The Late Archean biosphere: Implications of organic and inorganic geochemistry of marine shales and terrestrial paleosols

Open Access
Watanabe, Yumiko
Graduate Program:
Doctor of Philosophy
Document Type:
Date of Defense:
July 02, 2002
Committee Members:
  • Blair Hedges, Committee Member
  • Hiroshi Ohmoto, Committee Chair
  • Katherine Haines Freeman, Committee Member
  • Michael Allan Arthur, Committee Member
  • geochemistry
  • biosphere
  • organic matter
  • shale
  • Archean
  • paleosol
A very large number (>2,000) of data exist on the carbon isotope ratios of organic carbon in Precambrian sedimentary rocks, displaying a very range of –60 to –10 ‰ for the 2.8-2.6 Ga sedimentary rocks. These data have led many researchers to suggest the development of a globally methane-rich atmosphere-hydrosphere-biosphere system during this period, since the dcarbon isotope ratios less than about –35 ‰ thought to require methane cycling. However, other researchers have argued the development of methane-rich environments in the Precambrian, as well as the Phanerozoic, were only local phenomena. The disputes have continued primarily because most previous investigations have provided very little information on the nature (e.g., occurrences) of organic matter they have studied or on the geological, petrological, mineralogical, and geochemical characteristics of the rocks that host the organic matter. The goal of this study is to develop a better understanding of the nature (e.g., the types, abundances, and distributions) of organisms and the connections between the biosphere and Earth’s surface environments, including the redox state of the oceans and atmosphere during the late Archean. This goal has been approached through multidimensional investigations on two different types of organic carbon-rich rocks that are almost the same age (~2.7-2.6 Ga). The first set of samples includes the 2.7 Ga shales from the Kidd Creek area in the Abitibi district, Canada that were deposited under the influence of submarine hydrothermal activity in a deep (>1.5 km) ocean. The second set of samples are ~2.6 Ga soils from the Schagen area, East Transvaal district, South Africa that formed on land under a semi-arid climate. We have investigated the various characteristics (occurrence, chemical and isotopic composition) of the organic matter and its host rocks utilizing a variety of petrological, mineralogical, and geochemical techniques. The 40 shale samples from the Kidd Creek area we investigated were collected from five shale units (each 2- 50 m thick) that deposited in small depressions (< 2km diameter) within the period 2717-2690 Ma. We investigated the petrographical and mineralogical characteristics (especially the modes of occurrence of organic matter and sulfide minerals), the contents of 60 elements (major elements, heavy metals, and REEs), the contents of C, H, N and S in the bulk rock and extracted kerogen, the d13C value of extracted kerogen, the d13C and d18O values of carbonates, and the d34S value of sulfides. The d13C values of kerogen range from –38.8 to –17.0‰, and the organic carbon contents range from 0.03 to 11.63 wt%. Well-defined positive correlations have been recognized among the following parameters: (a) the d13C of kerogen, (b) the organic C content of shale, (c) the occurrence mode of organic matter in shale (e.g., dissemination, seams), (d) the C/P ratio of shale, (e) the Zn content of shale, and (f) the d13C of carbonate. We also compared the relationships between parameters (a) and (b) in shales from other parts of the 2.7 Ga greenstone belts in the Superior Province, Canada. Quantitative analyses of the correlations among the above parameters have led us to present the following suggestions for the nature of the biosphere, hydrosphere, and atmosphere 2.7 Ga ago: (1) the dominant primary producers in the oceans were most likely cyanobacteria with d13C values around –28 ‰; (2) the kinetic carbon isotopic effect accompanying the photosynthesis, DCO2-CH2O value, was about 8 ‰ greater than today due to higher pCO2 in the Archean atmosphere; (3) the normal, open oceans were oxygenated throughout the water column, thereby preventing the accumulation of organic carbon- and sulfide-rich sediments; (4) submarine depressions with anoxic bottom water were created locally and regionally in many parts of the oceans, especially in regions with active tectonics and volcanism; (5) many of the submarine depressions became sites of hydrothermal fluid discharge; (6) the plumes of hydrothermal fluids enhanced the recycling of phosphate between the anoxic bottom water and the oxic surface water, causing an increased activity of cyanobacteria, while an increased flux of the remnants of cyanobacteria in depressions caused an increased activity of anaerobic, heterotrophic, and mat-forming microbial communities; (7) the submarine microbial communities developed in depressions under the influence of low temperature, Zn-rich hydrothermal fluids were composed primarily of fermentative bacteria, methanogens, and sulfate-reducing bacteria with d13C values between –55 and –35 ‰; and (8) the submarine microbial communities developed in depressions under the influence of high temperature, Cu+Zn-rich hydrothermal fluids were inhabited primarily of thermophiles with d13C values between –20 and –10‰. These suggestions further lead to the following proposals: (9) contrary to the current popular theory, the productions of methane and organic matter with very low d13C values (<-35‰) did not occur globally in the Archean oceans; they were restricted to local (and regional) basins under the influence of low temperature hydrothermal fluids; and (10) because the O2 content of the bottom ocean water is dependent on the atmospheric pO2 level, suggestion (3) suggests that the pO2 level 2.7 Ga was already greater than ~0.5 PAL. The 2.6 Ga paleosols at Schagen, which developed on serpentinite, are characterized by relatively high contents of reduced-C (0.1 – 1.4 wt %) in the upper ~8 m section where thick (1-3 m) zones of massive carbonate (Fe-rich dolomite and calcite) are interbedded with 1-3 m-thick zones rich in clays (talc, chlorite, and ferri-stilpnomelane). We have carried out a systematic investigation of the petrographical, mineralogical, and geochemical characteristics of the carbonaceous matter and its host rocks on 34 samples collected from a 17 m profile. Mineralogical investigations have included XRD, EPMA, SEM, and TEM analyses of rocks, minerals, and extracted kerogen. Geochemical investigations have included chemical analyses to identify 60 elements (major, trace, and REE) and Fe3+/Fe2+ ratios, isotope analyses of carbonates (d13C, d18O, 87Sr/86Sr), chemical analyses of the kerogen extracts (C/H/S/N ratios), and carbon isotope analyses of extracted kerogens. Thermodynamic analyses of the formational conditions for the pertinent minerals (serpentine, talc, dolomite, calcite, and siderite) were also performed. The carbonaceous matter in the soils is intimately associated with clays (talc, chlorite, and ferri-stilpnomelane) and occurs mostly as seams (20 to 100 mm thick and a few cm long) that parallel the soil layers. These occurrences, along with the crystallographic structure (amorphous graphite), H/C ratios (0.02-0.05), and d13C values (-17.4 to –14.4 ‰) of the carbonaceous matter, suggest remnants of in-situ microbial mats developed during soil formation. After correcting for metamorphic effects, the original d13C values of the microbial mats are estimated to be around –18 ‰. The microbial mats, which may have been mostly composed of cyanobacteria, probably formed on the surface of silicate-rich soils under air. However, the microbial mats in the carbonate-rich soil zones probably formed at the bottom of an evaporitic, alkaline, and anoxic pond (or lake) and were probably composed mostly of anaerobic heterotrophs that utilized the remnants of cyanobacteria from the oxygenated surface water. The results of various investigations suggest the ~8m carbonaceous matter-rich soil section primarily grew upward from the original serpentinite surface through a combination of two major processes. The processes were (i) the precipitation of carbonates (mostly Fe-rich dolomite) by locally discharged groundwater and (ii) the accumulation of silicate-rich rock fragments (and minerals) transported from local sources (serpentinite and granite) by fluvial and/or aerial processes. The relative importance of these two processes fluctuated during soil formation due to changes in groundwater flow regimes. The groundwater probably originated from a granite-gneiss terrain and discharged in the serpentinite area, thereby creating an evaporating, alkaline pond where Fe-rich dolomite precipitated; the Fe-rich carbonates suggest the bottom water was reduced. During periods when groundwater was not discharging, the silicate-rich rock fragments from granitic sources were converted to clay-rich soils by the local rainwater. The enrichment of iron as ferric-bearing aluminous silicates (stilpnomelane) at the uppermost part of the soil section suggests this silicate-rich soil formation occurred under an oxygenated atmosphere, rather than under a water body. All results obtained from the above two types of research are consistent with an atmospheric evolution model that postulates an early (>2.7 Ga) development of an oxygenated atmosphere and complex biosphere in the oceans and on land.