Suspended Particulate Pb from the North Atlantic US GEOTRACES (GA-03) Section can be Estimated from Dissolved Pb Using a Two Phase Partition Coefficient Model

Edward A Boyle, Massachusetts Institute of Technology, Cambridge, MA, United States, Phoebe J Lam, University of California Santa Cruz, Department of Ocean Sciences, Santa Cruz, CA, United States, Daniel Ohnemus, University of Maine, Abigail Emery Noble, Gradient, Cambridge, MA, United States, Thomas M Church, University of Delaware, Newark, DE, United States, Sylvain Rigaud, Universite de Bordeaux; Earth Observatory of Singapore, Nanyang Technological University, Singapore, Singapore, Mark M Baskaran, Wayne State University, Detroit, MI, United States and Gillian Stewart, CUNY Queens College, School of Earth and Environmental Sciences, Flushing, NY, United States
Abstract:
Dissolved and particulate Pb were measured on more than 200 pairs of samples from the US GEOTRACES North Atlantic Transect (GA-03). Dissolved Pb measurements (Noble et al., 2015, DSR2 116:20) were made on samples collected by 0.2 micron Acropak filtration from the Go-Flo GEOTRACES Carousel and particulate Pb measurements (>51 micron and 0.8-51 micron, Ohnemus et al., 2015, DSR2 116:283) were made on filters collected by modified McLane pumps. Dissolved and particulate Pb-210 data was also obtained at six “super-stations” (Rigaud et al., 2015, DSR2 116:60). The dissolved Pb data (25-37 samples per station) were interpolated to create paired dissolved-particulate Pb concentrations at the depths of the McLane pumps (16 depths per station). Using these data for samples which include estimates of “MnO2 mass” and “lithogenic mass” (Lam et al., 2015, DSR2 116:303), and excluding four data points from the TAG hydrothermal plume, 90% of the variance in the <51 um particulate Pb data can be accounted for by a two-phase Kd (partition-coefficient) model with two constant Kd values through the entire section. The >51 um data is consistent with the same partition coefficients within the reservation that the much lower >51 um concentrations are close to detection-limit analytical noise. Where Pb-210 data is available, it is also consistent with this model although with apparently slightly lower partition coefficients. When individual stations are used in the regression, there appear to be small Kd differences between the eastern and western basin stations.
Back to: Trace Elements and Isotopes at the Interfaces of the Atlantic Ocean II Posters