Late Quaternary Changes in the Westerly Winds over the Southern Ocean: the sub-Antarctic peatland record and response

Abstract

The Southern Ocean plays a crucial role in modulating global atmospheric CO2 concentrations by acting as a large carbon sink. Upwelling driven by surface winds brings sequestered carbon from deep-water reservoirs to the ocean surface where it can saturate the potential sink and de-gas directly to the atmosphere. Specifically, the intensity and latitudinal position of the Southern Hemisphere Westerly Winds (SHWW) alters the magnitude of upwelling, and influences whether the Southern Ocean acts as a net sink or source of atmospheric CO2. In recent decades the SHWW have intensified and migrated further toward Antarctica, weakening the carbon sink. Over longer time-scales, the same mechanism has been implicated to explain the increase in atmospheric CO2 during the last deglaciation. Unfortunately, our ability to predict future changes in the SHWW is limited. Most existing knowledge of past wind-behaviour is based on archives located outside the core wind-belt, and hence spatial resolution in many areas of the Southern Ocean is poor. Consequently, more palaeo-data from inside the core wind-belt is needed to provide boundary conditions for earth system models. One such opportunity is provided by peatland ecosystems on the sub-Antarctic islands. Peatlands on exposed west-facing coasts receive oceanic base cations in concentrations proportional to the strength of the prevailing westerly winds. Reconstruction of bog-surface salinity could provide a direct proxy for past changes in wind behaviour, but methods to suitably extract the signal are lacking. In this thesis, we propose and test the idea that testate amoebae could record bog-surface salinity conditions. Using a land-sea transect on Marion Island (sub-Antarctica) we examine the ecology of these microorganisms and test their response to the contemporary conductivity gradient. We find that conductivity explained the most variance in community data, and show the potential of our novel proxy for palaeowind reconstruction in the wider sub-Antarctic region.