On the estimation of deep Atlantic ventilation from fossil radiocarbon records. Part II: (in)consistency with modern estimates

Abstract

Measurements of radiocarbon concentration (Δ14C) in fossil biogenic carbonates have been interpreted as reflecting a reduced ventilation of the deep Atlantic during the last ice age. Here we evaluate the (in)consistency of an updated compilation of fossil Δ14C data for the last deglaciation with the abyssal circulation in the modern Atlantic. A Δ14C transport equation, in which the mean velocity field is a modern field estimate and turbulent flux divergence is treated as a random fluctuation, is fitted to deglacial Δ14C records by using recursive weighted least squares. This approach allows us to interpret the records in terms of deviations from the modern flow with due regard for uncertainties in the fossil data, the Δ14C transport equation, and its boundary conditions. We find that the majority of fit residuals could be explained by uncertainties in fossil Δ14C data, for two distinct estimates of the modern flow and of the error variance in the boundary conditions. Thus, most, not all, deglacial data appear consistent with present-day ventilation rates. From 20% to 32% of the residuals exceed in magnitude the published errors in the fossil data by a factor of 2. Residuals below 4000 m in the western North Atlantic are all negative, suggesting that deglacial Δ14C values from this region are too low to be explained by modern ventilation. While deep water ventilation appeared different from today at some locations, a larger database and a better understanding of error (co)variances are needed to make reliable paleoceanographic inferences from fossil Δ14C records.