Margriet L Lantink, Joshua HFL Davies, Rick Hennekam, David McB. Martin, Paul Mason, Gert-Jan Reichart, Frederik J Hilgen

Recent evidence for astronomical-induced cycles in banded iron formations (BIFs) hints at the intriguing possibility of developing astrochronological, i.e. precise time-stratigraphic, frameworks for the earliest Proterozoic as also reconstructed for parts of the Mesozoic and Paleozoic. The ca 2.47-Ga Kuruman Iron Formation (Griqualand West Basin, South Africa) and Dales Gorge Member of the Brockman Iron Formation (Hamersley Basin, Western Australia) are of special interest in this regard, given their inferred temporal overlap and similar long-period eccentricity imprint. This suggests that these two BIFs may be correlated on the basis of their large-scale cycle patterns and using additional radio-isotopic age constraints.
To examine the possibility of establishing such a framework, we generated and analysed several high-resolution proxy records from both drill-core and outcrop, combined with chemical abrasion ID-TIMS U–Pb dating of presumed volcanically sourced zircon. Time-series analysis of these records yields a variety of spectral peaks, of which a prominent ~5 m and ~16 m cycle can be linked to the basic stratigraphic alternations and bundling. New and improved U–Pb ages of the Dales Gorge Member and Kuruman Iron Formation, respectively, indicate a comparable average sedimentation rate of 10–12 m/Myr for both BIF units. Based on this rate, we attribute the ~5 m cycle to the long 405-kyr eccentricity cycle. More tentatively, we interpret the ~16 m cycle as the very long 2.4-Myr eccentricity cycle, having a reduced period of ~1.3 Myr due to chaotic behaviour in the solar system. Other identified cycles (~580 kyr, ~700 kyr and ~1.8 Myr) might be explained in terms of weaker eccentricity components and/or as harmonics and combination tones of these cycles.
An initial attempt to establish cyclostratigraphic correlations between the Kuruman Iron Formation and Dales Gorge Member solely based on their characteristic cycle patterns proved unsuccessful, which may be due to a difference in stratigraphic recording of the astronomical signal between their different depositional environments. Next, we used the U–Pb ages to first constrain correlations at the scale of the ~16 m cycle, followed by a correlation of the basic ~5 m cycles. The resultant framework remains problematic and debatable at the individual 405 kyr cycle-level, and should merely be considered as a starting point for future studies. Particularly, our findings highlight the need for further investigations into how Milankovitch forcing influenced BIF sedimentation and paleoenvironmental conditions at a time when the Earth and solar system behaved fundamentally different from today.