Fe isotopes of a 2.4 Ga hematite-rich IF constrain marine redox conditions around the GOE

Abstract

The hematite- and manganese-rich Hotazel iron formation, Griqualand West basin, South Africa, was deposited at a key moment in time, close to the GOE between 2.4 and 2.3 Ga. It stratigraphically overlies the Ongeluk Formation, comprising thick flood basalts, which in turn interfinger with and cover the Makganyene Formation diamictites, the inferred remnants of the first Paleoproterozoic Snowball Earth interval. No extensive research has been conducted to date on the basal part of the Hotazel Formation due to poor exposure, though it constitutes an important link between a period of large-scale ice cover, extensive volcanism and the onset of atmospheric oxygenation. Here, we present a detailed petrographic, geochemical and Fe isotope study of a roughly 3-metre-long drill-core exposing the Ongeluk to Hotazel contact. Our results show that after the cessation of Ongeluk volcanism, primary precipitation of Fe(III) oxyhydroxides from the photic surface zone of the original basin became the dominant sedimentation mechanism. Negative δ56Fe values (between −0.26 and −0.50‰) in micro-drilled hematite-rich chert indicate that surface water δ56Fe compositions at the time of deposition were depleted. Yet, δ56Fe and bulk-rock Fe/Mn values are still substantially higher (1–2‰) than those reported higher up in the Mn-rich layers of the Hotazel sequence, suggesting that redox potentials were still comparatively limited during the earliest stages of the Hotazel depositional environment. The base of the Hotazel Formation thus forms a transitional interval between precipitation from essentially ferruginous seawater, as recorded in the Ghaap Group BIFs, and from isotopically and chemically highly evolved surface waters, as demonstrated by the Hotazel Mn-rich layers. In the absence of Fe(II) as a strong reducing agent, large volumes of photosynthetic oxygen may have eventually escaped into the atmosphere, leading to the onset of atmospheric oxygenation. Our results thus contradict previous models that place the onset of the GOE before the Hotazel Formation, concurrent with and mechanistically linked to the Makganyene and Ongeluk events. Instead, we show that the Hotazel basal sediments are still pre-GOE, consistent with their circa 2.4 Ga age and the continuation of MIF-S higher in the stratigraphic record.