Abstract
Cadmium (Cd) in the oceans closely mimics the behavior of the macronutrient phosphate (PO43-) and can be used in the enzyme carbonic anhydrase (CA), suggesting a biological uptake of Cd. This relationship between Cd and PO43- has been used extensively as a paleoproxy for historic nutrient cycling. However, the validity of this proxy is questionable due to the complexity of the Cd /PO43- relationship. To this end, Cd isotopic studies can provide critical insight into the mechanism controlling Cd uptake and may, in itself, be a useful paleoproxy for historic primary production.
In this study, multiple collector inductively coupled plasma mass spectrometry (MC-ICPMS), combined with double spiking techniques, was used to determine Cd isotopic compositions in surface waters from the subtropical convergence and Southern Ocean phytoplankton cultures. These analytical techniques were improved to increase the precision on low (<5 ng) Cd samples resolving small differences in natural Cd isotopic variations. These improvements included an investigation into the effects of organic resin-derived matrix on analytical accuracy, as percent-level isotopic shifts were observed in Cd standards analyzed after seawater samples. This study shows that resin-derived organics from Eichrom TRU resin are responsible for the observed anomalous shifts in Cd isotopic standards and that the effects behave in a non-mass dependent way. Oxidation of the samples before analysis eliminates this matrix effect on subsequent standard analyses.
With these improved methods, surface samples collected seasonally from the subtropical convergence and subantarctic were analyzed, showing limited Cd isotopic variability despite the observed 50-fold decrease in dissolved Cd concentration from winter to summer. This decrease was coincident with the drawdown of zinc (Zn) and macronutrients (NO3-, PO43-, Si(OH)4), strongly suggesting that Cd concentrations are dependent on biological activity. Cd was preferentially taken up relative to PO43- resulting in significant seasonal shifts in Cd/PO43- ratios. These results prove that neither Cd/PO43- ratios nor Cd isotopic compositions are, or will be, valid paleoproxies for nutrient cycling for subantarctic surface waters. The lack of Cd isotopic shifts contrasts the limited Cd isotopic data from North Pacific seawater where Cd isotopic compositions follow a kinetic biological uptake model. This indicates that multiple factors control the biological uptake of Cd, which are likely to influence the global Cd/PO43- relationship. Importantly, in regions where Zn and Cd concentrations are both low (<0.2 nmol kg-1) year round, diffusion limitation of both Cd and Zn may be governing biological uptake.
To provide further supporting evidence for this hypothesis, the Cd isotopic composition was also determined, for the first time, in a cultured marine phytoplankton (Proboscia inermis). The effects of Zn and/or Fe limitation on Cd uptake were investigated, and Cd isotopic compositions were used to calculate isotopic fractionation factors. Results showed that marine phytoplankton preferentially remove light isotopes from the medium, and the most fractionated Cd isotopic compositions were observed in Zn-replete samples despite higher cellular Cd concentrations in Zn-limited cultures. This suggests that phytoplankton have at least two different uptake mechanisms with one more highly selective for light isotopes than the other and that these mechanisms are dependent on Zn concentrations.