples from the OPH provide a window into surface conditions and redox balance in the aftermath of the initial rise of oxygen on Earth.

The interval of the OPH presented here lies between 2940 and 2115 m depth and corresponds to the ~2.0-billion-year-old Tulomozero Formation (age discussed in the supplementary materials). In other cores and outcrop exposures, this formation contains abundant pseudomorphic replacements of evaporite minerals (18, 19); original evaporite minerals are only preserved in the OPH where they define three major units. Unit 1 (2940 to 2833 m; 40% average core recovery in cored intervals) comprises dark red-pink, recrystallized halite with intraclasts of anhydrite, magnesite, and mudstone (Fig. 1, A and B), including ~10% various magnesium- and potassium-sulfate salts (Fig. 2). Unit 2 (2833 to 2330 m; 56% average core recovery in cored intervals) consists of decimeter- to meter-scale interlayered anhydrite, magnesite, and laminated dark gray to red mudstone (Fig. 1C) with minor glauberite, gypsum, and halite in its lower part. Unit 3 (2330 to 2115 m; 44% average core recovery in cored intervals) is typified by pink-tan, commonly microbially laminated dolostone (Fig. 1D), laminated red-brown-gray mudstone, and variable amounts of magnesite; quartz and dolomite pseudomorphs of calcium-sulfate minerals occur throughout the lower half of this unit, forming laths, nodules, discs, swallow-tail crystals, and chicken-wire fabric (fig. S2).

Fig. 1 Representative evaporite rocks of the Tulomozero Formation in the OPH.

(A) Cored intervals of halite with rounded gray intraclasts consisting of mudstone, anhydrite, and magnesite (box length is 1 m; 2900 m depth). (B) Halite with felted anhydrite grains and anhydrite-magnesite intraclasts (2854 m depth). (C) Magnesite-anhydrite (white) and halite overlain sharply by laminated red-gray mudstone with desiccation cracks; a magnesite-anhydrite bed is infilling a compacted desiccation crack at the top of the image (2528 m depth). (D) Laminated fine-grained dolostone (2304 m depth). White scale bars are 1 cm in length.

Fig. 2 Interpretive stratigraphy of the Tulomozero Formation in the OPH and associated mineralogical and geochemical data.

Calcium isotope data for samples influenced by former aragonite are shown by open circles with crosses. Additional δ34S data (gray filled circles) are from (30). Fms, formations; cc, calcite; arag, aragonite; dol, dolomite; SW, seawater; VCDT, Vienna Canyon Diablo Troilite. Methods are described in the supplementary materials.

Considering the Tulomozero Formation’s features in the OPH and its development across the 18,000 km2 of the Onega Basin (18, 19), the interpreted depositional setting is a restricted marine embayment with sabkha–coastal plain and shallow-marine environments (fig. S3). The OPH succession captures a decreasing degree of evaporation, from a state of halite and magnesium- and potassium-sulfate saturation (unit 1) through calcium-sulfate deposition (unit 2) and then to dolostone precipitation typical of a more open-marine setting (unit 3). The extent, thickness (>800 m in the OPH core), and mineral sequence of the evaporite succession are comparable to those of Phanerozoic evaporite basins.

The isotope geochemistry of the OPH evaporites presents an opportunity to investigate Earth’s ocean-atmosphere system ~2.0 billion years ago (Fig. 2). Quadruple sulfur isotope analyses of samples from units 1 and 2 reveal Δ33S and Δ36S values that are indistinguishable from zero, confirming that the atmosphere was oxic (6, 7) and that production of the Tulomozero Formation sulfate occurred well after that atmospheric transition. Triple oxygen isotope measurements of sulfates yield resolvably negative Δ17O values. Oxygen MIF derived from atmospheric O3/O2 photochemistry (20) cannot be ruled out, but the small-magnitude Δ17O signals preclude a unique interpretation (supplementary materials), and a quantification of Po2 is not possible.

The mass-dependent behavior of sulfur and calcium isotopes provides compositional constraints on ancient seawater. Sulfate δ34S values lie between 5 and 7 per mil (‰), except for the uppermost sample in unit 2 (discussed further in the supplementary materials). Given the small 34S enrichment during sulfate evaporite formation (10), the seawater sulfate δ34S composition is estimated to have been 4 to 6‰ during deposition of units 1 and 2. The homogeneous sulfur isotopic composition across ~400 m of OPH stratigraphy and the composition and sheer volume of evaporite minerals suggest that the OPH evaporites must have derived from seawater and preserve robust, primary isotopic signals. Additionally, the presence of halite and highly soluble bittern salts in unit 1 argues against interaction with large volumes of fluid and supports the interpretation of primary isotopic ratios for the major mineral-forming elements. Calcium isotope ratios show a clear stratigraphic relationship following mineralogical trends, with the highest δ44/40Ca values in unit 1, decreasing values in unit 2, and even lower values in unit 3. Three samples in unit 3 with the lowest δ44/40Ca values (–1.6 to –1.4‰, relative to modern seawater) also exhibit relatively high strontium content, with one sample containing minor relict aragonite; these observations indicate that the bulk sediment likely contained primary aragonite that has now largely been converted to other carbonate minerals (supplementary materials). Excluding those samples associated with aragonite, where mineralogy rather than evaporitic processes is the first-order control on calcium isotope ratios, the increase in δ44/40Ca values in the more highly evaporated facies is consistent with evaporite precipitation driving isotopic distillation of calcium, the magnitude of which is sensitive to the initial composition of seawater (21).

An estimate for seawater sulfate concentrations ~2.0 billion years ago can be derived from the observed sequence, mineralogy, and calcium isotope ratios of the OPH evaporites (Fig. 3). Constraints were assessed by comparison with batch evaporation simulations with varying initial ion concentrations. The relative concentrations of calcium and sulfate are the principal controls governing the precipitation sequence that is expressed in the OPH—as in modern evaporites—of carbonates (unit 3) followed by calcium sulfates (unit 2), halite, and eventually magnesium sulfates (unit 1). As such, the OPH preserves a reversed evaporite sequence where the degree of evaporation decreases stratigraphically upward, progressing from the most evolved brine at the base of the Tulomozero Formation toward more open-marine conditions. During a forward evaporite sequence, calcium precipitates as sulfate minerals and minor carbonate with an isotopic fractionation, so that calcium in the remaining brine becomes enriched in the heavier isotopes through Rayleigh distillation. If sufficient sulfate is present to remove a large fraction of the original calcium content of seawater, later calcium-bearing phases can record large δ44/40Ca enrichments (21). The ~1‰ δ44/40Ca range captured in the OPH between shallow-marine carbonates in unit 3 and halite-hosted anhydrite in unit 1 therefore places a lower limit on sulfate concentrations. Together with mineralogical constraints, and assuming modern concentrations of other major ions and a conservative interpretation of δ44/40Ca values (but see the supplementary materials for further discussion), the minimum sulfate concentration consistent with these observations is ~10 mmol/kg (Fig. 3).

Fig. 3 Constraints on seawater chemistry during deposition of the Tulomozero Formation.

Filled circles show results from batch evaporation simulations with variable calcium and sulfate concentrations; all other ions were as in modern seawater. Small open circles indicate failure to precipitate gypsum before halite. The color of filled circles indicates the fraction of initial calcium (fCa) removed at halite saturation. The conversion of fCa to δ44/40Ca range is based on a Rayleigh distillation model with α = 0.99905 (supplementary materials). M and S identify the compositions of modern and estimated Silurian seawater (8), respectively. Lines indicate constraints from OPH observations (arrows give directionality), and the blue shaded region shows the range of seawater compositions consistent with these constraints.

The OPH core provides quantitative evidence that marine sulfate concentrations ~2.0 billion years ago were at least a third those of modern seawater. This constraint validates assertions that a large Paleoproterozoic sulfate reservoir existed (18, 22) and increases fourfold the previous lower bound of 2.5 mmol/kg, derived from observing that gypsum evaporites precipitated before halite (22, 23). Although the ancient ocean volume is unknown, a sulfate concentration of 10 mmol/kg in a modern-sized ocean represents an oxidant reservoir equivalent to 23% of the present ocean-atmosphere oxidizing capacity (or 62% of the present atmospheric O2 inventory). The growth of such a reservoir from a sulfate concentration of <200 μmol/kg in the Neoarchean (24) would account for a net redox imbalance of at least 8 to 24 × 1010 mol/year in equivalent moles of O2 produced or organic carbon buried over 100 to 300 million years. The accumulation of such a sizable fluid oxidant reservoir within the given time constraints, compared with estimates of modern organic carbon burial [5 × 1012 mol/year (25)], can be explained by either a large and rapid decline in reductant sinks (i.e., sulfide) or a prolonged 2 to 5% imbalance over 108-year time scales. In either case, the observations suggest that a sustained increase in net O2 production occurred in the Paleoproterozoic.

The geologically rapid growth of a massive sulfate reservoir, with or without a commensurate increase in atmospheric O2, also has implications for feedbacks between the global biogeochemical cycles of O2 and CO2 and Earth’s climate. In particular, the oxidation of large amounts of reduced sulfur requires additional sources of carbon to offset the inferred organic carbon burial (the initial source of oxygen) and prevent catastrophic cooling (26). Models for Earth’s oxidation must balance these considerations, as well as the new evidence for a substantial oxidant reservoir in the form of marine sulfate. Additionally, although sulfate likely represented the largest oxidant reservoir during deposition of the Tulomozero Formation, its concentration subsequently decreased, so that evaporites ~1.9 billion years ago no longer precipitated gypsum before halite (12, 27) and a fundamental change in the sedimentary sulfur isotopic composition occurred (28). The apparently transient accumulation of surface oxidants is not yet well understood (2, 29) but implies a protracted reorganization of the global redox budget on the time scales of sedimentary recycling of pyrite and organic carbon (i.e., hundreds of millions of years). Regardless of the mechanisms involved, the observations presented here from the OPH core document a large oxidant pool ~2.0 billion years ago—a pivotal constraint on the history of Earth’s oxidation.

Supplementary Materials

Materials and Methods

Supplementary Text

Figs. S1 to S14

Table S1

References (3179)

Data S1 and S2

References and Notes

Acknowledgments: Thanks to T. H. Bui for assistance with sulfur isotope analyses and B. A. Wing for helpful discussions. This work was supported by a grant from the Simons Foundation (SCOL 339006 to C.L.B.) and by the European Research Council (ERC Horizon 2020 grant 678812 to M.C.), the Research Council of Norway (RCN Centres of Excellence funding scheme project 223259 to K.P. and A.L.), the Estonian Science Agency (PUT696 to K.K., A.L., K.P., and T.K.), and Princeton University (to J.A.H.). Core material from the OPH is maintained by the Institute of Geology, Karelian Research Centre, Petrozavodsk, Russia. Data presented in this study are available in the supplementary materials.

Stay Connected to Science

Navigate This Article