Diatom silicon isotopes as a proxy for silicic acid utilisation: A Southern Ocean core top calibration

Despite a growing body of work that uses diatom δ30Si to reconstruct past changes in silicic acid utilisation, few studies have focused on calibrating core top data with modern oceanographic conditions. In this study, a microfiltration technique is used to divide Southern Ocean core top silica into narrow size ranges, separating components such as radiolaria, sponge spicules and clay minerals from diatoms. Silicon isotope analysis of these components demonstrates that inclusion of small amounts of non-diatom material can significantly offset the measured from the true diatom δ30Si. Once the correct size fraction is selected (generally 2–20 μm), diatom δ30Si shows a strong negative correlation with surface water silicic acid concentration (R2 = 0.92), highly supportive of the qualitative use of diatom δ30Si as a proxy for silicic acid utilisation. The core top diatom δ30Si matches well with mixed layer filtered diatom δ30Si from published in situ studies, suggesting little to no effect of either dissolution on export through the water column, or early diagenesis, on diatom δ30Si in sediments from the Southern Ocean. However, the core top diatom δ30Si shows a poor fit to simple Rayleigh or steady state models of the Southern Ocean when a single source term is used. The data can instead be described by these models only when variations in the initial conditions of upwelled silicic acid concentration and δ30Si are taken into account, a caveat which may introduce some error into quantitative reconstructions of past silicic acid utilisation from diatom δ30Si.