Downloads
Cite as
F. A. Haumann (2016): Southern Ocean response to recent changes in surface freshwater fluxes. Doctoral thesis. ETH Zurich.
doi:10.3929/ethz-b-000166276.
Summary
Earth’s climate bears a close relation to the vertical exchange of water masses in the global ocean.
This relation originates from the transport of heat, carbon, and nutrients with the subduction and
upwelling of water masses. The majority of this vertical exchange occurs in the Southern Ocean,
where carbon- and nutrient-rich deep waters re-surface. To date, the mechanisms, which control
long-term changes in this upwelling, are not firmly established. Both changes in either surface
winds or surface freshwater fluxes could hypothetically alter the upwelling. Strong meridional
gradients in the surface wind field propel a surface divergence that pulls waters from the deep.
This process is facilitated by a marginally stable vertical density stratification, which is mainly set
up by salinity and therefore by surface freshwater fluxes. So far, the exploration of stratification
changes has been limited by the availability of observational freshwater flux data and by their
poor representation in climate models. In this dissertation, I investigate recent changes in surface
freshwater fluxes and their effect on the hydrography, circulation, as well as the vertical exchange
of heat and carbon in the Southern Ocean. A particular focus of this thesis is on yet unconstrained
freshwater fluxes which originate from the seasonal formation and melting of sea ice.
In the first part of this thesis, I provide the first comprehensive data set of annual freshwater fluxes arising from the formation, transport, and melting of sea ice in the Southern Ocean over the time period from 1982 to 2008. For this purpose, I combine numerous satellite, in-situ, and reanalysis data of sea-ice concentration, thickness, and drift. The resulting freshwater flux estimates reveal that 410±110 mSv (1 mSv = 103 m3 s−1) of freshwater are removed and added to the surface waters by the formation and melting of sea ice each year. Compared to the available data of the atmospheric and land-ice freshwater fluxes, sea ice provides the dominant freshwater flux in the seasonally ice-covered region of the Southern Ocean. Most of the sea ice forms in the coastal ocean; and 130±30 mSv of this sea ice are transported towards the sea-ice edge, where it melts. Thereby, it increases the salinity in coastal and bottom waters and lowers the open-ocean surface and intermediate water salinity. This northward transport of freshwater has increased by 20±10% over the observational period, which corresponds to a freshening of −0.02±0.01 g kg−1 per decade in the open-ocean waters. From this analysis, I conclude that the increased northward transport of freshwater by sea ice explains the majority of the observed freshening of the open-ocean waters. In fact, ocean salinity data shows that the largest freshening signal coincides with the region of largest increased sea-ice melting in the Pacific sector.
In order to better understand the ocean’s response to changing freshwater fluxes, I am using the newly derived sea-ice–ocean freshwater fluxes to constrain a regional ocean circulation model in the second part of this thesis. By perturbing the model with the observed recent changes in these fluxes, I find that the increased northward sea-ice transport could also be responsible for most of the observed surface cooling in the Southern Ocean that occurred over recent decades, despite global warming. The model simulations suggest that this cooling originates from a freshening and enhanced surface density stratification in the upwelling region that delays and shoals the deep winter-time mixing. As a consequence, the total heat capacity of the mixed layer decreases and less warm deep water enters the surface layer, which results in a surface cooling and sub-surface warming. Moreover, about 25% less carbon-dioxide (CO2) is released to the atmosphere from the upwelling region. This response can be explained, on the one hand, by the surface cooling that increases the solubility of CO2 in seawater and, on the other hand, by reduction in upwelling. The reduced CO2 release is opposed by a reduced subduction of CO2 into Antarctic Intermediate Water and Subantarctic Mode Water due to the increasing stratification. These findings suggest that the increased surface density stratification could explain why the Southern Ocean carbon sink has not saturated but has rather strengthened over recent decades, despite increasing surface winds.
In conclusion, the insights gained from my dissertation point towards much higher sensitivity of the upwelling in the Southern Ocean to changes in the sea-ice freshwater fluxes than previously assumed. My combined analyses of observational data and model experiments elucidate that sea ice effectively re-shovels freshwater from the lower overturning cell to the upper overturning cell and thereby increases the meridional and vertical salinity and density gradients in the Southern Ocean. In the long-term, changes in this system could alter the atmospheric CO2 concentration and therefore the global climate. A potential warming of the sea-ice region in the future could reverse the changes observed over recent decades and enhance the release of CO2 from the ocean to the atmosphere, which might amplify global warming. In contrast, in colder glacial climates, increased sea-ice formation could reduce the upwelling of carbon-rich waters to the surface ocean and lower the atmospheric CO2 concentrations. This interpretation of my results is in line with the hypothesis that past glacial–interglacial variations in the global atmospheric CO2 concentration could originate from changes in the Southern Ocean density stratification.
In the first part of this thesis, I provide the first comprehensive data set of annual freshwater fluxes arising from the formation, transport, and melting of sea ice in the Southern Ocean over the time period from 1982 to 2008. For this purpose, I combine numerous satellite, in-situ, and reanalysis data of sea-ice concentration, thickness, and drift. The resulting freshwater flux estimates reveal that 410±110 mSv (1 mSv = 103 m3 s−1) of freshwater are removed and added to the surface waters by the formation and melting of sea ice each year. Compared to the available data of the atmospheric and land-ice freshwater fluxes, sea ice provides the dominant freshwater flux in the seasonally ice-covered region of the Southern Ocean. Most of the sea ice forms in the coastal ocean; and 130±30 mSv of this sea ice are transported towards the sea-ice edge, where it melts. Thereby, it increases the salinity in coastal and bottom waters and lowers the open-ocean surface and intermediate water salinity. This northward transport of freshwater has increased by 20±10% over the observational period, which corresponds to a freshening of −0.02±0.01 g kg−1 per decade in the open-ocean waters. From this analysis, I conclude that the increased northward transport of freshwater by sea ice explains the majority of the observed freshening of the open-ocean waters. In fact, ocean salinity data shows that the largest freshening signal coincides with the region of largest increased sea-ice melting in the Pacific sector.
In order to better understand the ocean’s response to changing freshwater fluxes, I am using the newly derived sea-ice–ocean freshwater fluxes to constrain a regional ocean circulation model in the second part of this thesis. By perturbing the model with the observed recent changes in these fluxes, I find that the increased northward sea-ice transport could also be responsible for most of the observed surface cooling in the Southern Ocean that occurred over recent decades, despite global warming. The model simulations suggest that this cooling originates from a freshening and enhanced surface density stratification in the upwelling region that delays and shoals the deep winter-time mixing. As a consequence, the total heat capacity of the mixed layer decreases and less warm deep water enters the surface layer, which results in a surface cooling and sub-surface warming. Moreover, about 25% less carbon-dioxide (CO2) is released to the atmosphere from the upwelling region. This response can be explained, on the one hand, by the surface cooling that increases the solubility of CO2 in seawater and, on the other hand, by reduction in upwelling. The reduced CO2 release is opposed by a reduced subduction of CO2 into Antarctic Intermediate Water and Subantarctic Mode Water due to the increasing stratification. These findings suggest that the increased surface density stratification could explain why the Southern Ocean carbon sink has not saturated but has rather strengthened over recent decades, despite increasing surface winds.
In conclusion, the insights gained from my dissertation point towards much higher sensitivity of the upwelling in the Southern Ocean to changes in the sea-ice freshwater fluxes than previously assumed. My combined analyses of observational data and model experiments elucidate that sea ice effectively re-shovels freshwater from the lower overturning cell to the upper overturning cell and thereby increases the meridional and vertical salinity and density gradients in the Southern Ocean. In the long-term, changes in this system could alter the atmospheric CO2 concentration and therefore the global climate. A potential warming of the sea-ice region in the future could reverse the changes observed over recent decades and enhance the release of CO2 from the ocean to the atmosphere, which might amplify global warming. In contrast, in colder glacial climates, increased sea-ice formation could reduce the upwelling of carbon-rich waters to the surface ocean and lower the atmospheric CO2 concentrations. This interpretation of my results is in line with the hypothesis that past glacial–interglacial variations in the global atmospheric CO2 concentration could originate from changes in the Southern Ocean density stratification.
Copyright © 2016, F. A. Haumann, ETH Zurich, Switzerland