Water Mass Structure and Bottom Water Formation in the Ice-Age Southern Ocean

DESCRIPTION: Scientists established more than 30 years ago that the climate-related variability of carbon dioxide levels in the atmosphere over Earth's ice-age cycles was regulated by the ocean. Hypotheses to explain how the ocean regulates atmospheric carbon dioxide have long been debated, but they have proven to be difficult to test. Work proposed here will test one leading hypothesis, specifically that the ocean experienced greater density stratification during the ice ages. That is, with greater stratification during the ice ages and slower replacement of deep water by cold dense water formed near the poles, the deep ocean would have held more carbon dioxide, which is produced by biological respiration of the organic carbon that constantly rains to the abyss in the form of dead organisms and organic debris that sink from the sunlit surface ocean. To test this hypothesis, the degree of ocean stratification during the last ice age and the rate of deep-water replacement will be constrained by comparing the radiocarbon ages of organisms that grew in the surface ocean and at the sea floor within a critical region around Antarctica, where most of the replacement of deep waters occurs. Completing this work will contribute toward improved models of future climate change. Climate scientists rely on models to estimate the amount of fossil fuel carbon dioxide that will be absorbed by the ocean in the future. Currently the ocean absorbs about 25% of the carbon dioxide produced by burning fossil fuels. Most of this carbon is absorbed in the Southern Ocean (the region around Antarctica). How this will change in the future is poorly known. Models have difficulty representing physical conditions in the Southern Ocean accurately, thereby adding substantial uncertainty to projections of future ocean uptake of carbon dioxide. Results of the proposed study will provide a benchmark to test the ability of models to simulate ocean processes under climate conditions distinctly different from those that occur today, ultimately leading to improvement of the models and to more reliable projections of future absorption of carbon dioxide by the ocean.

OUTCOMES: In the cores with the highest accumulation rate we found clear evidence for a change in redox status of the sediments at the end of the last ice age approximately 20,000 years ago. Specifically, we found a large peak in the manganese concentration of sediments deposited just after the end of the last ice age. We have combined these results with unpublished results from the equatorial Pacific, together with published results from cores collected along the margin of Antarctica, to show that the deep water of entire the Pacific Ocean and Southern Ocean had much lower oxygen concentrations during the Pleistocene ice ages than is the case today. Lower oxygen concentrations translate directly into increased storage of carbon dioxide. These findings are consistent with a growing body of evidence for greater storage of carbon in the deep sea during the Pleistocene ice ages, thereby explaining the remarkable correlation between atmospheric carbon dioxide and temperature of Antarctica over the last 800,000 years that has been extracted from ice cores.  In addition, Columbia undergraduate Kelly-Marie Powell used our cores to test the hypothesis that iron supplied to the iron-starved waters of the Southern Ocean by debris released from melting ice bergs (ice-rafted debris, IRD) would stimulate the growth of phytoplankton and thereby increase the amount of carbon sequestered in the deep sea.  She unambiguously falsified this hypothesis by showing that there is no correlation between the abundance of IRD and indicators of biological productivity.  From the analysis of seawater samples we found that lateral transport by isopycnal mixing is a large term in the mass budgets of 230Th and 231Pa in the Southern Ocean.  This discovery will require that certain conclusions about climate-related changes in ocean circulation based on sedimentary 231Pa/230Th ratios, relying on the assumption that net lateral redistribution of 231Pa occurs primarily or even entirely by advection, be reinterpreted. Under circumstances where is the mass budget of 230Th is affected by the lateral transport, as was found for the Southern Ocean (see above), the use of 230Th as a constant flux proxy for determining sediment accumulation rate has been thought to be compromised.   However, PhD student Frankie Pavia found a way to correct for the lateral transport of 230Th and thereby calculate thorium normalized fluxes of dust to this region of the Southern Ocean where dust supply is extremely low and, consequently, iron limitation is thought to regulate the efficiency of the biological carbon pump.