Woods Hole Oceanographic Institution

Phoebe J Lam

»Inverse methods for particle dynamics
»The speciation of marine particulate iron
»The dynamic biological pump
»The continental margin is a key source of iron
»High Biomass Low Export regimes in the Southern Ocean
»Wintertime phytoplankton bloom in the Subarctic Pacific supported by continental margin iron
»Differential effects of iron additions on organic and inorganic carbon production by phytoplankton

Olivier Marchal, Phoebe J. Lam,

What can paired measurements of Th isotope activity and particle concentration tell us about particle cycling in the ocean?

, Geochimica et Cosmochimica Acta, 90: 126-148, 2012

The ability of paired measurements of thorium isotope activity and particle concentration to constrain rate constants of sorption reactions and particle dynamics in the ocean is examined. This study is motivated by GEOTRACES and other sam- pling programs where Th and particle data are gathered in various oceanic environments. Our approach relies on inversions with a model of trace metal and particle cycling in the water column. First, the model is used to simulate vertical profiles of (i) the activity of three Th isotopes (228,230,234Th) in the dissolved phase, small suspended particles, and large sinking particles, and (ii) the concentration of small and large particles. The simulated profiles are then subsampled and corrupted with noise to generate a pseudo data set. These data are combined with the model with arbitrary values of rate constants of Th adsorption, Th desorption, particle sinking, particle remineralization, and particle (dis)aggregation in an effort to recover the actual values used to generate the data. Inversions are performed using a least-squares technique with varying assumptions about data noise, data sampling, and model errors.

We find that accurate and precise recovery of rate parameters is possible when all data have a relative error of less than 20%, vertical sampling is dense enough to resolve activity and concentration gradients, and model errors are negligible. Esti- mating cycling rates from data with larger errors and (or) at locations where model assumptions are not tenable would remain challenging. On the other hand, the paired data set would improve significantly the relative precision of rate parameters com- pared to that of prior estimates (P100%), even with current data uncertainties and significant model errors. Based on these results, we advocate the joint measurement of all three Th isotopes, 228Ra, and particles collected by in situ filtration within GEOTRACES and other sampling programs targeted at the study of particle processes in the ocean.

FILE » revised_full_manuscript_Marchal_and_Lam.pdf
Final, revised manuscript, not copyedited.

© Woods Hole Oceanographic Institution
All rights reserved