Iron supply and demand in an Antarctic shelf ecosystem

© The Author(s), 2015. This article is distributed under the terms of the Creative Commons Attribution License. The definitive version was published in Geophysical Research Letters 42 (2015): 8088–8097, doi:10.1002/2015GL065727. The Ross Sea sustains a rich ecosystem and is the most productive secto...

Full description

Bibliographic Details
Published in:Geophysical Research Letters
Main Authors: McGillicuddy, Dennis J., Sedwick, Peter N., Dinniman, M. S., Arrigo, Kevin R., Bibby, Thomas S., Greenan, Blair J. W., Hofmann, Eileen E., Klinck, John M., Smith, Walker O., Mack, Stefanie L., Marsay, Christopher M., Sohst, Bettina M., van Dijken, Gert L.
Format: Article in Journal/Newspaper
Language:English
Published: John Wiley & Sons 2015
Subjects:
Online Access:https://hdl.handle.net/1912/7657
Description
Summary:© The Author(s), 2015. This article is distributed under the terms of the Creative Commons Attribution License. The definitive version was published in Geophysical Research Letters 42 (2015): 8088–8097, doi:10.1002/2015GL065727. The Ross Sea sustains a rich ecosystem and is the most productive sector of the Southern Ocean. Most of this production occurs within a polynya during the November–February period, when the availability of dissolved iron (dFe) is thought to exert the major control on phytoplankton growth. Here we combine new data on the distribution of dFe, high-resolution model simulations of ice melt and regional circulation, and satellite-based estimates of primary production to quantify iron supply and demand over the Ross Sea continental shelf. Our analysis suggests that the largest sources of dFe to the euphotic zone are wintertime mixing and melting sea ice, with a lesser input from intrusions of Circumpolar Deep Water and a small amount from melting glacial ice. Together these sources are in approximate balance with the annual biological dFe demand inferred from satellite-based productivity algorithms, although both the supply and demand estimates have large uncertainties.