Significance of ⁵⁶Fe depletions in late-Archean shales and pyrite

Sedimentary rocks and minerals formed during the final two-hundred million years of the Archean Eon (2.7 to 2.5 billion years ago, or Ga) are more depleted in ⁵⁶Fe than at any other time in Earth's past. Three hypotheses are proposed to explain these ⁵⁶Fe depletions: (1) a very negative late-Archean seawater δ⁵⁶Fe value, (2) “shuttling” of isotopically light Fe across the chemocline in redox-stratified settings, and (3) pyrite formation in an Fe(II)-rich ocean. Each of these scenarios has different implications for the initial oxidation of Earth's surface, the climax of which – the Great Oxidation Event – immediately post-dates the appearance of these ⁵⁶Fe depletions in the rock record. To help inform this debate, we measured the Fe isotope ratios of 120 shale and pyrite samples from Western Australia (Mt. McRae Shale and Jeerinah Formation) and South Africa (Klein Naute Formation) deposited between ∼2.65 Ga and ∼2.50 Ga. As in previous studies, we also find very strong sedimentary ⁵⁶Fe depletions, to as low as δ⁵⁶Fe = −2.06 ± 0.08‰ in bulk shales and δ⁵⁶Fe = −2.31 ± 0.08‰ in pyrite. Some, but not all, of the severest ⁵⁶Fe depletions appear alongside evidence of an Fe shuttle and local pyrite formation. These processes need not be mutually exclusive, and some combination of them likely played a partial, probably faciliatory role in driving some strong ⁵⁶Fe depletions in our dataset. Most interestingly, and with little exception, the severest ⁵⁶Fe depletions appear in samples deposited farther from shore under H₂S-rich and anoxic (“euxinic”) conditions. We find it difficult to explain this connection without invoking the persistent presence of a very negative global seawater δ⁵⁶Fe value during the latest Archean, one that was most consistently captured in sediments formed in distal euxinic settings. In order to impart this isotopic effect on seawater, the global seawater Fe(II) reservoir needed to have been partially oxidized during at least the final few hundreds of millions of years leading up to the Great Oxidation Event. Our new data add support to the idea that Earth's initial oxidation was a long and protracted process rather than a rapid event.