Limited oxygen production in the Mesoarchean ocean Frantz Ossa Ossaa,b,1, Axel Hofmannb, Jorge E. Spangenbergc, Simon W. Poultond, Eva E. Stüekene, Ronny Schoenberga, Benjamin Eickmanna,b, Martin Willef, Mike Butlerg, and Andrey Bekkerb,h aDepartment of Geosciences, University of Tuebingen, 72074 Tuebingen, Germany; bDepartment of Geology, University of Johannesburg, 2092 Johannesburg, South Africa; cInstitute of Earth Surface Dynamics, University of Lausanne, 1015 Lausanne, Switzerland; dSchool of Earth and Environment, University of Leeds, Leeds LS2 9JT, United Kingdom; eSchool of Earth & Environmental Sciences, University of St. Andrews, St. Andrews KY16 9AL, United Kingdom; fInstitute of Geological Sciences, University of Bern, 3012 Bern, Switzerland; gEnvironmental Isotope Laboratory, IThemba LABS, 2050 Johannesburg, South Africa; and hDepartment of Earth Sciences, University of California, Riverside, CA 92521 Edited by Mark H. Thiemens, University of California, San Diego, La Jolla, CA, and approved February 28, 2019 (received for review October 31, 2018) + 15 The Archean Eon was a time of predominantly anoxic Earth surface temporary NH4 oxidation, and thus the δ N record has been conditions, where anaerobic processes controlled bioessential ele- used to infer the development of locally oxygenated surface- ment cycles. In contrast to “oxygen oases” well documented for the ocean environments after ∼2.7 Ga (5, 15, 17, 18, 25). Neoarchean [2.8 to 2.5 billion years ago (Ga)], the magnitude, spatial Independently, stable-isotope systematics of redox-sensitive extent, and underlying causes of possible Mesoarchean (3.2 to 2.8 Ga) elements such as iron (Fe), Mo, uranium, and sulfur, as well as surface-ocean oxygenation remain controversial. Here, we report locally enhanced manganese (Mn) (oxyhydr)oxide precipitation, δ15N and δ13C values coupled with local seawater redox data for support an earlier emergence of oxygenic photosynthesis and Mesoarchean shales of the Mozaan Group (Pongola Supergroup, episodic development of oxygen oases in the Mesoarchean sur- South Africa) that were deposited during an episode of enhanced face ocean (7–9, 11), well before currently accepted evidence for Mn (oxyhydr)oxide precipitation between ∼2.95 and 2.85 Ga. Iron oxidative nitrogen cycling. Furthermore, phylogenomic estimates and Mn redox systematics are consistent with an oxygen oasis in based on molecular clocks also suggest that cyanobacterial stems δ15 the Mesoarchean anoxic ocean, but N data indicate a Mo-based capable of oxygenic photosynthesis might find their roots in the diazotrophic biosphere with no compelling evidence for a signifi- Archean, with a development and progressive diversification start- cant aerobic nitrogen cycle. We propose that in contrast to the ing as early as ∼3.5 Ga (26–28). However, the factors that caused Neoarchean, dissolved O2 levels were either too low or too limited a delay in pervasive oxygenation of the atmosphere–hydrosphere in extent to develop a large and stable nitrate reservoir in the system after the establishment of oxygenic photosynthesis earlier in EARTH, ATMOSPHERIC, AND PLANETARY SCIENCES Mesoarchean ocean. Since biological N2 fixation was evidently active the Archean remain poorly constrained, particularly with regard to in this environment, the growth and proliferation of O -producing 2 the role of their two main biolimiting nutrients, N and P (19–23). organisms were likely suppressed by nutrients other than nitrogen Modeling studies have demonstrated that low dissolved P concen- (e.g., phosphorus), which would have limited the expansion of trations would severely suppress the rate of oxygenic photosynthesis oxygenated conditions during the Mesoarchean. and ultimately the spatial extent of Archean oxygen oases (29). However, there is currently no consensus on dissolved P concen- oxygen oasis | nitrogen isotopes | nutrient limitation | oxygenic – – photosynthesis | Mesoarchean trations in the Archean ocean (21 23, 30 32). To assess controls on the spatial development and intensity of Earth’s first oxygen oases, we measured nitrogen (δ15N) and dramatic rise in atmospheric oxygen level during the Great Oxidation Event (GOE) at ∼2.4 billion years ago (Ga) is A Significance marked by the disappearance of mass-independent fractionation of sulfur isotopes, oxidation of detrital pyrite and uraninite, and “ ” the appearance of red beds, reflecting the irreversible transition Episodic development of oxygen oases during the Archean from an anoxic to an oxic world (1, 2). While it is widely accepted Eon characterizes the hundreds of millions of years transition – that oxygenic photosynthesis was a first-order control on the to permanent oxygenation in the atmosphere hydrosphere ∼ – GOE (3), Archean shallow-marine “oxygen oases” and “whiffs” system at the Great Oxidation Event ( 2.4 2.3 Ga). One of these well-characterized oxygen oases is recorded in Meso- of atmospheric oxygen (O2) have been proposed to have oc- curred up to several hundred million years before the GOE (4– archean sediments of the Pongola Supergroup. We show that 18). However, while processes that drove oxygen production in contrast to the Neoarchean, biological oxygen production in during transient and localized oxygenation events in the Neo- a shallow ocean having Mo-based nitrogen fixation was not archean (2.8 to 2.5 Ga) are supported by a wide range of geo- sufficient to result in a dissolved nitrogen reservoir that would chemical proxies (e.g., refs. 4–6 and 13–18), those from the carry the isotopic effects of an aerobic nitrogen cycle. Never- theless, it appears that low concentrations of bioavailable Mesoarchean (3.2 to 2.8 Ga) are constrained by only a limited phosphorus, rather than nitrogen, suppressed the growth and number of studies (7–12). expansion of oxygenic photosynthesizers and may explain The nitrogen (N) cycle from the Paleoarchean up to ∼2.7 Ga is why pervasive and permanent oxygenation was delayed dur- widely considered to have been dominated by bioavailable am- + ing the Archean Eon. monia (NH4 ) under anoxic water-column conditions (15, 16). + Oxidation of NH4 would have been suppressed in an early Ar- Author contributions: F.O.O., A.H., R.S., and A.B. designed research; F.O.O., A.H., and A.B. chean ocean characterized by extremely low O2 concentrations performed research; F.O.O. contributed new reagents/analytic tools; F.O.O., A.H., J.E.S., (15–17). Free O2 is produced through oxygenic photosynthesis, the S.W.P., E.E.S., M.B., and A.B. analyzed data; and F.O.O., A.H., J.E.S., S.W.P., E.E.S., R.S., B.E., rate of which is mainly controlled by the concentrations of bio- M.W., M.B., and A.B. wrote the paper. available N and phosphorus (P) (19–24). While the sedimentary The authors declare no conflict of interest. δ15N record suggests that N was bioavailable and that diazotrophic This article is a PNAS Direct Submission. molybdenum (Mo)-based nitrogenase dominated N2 fixation in Published under the PNAS license. the Mesoarchean, the record also places a robust minimum age for 1To whom correspondence should be addressed. Email: [email protected]. the occurrence of aerobic N cycling at ∼2.72 Ga in the Neoarchean This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. (e.g., refs. 15 and references therein). Indeed, prominent 1073/pnas.1818762116/-/DCSupplemental. N isotope excursions in the Neoarchean provide evidence for www.pnas.org/cgi/doi/10.1073/pnas.1818762116 PNAS Latest Articles | 1of6 Downloaded by guest on October 2, 2021 13 organic carbon (δ Corg) isotopes, local water-column redox S1). It has mostly high ratios of highly reactive Fe to total Fe (FeHR/ proxies (Fe speciation and Mn concentrations), and elemental FeT), and high Fe/Al ratios (see SI Appendix, Methods for detailed data for shales of the ∼2.95- to 2.85-Ga Mozaan Group, Pongola analytical techniques). Sequence II was deposited in a deep subtidal Supergroup, South Africa (SI Appendix includes geologic setting but above fair-weather wave base setting and shows a progressive and all data). Our aim is to clarify the factors that controlled the decrease in Mn contents and Mn/Fe, FeHR/FeT, and Fe/Al ratios, nature and development of oxygen oases in the Mesoarchean. whereas Mn and Fe contents are higher than those in average SI Appendix Results and Discussion Pongola shales (ref. 33 and ,TableS1). The uppermost sequence III represents deepening to between fair-weather and Water Column Redox Reconstruction. Our samples span from a shallow- storm wave base and is characterized by persistently low Mn contents marine (above wave base) depositional setting in the White Mfolozi Inlier to a deeper-water (below wave base) equivalent in the Non- andMn/Fe,FeHR/FeT,andFe/Alratios,withMnandFecontents SI Appendix goma area and comprise three sequences deposited at different similar to those in average Pongola shales (ref. 33 and , water depths (Fig. 1 and SI Appendix, Figs. S1 and S2 and Table S1). Table S1). In the more distal, deeper-water setting of the Nongoma In the White Mfolozi Inlier, sequence I, deposited in the most area, where distinct compositional trends were not observed, Fe/Al proximal, intertidal to shallow subtidal setting, is characterized by ratios tend to be high, but Mn contents remain low and Mn/Fe ratios high Mn contents and Mn/Fe ratios compared with average values shift to values lower than the average for shales of the Pongola for shales of the Pongola Supergroup (ref. 33 and SI Appendix,Table Supergroup (ref. 33 and SI Appendix,TableS1). Fig. 1. Geochemical data for shale samples plotted along the lithostratigraphic columns of the studied sections of the Mozaan Group (Pongola basin) from the shallow part of the Ntombe Formation in the White Mfolozi Inlier (A), and its deeper-water equivalent in the Nongoma area (B) (see SI Appendix, Fig. S2 for details). Sequences are defined based on water-depth indicators and chemostratigraphic data.