INVITED: WHEN DID OXYGEN APPEAR IN THE ATMOSPHERE?

Signatures of mass-independent fractionation (MIF) of the sulfur isotopes in sedimentary sulfide and sulfate minerals of Archean age suggest that the Archean atmosphere was essentially devoid of oxygen.  The value of Δ³³S is a measure of the degree of MIF. Pyrites in the Mt. McRae Shale, in the Hamersley Basin, Western Australia yield Δ³³S values from –2.0 to +6.9‰. This large variation is consistent with at most a few ppm O₂ in the atmosphere during the deposition of the shale between 2479 ±3 and 2561 ±8 Ma. Sulfur in pyrite of the overlying Brockman Iron Formation, whose minimum age is 2470 Ma, has a Δ³³S range of –1.3‰ to +0.1‰.  This range of Δ³³S is well outside the range of 0.0 to +0.3 ‰ that is measured at the Geophysical Laboratory for a variety of biogenic and hydrothermal sulfide and sulfate minerals formed at times of abundant atmospheric O₂.  It is likely, therefore, that the level of atmospheric O₂ rose above the ppm level after 2470 Ma. Sulfur in pyrite from the 2316 ±7 Ma Timeball Hill and Rooihogte formations in South Africa show very little evidence of contemporary MIF.  This is consistent with a rise of atmospheric O₂ above the ppm level between 2470 and 2316 Ma.

Many Archean sulfides and sulfates have a considerable range of Δ³³S values.  Some of these values are close to 0 ‰.  This may not imply the presence of atmospheric O₂ because in all cases studied to date contemporary sulfides and/or sulfates with Δ³³S values indicative of MIF are also present.  The available data therefore suggest that O₂ levels in the atmosphere were no more than a few ppm during the Archean and rose significantly between 2470 and 2316 Ma.

This conclusion is consistent with that drawn from other lines of evidence.  Paleosols developed on basalts during the late Archean and on greenstones at the erosion surface below the Huronian Matinenda formation suggest that atmospheric O₂ levels were less than ca.10^-3.5 atm.  The Pronto paleosol, developed on the same erosion surface, may indicate the presence of small amounts of atmospheric O₂ during its development.  The ages of these paleosols is not well known but is probably close to 2450 Ma.  The Pronto paleosol may therefore indicate that atmospheric O₂ rose shortly after the last large MIF signal at 2470 Ma.