Boron isotopic compositions of corals: Seawater or diagenesis record?

Abstract

The first analyses of modern and fossil corals (calcite-free) using a positive ion (Cs 2BO 2 +) TIMS technique and a chemical procedure for B extraction are reported. The δ 11B for modern corals fluctuates between 23.3 and 25.5‰. The range of variation of fossil corals from the last interglacial period is greater, from 23.5 to 27.3‰, and the lowest δ 11B corresponds to the corals close to Last Glacial Maximum. Boron concentrations show a variation from 49 to 58 ppm in the recent corals, and from 39 to 52 ppm in ancient corals. For the ancient corals, linear correlations are observed in both δ 11B vs. B concentration and δ 11B vs. 1/B plots, but these correlations are not observed for modern corals. Two interpretations are presented to account for these observations and to test the potential of δ 11B for reconstructing paleo-ocean pH. The first interpretation assumes a thermodynamic partitioning of B between seawater and carbonates and shows that both boron isotopic compositions and concentrations can be used to calculate the pH of seawater in which corals grow. For our data, such a model requires that both the pH and the ΣCO 2 of the deglacial ocean was lower than present-day values. The second interpretation is based on the correlation observed for ancient corals between boron isotopic compositions and 1/B. We suggest that the measured boron isotopic compositions of ancient corals are caused by the mixing of a primary aragonite (the coral skeleton) and secondary aragonite precipitated from 10B-enriched reef pore fluids. The characteristics (δ 11B and B concentrations) of the diagenetic aragonite and the fluids from which it precipitates can be constrained. If this proves correct, only the non-recrystallised parts of ancient corals can be used for paleo-pH reconstructions. These recrystallisation processes did not affect the 234U/ 238U ratios of the corals.