Controls on dissolved and particulate iron distributions in surface waters of the Western Antarctic Peninsula shelf

The Western Antarctic Peninsula (WAP) displays high but variable productivity and is also undergoing rapid change. Long-term studies of phytoplankton communities and primary production have suggested transient limitation by the micronutrient iron (Fe), but to date no data have been available to test this hypothesis. Here, we present the first spatially extensive, multi-year measurements of dissolved and particulate trace metals in surface waters to investigate the key sources and sinks of Fe in the central WAP shelf. Surface samples of dissolved and particulate metals were collected throughout the 700 x 200 km grid of the Palmer Long-Term Ecological Research program in three consecutive austral summers (2010 - 2012). Iron concentrations varied widely. Both dissolved and particulate Fe were high in coastal waters (up to 8 nmol kg(-1) and 42 nmol kg(-1), respectively). In contrast, very low Fe concentrations (< 0.1 nmol kg(-1)) were widespread in mid- to outer-shelf surface waters, especially in the northern half of the sampling grid, suggesting possible Fe limitation of primary production on the shelf. Sea ice and dust inputs of Fe were minor, although their relative importance increased with distance from shore due to the larger near-shore sources. Sedimentary inputs were inferred from manganese distributions; these were more significant in the northern portion of the grid, and showed interannual variation in intensity. Overall, the interannual distribution of Fe was most closely correlated to that of meteoric water (glacial melt and precipitation). Although the Fe concentrations and relative contributions of dissolved and particulate Fe attributed to meltwater were variable throughout the sampling region, increasing glacial meltwater flux can be expected to increase the delivery of Fe to surface waters of the coastal WAP in the future.