Presenter: Perrin Davidson (University of Chicago)

Description:
The biological carbon pump (BCP) performs a significant role in mitigating increasing anthropogenic CO₂ emissions and consequent heating effects. Quantifying the contribution of the BCP to carbon sequestration has been the topic of several recent studies, with the average yearly particulate organic carbon (POC) flux out of the upper ocean ranging from 4 to 12 PgC year^-1. Here we use the commonly employed Thorium-234 (²³⁴Th) proxy to determine POC fluxes throughout the world's oceans. The disequilibrium between ²³⁴Th (half-life ~24.1 days), a highly particle-reactive radionuclide, with its parent, Uranium-238 (²³⁸U, half-life ~4.5 x 10⁹ years), can be used to quantify ²³⁴Th fluxes out of the surface ocean. Along with the particulate ratio of ²³⁴Th with POC (²³⁴Th:POC), these measurements have been used to estimate POC fluxes throughout the water column. Using a global database, over 60,000 global measurements of ²³⁴Th, ²³⁸U, and ²³⁴Th:POC were regressed to a 2.8-degree grid using a suite of Machine Learning (ML) and Minimum Variance algorithms. This gridded data was then used to drive a 3D global model of POC fluxes out of the regionally varying Euphotic Zone (EZ) using a set of sparse, implicit and explicit transport matrices derived from a global, coupled General Circulation Model.  Our model captures known major features of ²³⁴Th, with an average of 2.03±0.18 dpm L^-1 globally and greatest activities in ocean gyres. The model estimates an average of ~7.4 PgC year^-1 of POC out of the EZ, with an average flux of 52±35 mgC m^-2 day^-1 and greatest fluxes in the coastal, equatorial, and extreme latitude regions. Furthermore, our model predicts that the Pacific Ocean contributes to ~60% of the annual flux, with ~4.2 PgC year^-1, followed by the Atlantic Ocean, with a ~30% contribution of ~2.2 PgC year^-1.