Evidence for hot early oceans? (Reply)

Abstract

Robert & Chaussidon The possibility of a change in δ¹⁸O values of the oceans has been discussed for the past 30 years¹ and the comment by Shields and Kasting² does not really bring any new insight into this issue. The carbonate δ¹⁸O curve³ is indistinguishable from that of cherts and shows a huge scatter caused by local geological processes. This scatter in δ¹⁸O (up to 20‰ at a given age) far exceeds the sensitivity of the isotope thermometer (about 2‰ per 10 °C). It therefore prevents any precise additional test of the model and no consensus was ever reached in the literature on this debate from oxygen-isotope data alone. Our approach⁴, by contrast, was to try to find another isotopic proxy (δ³⁰Si) that could bring new and independent constraints to test the hypothesis of hot oceans in the early Precambrian. Our interpretation of the δ³⁰Si values in terms of temperature is fully coherent not only with the δ¹⁸O of cherts but also with that of marine carbonates, and certainly does not overlook these data. In addition, we indicated⁴ that our sample set has not the time resolution that would be necessary to look for an effect on silicon isotopes of Precambrian global glaciations. No chert samples from such periods were analysed.