Meteorology and oceanography of the Atlantic sector of the Southern Ocean—a review of German achievements from the last decade
- 3.5k Downloads
In the early 1980s, Germany started a new era of modern Antarctic research. The Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research (AWI) was founded and important research platforms such as the German permanent station in Antarctica, today called Neumayer III, and the research icebreaker Polarstern were installed. The research primarily focused on the Atlantic sector of the Southern Ocean. In parallel, the German Research Foundation (Deutsche Forschungsgemeinschaft, DFG) started a priority program ‘Antarctic Research’ (since 2003 called SPP-1158) to foster and intensify the cooperation between scientists from different German universities and the AWI as well as other institutes involved in polar research. Here, we review the main findings in meteorology and oceanography of the last decade, funded by the priority program. The paper presents field observations and modelling efforts, extending from the stratosphere to the deep ocean. The research spans a large range of temporal and spatial scales, including the interaction of both climate components. In particular, radiative processes, the interaction of the changing ozone layer with large-scale atmospheric circulations, and changes in the sea ice cover are discussed. Climate and weather forecast models provide an insight into the water cycle and the climate change signals associated with synoptic cyclones. Investigations of the atmospheric boundary layer focus on the interaction between atmosphere, sea ice and ocean in the vicinity of polynyas and leads. The chapters dedicated to polar oceanography review the interaction between the ocean and ice shelves with regard to the freshwater input and discuss the changes in water mass characteristics, ventilation and formation rates, crucial for the deepest limb of the global, climate-relevant meridional overturning circulation. They also highlight the associated storage of anthropogenic carbon as well as the cycling of carbon, nutrients and trace metals in the ocean with special emphasis on the Weddell Sea.
KeywordsPolar meteorology Polar oceanography Antarctica Southern Ocean Weddell Sea
Antarctic Bottom Water
Antarctic Intermediate Water
Antarctic Circumpolar Current
Antarctic mesoscale prediction system
Atmosphere-ocean general circulation model
Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research
Cold air outbreak
Convective boundary layer
Circumpolar Deep Water
Consortium of small-scale modelling
Deutsche Forschungsgemeinschaft (German Research Foundation)
Dissolved organic carbon
Dissolved organic matter
German Meteorological Service
General circulation model of the Max Planck Institute for Meteorology (Hamburg)
ECHAM/MESSy Atmospheric Chemistry
Finite Element Sea ice Ocean Model
Filchner-Ronne Ice Shelf
Regional climate model combining HIRLAM and ECHAM
High-resolution limited area model
High-Salinity Shelf Water
Large Eddy Simulation
Larsen Ice Shelf
Moderate-Resolution Imaging Spectroradiometer
Northern Annular Mode
National Centres for Environmental Prediction
Precipitation minus evaporation
Potential anthropogenic climate changes
Southern Annular Mode
Subantarctic Mode Water
Sea ice concentration
Sea ice extent
Schwerpunktprogramm (DFG Priority Program)
Solar zenith angle
Warm Deep Water
World Meteorological Organization
Weddell Sea Bottom Water
Weddell Sea Deep Water
Two hundred years ago, the Southern Ocean was viewed as an inaccessible, stormy and icy sea, challenging dauntless explorers, hunters and whalers. Today, the Southern Ocean is considered to be an important tessera of the climate puzzle hosting numerous processes of global importance. It connects the three major oceans through the most vigorous ocean current on Earth, the Antarctic Circumpolar Current, ventilates most of the world ocean abyss, participates in the global carbon cycle and affects the rate of global sea level rise due to the interaction with the Antarctic Ice Sheet. However, the Southern Ocean not only acts but also reacts to sea floor topography and the polar atmosphere. The former guides currents around the continent, warm waters of open ocean origin into ice shelf cavities, dense shelf waters down the continental slope and determines the kind of water able to escape the basins of the marginal seas. The latter provides energy for driving the ocean currents, controls the surface fluxes of heat and moisture, regulates the exchange of natural and anthropogenic gases and determines sea ice properties and coverage with consequences for primary production, ocean-air fluxes and water mass characteristics. The atmosphere links air-ice-ocean interaction with the stratosphere, e.g. changes in stratospheric circulation, due to ozone loss, to tropospheric circulations and sea ice coverage. Furthermore, the atmosphere is the fastest connection between Antarctica and the mid-latitudes, and the tropics.
Due to the potential of the Weddell Sea region for innovative discoveries and the relative vicinity to home ports, past (W. Filchner, Deutschland Expedition, 1911–1913) and recent German investigations focussed on the Weddell Sea and the fringing continent including ice shelves and the hinterland. Since 1981, continental research in the Weddell Sea area is based on the Ekstrømisen at the German permanent station, today called ‘Neumayer III’, and the summer Kohnen Station on Drauning Maud Land (Fig. 1). Since 1982, expeditions to the Weddell Sea (and beyond) are supported immensely by the icebreaker Polarstern, which provides the platform for multi-disciplinary polar research even during the austral winter months. Two research aircraft are available for scientific campaigns during the summer season. These research platforms (stations, icebreaker, aircraft) are run by the Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research (AWI) and allow the operation of modern devices ranging from ROVs to recently used unmanned air vehicles (UAV, Jonassen et al. 2015) for atmospheric boundary layer research.
Here, we review the achievements of recent investigations of the polar atmosphere and ocean with the focus on the Weddell Sea, primarily funded by the Priority Program (Schwerpunktprogramm, SPP-1158) ‘Antarctic Research’ of the German National Science Foundation (Deutsche Forschungsgemeinschaft, DFG). The main aim of the SPP is to give scientists from universities access to Antarctic stations and research platforms, thereby fostering and intensifying the cooperation between scientists from different German universities and the AWI as well as other institutes worldwide involved in polar research. This paper summarizes the main findings of the last decade.
2 Climate relevant processes in the Antarctic atmosphere
The atmosphere is a key component of the Antarctic climate system. Atmospheric processes ranging from micrometres to thousands of kilometres are responsible for the interaction and transports of momentum, energy and matter at the interface of ocean and ice surfaces, exchange between tropics, mid-latitudes and the Antarctic and the exchange between stratosphere and troposphere. Research in the realm of the Antarctic atmosphere contributed to the understanding of the interaction of the changing ozone layer with large-scale atmospheric circulations and changes in the sea ice cover (Section 2.1). The complex interaction between the radiation field as a function of wavelength, snow reflection and clouds was investigated (Section 2.2). Studies using climate and weather forecast models improved knowledge of the climate change signals associated with synoptic cyclones (Section 2.3) and the water cycle (Section 2.4). Investigations of the atmospheric boundary layer focused on the interaction between atmosphere, sea ice and ocean particularly for leads and polynyas (Section 2.5).
2.1 Ozone-related changes in atmospheric circulation and sea ice extent
The stratospheric ozone depletion is most evident in the ozone hole that appears each austral spring over Antarctica. The ozone depletion led to a 6 °C cooling of the lower stratosphere over the South Pole and an associated intensification of the stratospheric vortex in spring. The interaction of these stratospheric changes with the troposphere and particularly with the sea ice extent in the Antarctic was a focus of international research during the last decade. Since 1992, weekly ozone soundings have been performed at the German Antarctic research station Neumayer (König-Langlo and Loose, 2007). These measurements continue the time series that started already in 1985 at the neighbouring German research base Georg-Forster-Station (König-Langlo and Gernandt, 2009). Ozone sensors (ECC 5A/6A) mounted on RS80/RS90/RS92 radiosondes (Vaisala) have been used. High ozone partial pressures are measured at altitudes around 20 km—called ‘ozone layer’—from December/January to the end of August. During Antarctic spring (September to November), the ozone layer vanishes more or less completely.
Between 2001 and 2004, the overall springtime ozone concentrations measured above an altitude of 20 km were rising again. This effect was interpreted as the beginning of the recovery of the ‘ozone hole’ as a response to the worldwide ban of nearly any CFC product in the Montreal Protocol in 1987. However, the measurements in the following years from Neumayer II (Hoppel et al. 2005) showed that the recovery of the ozone hole did not start at that time. Especially the very high temperatures and ozone concentrations during Spring 2002 (Fig. 2) could be explained as a consequence of a dynamic breakdown of the Antarctic stratospheric vortex during winter. It took a whole decade until the measurements from Neumayer and other Antarctic stations indicated an ongoing recovery of the ozone layer. In September 2014, the World Meteorological Organization (WMO) officially posted the success of the Montreal Protocol from 1987 (WMO 2014). However, the observed increase of global annual mean total ozone of 1 % between 2000 and 2013 compared to the large interannual variability found in that time period did not allow to conclude a recovery of the Antarctic ozone.
The trends of sea ice extent (SIE) during the last decades are different for both hemispheres. Arctic sea ice has dramatically decreased in the recent past. Estimates from satellite measurements find a negative trend of about 10 % per decade since 1979 (Comiso et al. 2008). In contrast, annual mean Antarctic sea ice increased by about 1 % per decade for the years 1978–2006 (Turner et al. 2009). While the Arctic sea-ice retreat has been associated with the warming of the troposphere caused by increasing greenhouse gas (GHG) concentrations (IPCC 2013), the ozone depletion by man-made halogens in the polar stratosphere and its impact on tropospheric circulation have been suggested as the driving mechanism to explain the observed Antarctic changes (e.g. Thompson and Solomon 2002). However, there is still low confidence in the scientific understanding of the observed increase in Antarctic SIE since 1979, due to missing knowledge of internal variability and competing explanations for the causes of change (IPCC 2013). With further increasing GHGs and an expected recovery of polar ozone at the end of the twenty-first century, projections of future polar climate and its hemispheric differences are highly uncertain. To understand and project the interactions between the atmosphere, oceans and the cryosphere as well as the chemical and radiative effects of natural and anthropogenic climate gases throughout the troposphere and stratosphere, complex numerical models need to be applied. A new atmosphere-ocean version of the ECHAM/MESSy Atmospheric Chemistry (EMAC) chemistry-climate model was used in a study, which combines the EMAC model (Jöckel et al. 2006) with the Max Planck Institute-Ocean Model (MPI-OM, Jungclaus et al. 2006).
A comparative analysis of the Arctic region reveals a similar behaviour on interannual time scales (not shown) in which the SIE is positively correlated with the Northern Annular Mode (NAM)—the northern hemisphere equivalent of the SAM—consistent with previous studies (e.g. Rigor et al. 2002). However, on time scales longer than 11 years a much more coherent pattern of behaviour emerges in which periods of a positive polarity of the NAM are associated with reduced SIE and vice versa (not shown). This happens because the NAM is an important driver for the Atlantic meridional overturning circulation (Eden and Jung 2001); and on time scales longer than 11 years, an enhanced overturning warms the high northern latitudes leading to reduced SIE.
2.2 Progress in the measurements of sky luminance and sky radiance
Variations in the ozone concentrations have a strong impact on the incoming radiation on the earth surface. It is well known that the direct impact is strongly wavelength-dependent. While the influence of ozone concentrations on the radiation is weak in the visible part of the spectrum, it can be dominant in parts of the infrared and in the shortwave ultraviolet (UV) spectrum. An improved understanding of the radiation is therefore helpful to determine the impact of current and future changes on the climate in Antarctica. Angular distribution of solar radiance and its spectral characteristics are key radiative quantities to study the impact of climate changes in Antarctica (Cordero et al. 2013, 2014). These quantities and the absorption characteristics of snow determine how much radiation is reflected back to space and how much snow melts should temperatures rise.
In the last decade, the observational capabilities for measuring radiance and luminance were significantly improved. In contrast to the ‘old’ scanning instruments that needed one day to measure the spectral radiance field, the newly developed instruments are now capable to measure sky radiance in dependence of zenith and azimuth angle in more than 100 directions simultaneously within a second (Riechelmann et al. 2013; Tohsing et al. 2013; Seckmeyer et al. 2010). These instruments have already improved our understanding of climate change in both the Arctic and Antarctic, and we expect that also for the future.
Sky luminance and spectral radiance have been characterized at the Neumayer Station during the austral summer 2003/2004 (Wuttke and Seckmeyer 2006). The high reflectivity of the surface (albedo) in Antarctica, reaching values up to 100 % in the UV and visible part of the solar spectrum due to snow cover (Wuttke et al. 2006) modifies the radiation field considerably when compared to mid-latitudes. A dependence of luminance and spectral radiance on solar zenith angle (SZA) and surface albedo was identified. For snow and cloudless sky, the horizon luminance exceeds the zenith luminance by as much as a factor of 8.2 and 7.6 for a SZA of 86° and 48°, respectively. Thus, a snow surface with high albedo can enhance horizon brightening compared to grass by a factor of 1.7 for low sun at a SZA of 86° and by a factor of 5 for high sun at a SZA of 48°. Measurements of spectral radiance show increased horizon brightening for increasing wavelengths and, in general, a good agreement with model results. However, large deviations are found between measured and modelled values especially in the infrared range that are only partly explained by measurement uncertainties. Progress is expected by future studies with the faster instruments available now.
2.3 Extra-tropical cyclones and storms and their impact on the southern polar hemisphere
Extra-tropical cyclones (ETCs) in the mid to high latitudes of the southern hemisphere (SH) are a fundamental part of the atmospheric energy and momentum transport. ETCs are essential for the meridional exchange processes, which are necessary to maintain the hemispheric and, thus, global energy balance of the Earth’s climate system. In contrast to the northern hemisphere, where stationary waves in the mid-troposphere are important for the poleward energy transport, the atmospheric energy transport in the SH is mainly accomplished by transient and shorter baroclinic waves, evident as travelling ETCs at the surface. In the absence of orographic influences, which lead to distinct areas of baroclinic cyclogenesis in the northern Atlantic and Pacific, the SH transient waves have a more circumpolar distribution. At present, potential anthropogenic climate changes (PACC) are investigated by means of a multi-model ensemble of state-of-the-art coupled atmosphere-ocean general circulation model (AOGCM) simulations, especially to understand the nature and causes of potential changes in future ETCs and their impact on Antarctica.
The potential anthropogenic climate change also impacts poleward moisture fluxes in the SH. Distinguishing between thermodynamic and dynamic influences of PACC on moisture fluxes, PACC signals of the responsible waves in the synoptic scale show a poleward shift related to the poleward shift of ETCs (Grieger et al. 2015). Antarctic net precipitation was calculated by means of the vertically integrated moisture flux. Grieger et al. (2015) found signals of increasing net precipitation whereas the dynamical part of net precipitation decreased. They explained these findings with the low variability of synoptic-scale waves, which generally decreased, especially off the coast of West Antarctica, and suggested that this is related to a changing variability of the ABS low.
2.4 Components of the water cycle and their role for ice and snow accumulation
Precipitation is the dominant term among the various components of surface snow accumulation. Information from in situ observations for precipitation events are hardly available due to the difficulties in measuring precipitation under Antarctic conditions, even when using automatic weather stations (e.g. van den Broeke et al. 2004; Welker et al. 2014). Therefore, an annual snow accumulation climatology has been frequently used as an indicator for precipitation (King and Turner 1997). Later on, precipitation climatologies have been derived from model data, such as reanalysis data (Bromwich et al. 2011), mesoscale weather forecasts using the Antarctic Mesoscale Prediction System (Bromwich et al. 2005) as shown by Schlosser et al. (2008) or by regional climate models (van de Berg et al. 2006; Lenaerts et al. 2012). Recently, Palerme et al. (2014) have used satellite products from CloudSat to derive precipitation. Highest precipitation rates of more than 500 mm per year were found along the coastal escarpment; on the ice sheet plateau, the rates drop to less than 100 mm per year.
2.4.1 Precipitation processes related to synoptic disturbances
Large portions of precipitation in the coastal regions fall during episodes of passing fronts of cyclonic systems (King and Turner 1997). If precipitation, and hence the accumulation, occurs preferentially during particular months of the year, the temperature derived from ice core analysis will be biased towards the conditions prevailing during these days in the region around the drilling site. Thus, it is necessary to know the amount and timing of precipitation and to investigate precipitation events on short time scales with high spatial resolution.
A high spatial and temporal resolution is needed particularly for the simulation of high precipitation events. A selected weather situation, formation and horizontal distribution of clouds and precipitation in Queen Maud Land have been investigated by Wacker et al. (2009), using a high-resolution non-hydrostatic weather forecast model COSMO (consortium of small-scale modelling; Steppeler et al. 2003). This model was initially developed at the German Meteorological Service (DWD) and applied in that study for the first time for Antarctic conditions. A 12-day episode from February 1999 is addressed, when low-pressure systems and related fronts travelled along the coast of Queen Maud Land. Emphasis was placed on the temporal evolution and horizontal distribution of clouds and precipitation in high spatial resolution (7-km horizontal mesh size).
All mentioned studies on precipitation in Antarctica using mesoscale forecast models, show some success in the simulation of general spatial and temporal patterns. Nevertheless, all model precipitation data for Antarctica reveal deficiencies, such as the supposed overestimation (Fig. 8). This argues for a continuing effort in both numerical model development and precipitation observations.
2.4.2 Regional climate model simulations of the water cycle
The regional accumulation changes are largely driven by changes in the transient activity around the Antarctic coasts due to the varying Antarctic oscillation (AAO) phases. The monthly mean AAO index from 1958 to 1998 is based on the first principal component of the National Centres for Environmental Prediction (NCEP) 850 hPa extratropical height field (20° S–90° S) adapted from Thompson and Wallace (2000). During positive AAO, more transient pressure systems travel towards the continent; thus, West Antarctica and parts of southeastern Antarctica gain more precipitation and mass. Over central Antarctica, the prevailing anticyclone causes a strengthening of polar desertification connected with a reduced surface mass balance in the northern part of East Antarctica. The shifts in the AAO pattern is accompanied by changes in the short baroclinic waves. The associated heat and humidity fluxes (on time scales 2–6 days at 850 hPa) between positive and negative AAO periods show most pronounced shifts over the Southern Ocean. During positive AAO phases, increased heat and humidity transports, due to synoptical cyclones towards the Antarctic continent, appear over the Pacific Ocean, while over the Atlantic Ocean a reduction occurs (Dethloff et al. 2010).
2.5 The atmospheric boundary layer and interactions with sea-ice and ocean
The grid size of the newest generation of weather prediction models reaches the kilometre range and might decrease in the future to even smaller values, approaching finally the scale of large convective eddies in very shallow convection. The turbulent processes at grid sizes around 1 km become partly resolved and partly parameterized. When flux parameterizations in this overlapping region (often called grey zone or terra-incognita; Wyngaard (2004); Mironov (2009)) are not adjusted to the higher resolution, fluxes can be double-counted. Studies on convective mesoscale processes that occur frequently in both polar regions are presented in the following with a special focus on the difficulties related to the grey-zone problem. Due to the similar environment, process studies that have been carried out in the Arctic are very relevant for the Antarctic as well.
2.5.1 Convection over leads
Thus, it is important to account for convection over leads in climate models. The horizontal scale of leads and plumes is much below the grid size of climate models so that the fluxes over grid boxes with partial sea ice coverage usually depend linearly on the sea ice concentration. However, Andreas and Cash (1999) emphasize that fluxes per unit area are larger over small leads than over large ones, and Lüpkes and Gryanik (2015) show that the transfer of momentum and heat depends nonlinearly on sea ice concentration.
The application of the new closure showed that regions of down-gradient transport and counter-gradient transport are well reproduced as in the corresponding LES (Fig. 11). Hence, a good agreement resulted also for the wind and temperature fields (see Lüpkes et al. 2008b), which were strongly affected by the convection over leads. These investigations show that the above parameterizations can be considered as a first step towards an adjustment of convection parameterizations to the grey zone in case of shallow convection over leads. Forthcoming climate models should account for the nonlinear dependence of fluxes on the sea ice concentration.
2.5.2 Cold-air outbreaks
The patterns of organized convection, which are visible by cloud streets, are observed over Greenland and the Barents Sea in more than 50 % of the time during winter (Brümmer and Pohlmann 2000). Although similar statistics for the Antarctic are not yet available, satellite images show that CAOs with organized convection are a frequent phenomenon during off-ice flow, e.g. over the Bellingshausen Sea. Studies of CAOs in the Arctic can be considered as representative also for the Antarctic. Due to their impact on the exchange processes between ocean and atmosphere, it is important that they are well represented by parameterizations used in numerical weather and climate models.
Several studies (e.g. observational studies by Brümmer 1999; Hartmann et al. 1997) show that roll convection can substantially contribute to vertical transport of heat, moisture, and momentum. Therefore, it is often argued that vertical transport might be enhanced by rolls (e.g. Kristovich et al. 1999). However, as discussed in Gryschka et al. (2014), this needs to be proven. One reason is that it is not possible to find an observed reference case without rolls for the same large-scale forcing. A second reason is that, so far, it was assumed that rolls in CAOs are developing by a pure self-organization mechanism so that again—for the same forcing—no case can exist without rolls. However, Gryschka et al. (2008) have shown with the LES model PALM (Maronga et al. 2015) that this assumption is not generally valid. They introduced and distinguished the terms ‘forced roll convection’ and ‘free roll convection’. Free rolls develop by a pure self-organization of the flow only for small values of the stability parameter −z i/L (<10), where z i is the height of the convective mixed layer and L is the Obukhov length. Forced rolls are triggered by upstream heterogeneities in the surface temperature (e.g. in the marginal ice zone in case of CAOs) and can develop also for much larger values of −z i/L. Because in CAOs, −z i/L typically is much larger than 10 (e.g. in most of the observed CAOs in Brümmer (1999)), it can be expected that forced rolls are the dominant type of rolls within CAOs situations.
These findings allow a comparison of LES runs with and without rolls using the same large-scale meteorological forcing. This can be achieved by prescribing the surface temperatures with or without upstream heterogeneities as was done in the LES parameter study of Gryschka et al. (2014). The authors carried out 27 LES for 12 different CAO scenarios, covering a wide parameter range typical for CAOs. The stationary model domain was large enough to cover the development of the convective boundary layer (CBL) and roll convection for a wide distance over the ocean (up to 160 km from the sea ice edge), while the resolution of 50 m was fine enough to resolve small-scale unorganized turbulence. For each scenario, a roll and non-roll case was simulated and the wavelength of the rolls varied.
There are also studies with mesoscale non-eddy resolving models (Wacker et al. 2005; Chechin et al. 2013) which demonstrate that the observed development of the convective boundary layer during CAOs, including turbulent fluxes, are well reproduced with grid sizes of 4–15 km when adequate non-local turbulence closures are used. These closures do not need to account explicitly for roll convection, which can be explained by the findings of Gryschka et al. (2014) described above. Furthermore, Chechin et al. (2013) found that the formation of a wind maximum in the convective layer over open water with strong impact on the turbulent fluxes requires grid sizes smaller than 30 km for an accurate reproduction.
2.5.3 Katabatic wind and polynyas
The katabatic wind, developing on the Antarctic Ice Sheet, represents a key factor for the exchange of energy and momentum between the atmosphere and the underlying surface. Considering the large area of the Antarctic continent, this wind system plays also an important role in the global energy and momentum budget. A generally stable stratification over the ice slopes leads to the development of a katabatic wind system, which enhances air/snow interaction processes and influences the surface mass balance (van Lipzig et al. 2004; van de Berg et al. 2006). Wind speeds up to gale force are often observed in confluence zones and regions of steep topography near the Antarctic coast (Loewe 1972). A special point of interest is the modification of katabatically generated air flows when passing over the coastline and interacting with the sea ice or open water surface by forming or maintaining polynyas, which represent important areas of sea ice production and brine formation.
In the last two decades, a couple of numerical investigations of the katabatic wind system in Antarctica have been performed using 3D meso-scale models (e.g. Heinemann 1997; van Lipzig et al. 2004; Parish and Bromwich 2007; Dethloff et al. 2010). It was found that a high model resolution of at least 15 km is needed to capture the katabatic wind in regions near the coast or in topographically structured areas such as the Antarctic Peninsula.
The correct representation of the atmospheric forcing on polynya formation is crucial for the quantification of dense shelf water formation in the coastal polynyas of the Weddell Sea. Haid et al. (2015) found a large sensitivity of coastal polynya formation in the southwestern Weddell Sea to the atmospheric forcing for the sea ice-ocean model FESOM, using different coarse resolution global atmospheric analyses/reanalysis data and high-resolution COSMO model data. Major differences occur in mountainous areas where wind is strongly guided by surface topography. Particularly at Coats Land and along the Antarctic Peninsula, the use of high-resolution forcing results in a substantial improvement of the representation of polynya formation processes.
2.6 Sea ice production in Weddell Sea polynyas
Open water and thin sea ice areas, associated with wintertime polynyas in the coastal areas of the southern Weddell Sea, represent an enormous energy source for the atmosphere but also a large source of High-Salinity Shelf Water (HSSW), which plays a major role for the deep and bottom water formation and ocean circulation under the Filchner-Ronne Ice Shelf (Haid et al. 2015). HSSW production is directly related to sea ice production, which, therefore, is an important quantity in sea-ice/ocean modelling. However, direct measurements of sea ice production in the coastal polynyas are rare. This lack of in situ data is partly remedied by satellite-based studies. The majority of recent satellite studies of sea-ice production in Weddell Sea polynyas rely on passive-microwave sensors (Kern 2009; Drucker et al. 2011; Nihashi and Ohshima 2015). The main disadvantages of passive-microwave retrievals of sea ice production are the coarse resolution and the fact that thin-ice covered polynyas often remain undetected.
The comparison of sea ice production between our study and other recent model or satellite-based studies in the southern Weddell Sea shows that our data yield generally lower sea ice production values; e.g. the model results of Haid and Timmermann (2013) for the Antarctic Peninsula are much higher. Drucker et al. (2011) found an average of 99 km3 for the Ronne Ice Shelf, 112 km3 for the Brunt Ice Shelf, and 30 km3 for the region around the grounded iceberg A-23A. These estimates are three times higher than ours. There may be several reasons for these large differences: (i) Coarse resolution data like passive microwave tend to yield higher sea ice production rates (Nihashi and Ohshima 2015). (ii) Differences exist in the parameterization of turbulent and radiative atmospheric fluxes of different methods, e.g. if the transfer coefficients are not stability dependent. (iii) The atmospheric data driving the thin-ice retrieval has a large impact on the ice production, particularly if there is a cold bias as it is known for National Centres for Environmental Prediction (NCEP) reanalyses products (Lindsay et al. 2014). Since we are using state-of-the-art parameterizations (Adams et al. 2013) and ERA-Interim reanalysis data, we think that our high-resolution (2 km) data set represents a new benchmark for the comparison with model estimates of sea ice production in Weddell Sea polynyas.
3 Changes in the Southern Ocean and its reverberation in cryosphere and biosphere
Changes in the Southern Ocean significantly influence global climate in many ways and on a large range of time scales. The capacity of the ocean to store heat and carbon is a prerequisite to understand the strength of the fluctuations in global atmospheric temperature and CO2 during the last several 100,000 years. The Southern Ocean is a key region for the exchange of energy and gases between atmosphere and the deep ocean. Here, surface water is converted to deep and bottom water and the deepest branch of the global meridional overturning is formed. Compared to other ocean basins, the warming rate in the Southern Ocean is the highest. Interaction of relative warm water with ice sheets is thought to be one of the major processes causing sea level rise and ice sheet mass loss; today, about 20 % of the global sea level rise is already attributed to this process. Warming of the Southern Ocean might also influence the formation rates of deep and bottom water as well as the uptake of anthropogenic CO2. Another peculiarity of the Southern Ocean is the dominant role of iron availability for the control of biological carbon fixation and the interaction with processes involving microorganisms on the transformation and transport from surface waters to sediments. Even though chemical and biological interactions affect and interconnect the biogeochemical cycles of carbon and trace elements of the Southern Ocean, so far, few studies have addressed this complex research area.
In the last decades, research in the Southern Ocean contributed to new insights into the interaction between ocean and ice sheets, the changes in water mass characteristics, ventilation, and formation rates and associated storage of anthropogenic carbon as well as into the cycling of carbon, nutrients, and trace metals. Most oceanic field studies have been conducted in the Atlantic sector of the Southern Ocean, especially in the Weddell Sea. Here, ocean water has access to two large ice shelves, the Filchner-Ronne Ice Shelf (FRIS) and the Larsen Ice Shelf (LIS). Interaction between the ocean and these ice shelves is crucial for the formation of Weddell Sea Deep Water (WSDW) and Weddell Sea Bottom Water (WSBW), the main precursors for Antarctic Bottom Water (AABW), which forms the deepest limb of the global, climate-relevant meridional overturning circulation. However, the importance, local variability, and evolution of deep and bottom water formation under changing climate conditions as a long-term sink for atmospheric anthropogenic carbon (C ant) is still under debate, particularly in the Weddell Sea.
3.1 Freshwater input through ice shelf basal melting
In a steady state world, the snow accumulation on top of the (up to 4000-m thick) Antarctic Ice sheet (Section 2.4) is balanced by the transfer of mass to the ocean, either by iceberg calving at ice shelf fronts or by melting of ice shelf bases (Jacobs et al. 1992). Recent studies, based on remote sensing, have revealed that both transfer processes are of equal relevance, and the basal mass loss in the Weddell Sea was found to be 118 ± 52 Gt year−1 (Depoorter et al. 2013). Independent evidence comes from earlier estimates of Huhn et al. (2008), who used the distribution of He and Ne isotopes to infer the contribution of basal melt to the Weddell Sea water masses. Combined with their estimates of the formation rates of deep and bottom water they reported 35 ± 19 Gt year−1 for LIS and 123 ± 53 Gt year−1 for FRIS.
This self-regulating process can only be bypassed if the heat for melting is provided by the open ocean. Such situation is known for the Amundsen and Bellingshausen Seas, where at mid-depths Circumpolar Deep Water (CDW), originating from the Antarctic Circumpolar Current (ACC), penetrates onto the continental shelf and cascades down towards the ice shelves draining the West Antarctic Ice Sheet (Nakayama et al. 2013). This heat supply causes the highest basal melt rates so far known in Antarctica with estimates reaching up to 100 m per year underneath the Pine Island Glacier Ice Shelf (Dutrieux et al. 2014). The resulting freshwater plume primarily influences the surface layer but can also be mixed into the CDW due to convection in front of the ice shelf, thus reducing the heat content of the inflowing warmer deep water. One might speculate that the large volume of glacially derived freshwater is one of the reasons for the Amundsen and Bellingshausen Seas not being a site for deep and bottom water formation today. Part of the meltwater injected into Amundsen Sea is advected northward and into the ACC (Kusahara and Hasumi 2014) while the other part is transported with the coastal current into the Ross Sea. Here, according to a numerical study (Nakayama et al. 2014), the meltwater is likely to be responsible for the observed long-term decrease of shelf water salinity (Jacobs et al. 2002). The freshening in the Ross Sea, in turn, is suggested to have caused the decline in the salinity of the AABW in the Australian Antarctic Basin (Rintoul 2007). The latter demonstrates the intricate transfer of surface signals, due to changes in the mass balance of the Antarctic Ice Sheet, into the deep ocean.
In the large embayments, such as the Weddell and Ross Sea, the circulation in the large sub-ice shelf cavities is driven by HSSW (Fig. 19) formed in polynyas and leads (see Section 2.6) near the ice shelf fronts (Tamura et al. 2008; Haid and Timmermann 2013). Although at surface freezing temperature, HSSW fuels area-mean basal melt rates of a few decimetres per year (several meters per year locally along the deep grounding lines) due to the pressure dependence of the freezing point. The resulting meltwater plume is super-cooled, i.e. below the in situ surface freezing point, and less buoyant than the ambient shelf water, thus following the bottom bathymetry on its way to the continental shelf break (Fig. 19). In the southern Weddell Sea, the freshwater of glacial origin predominantly gets incorporated into the WSBW that is confined to the deep Weddell Basin. Basal melt water that is more directly fed into the World Ocean originates from the LIS fringing the western Weddell Sea (Huhn et al. 2008). Part of the LIS freshwater remains on the northwestern Weddell Sea continental shelf; the observed freshening of the winter shelf water column near the tip of the Antarctic Peninsula (Hellmer et al. 2011) as well as the salinity decrease of the deep waters in the central basin of Bransfield Strait (Garcia and Mata 2005) might have been caused by an increase of LIS basal melting.
For LIS, however, the basal mass loss does not change much over the next centuries; it even decreases in the twenty-second century (Fig. 20, Timmermann and Hellmer 2013). The main reason is the presence of cold, fresh meltwater from FRIS that prevents WDW intrusions at LIS. As demonstrated for the Amundsen and Ross Seas (Nakayama et al. 2014), this indicates again a link between the different ice shelves via the continental shelf and slope circulations and calls for circumpolar rather than regional approaches if the interaction of the Southern Ocean with Antarctic ice shelves is investigated.
3.2 Water mass and ventilation changes in the Weddell Sea
The only pathway by which the interior Weddell Sea gains significant amounts of heat is the inflow of warm (>1.3 °C) and saline CDW of the ACC (Fig. 18) with the southward oriented eastern limb of the Weddell Gyre between 20° E and 30° E (Schröder and Fahrbach 1999). When CDW enters the Weddell Sea proper across the Prime Meridian its warm core occupies the depth range between 150 and 600 m. The inflow is split into two branches, one centred between 63° S and 64° S associated with the topographic slope north of Maud Rise and the other between 68° S and 69° S associated with the continental slope of Antarctica (Cisewski et al. 2011). Downstream of Maud Rise the two cores mix mainly isopycnically (Leach et al. 2011) to form the so-called Warm Deep Water in the Weddell Sea. CDW is predominately fed by the North Atlantic Deep Water but also contains deep waters from other ocean sources.
WDW takes a crucial role in the ocean-ice shelf interaction, hence basal melting, and in the formation of deep and bottom water in the Weddell Sea, the major source (about 60 %, Orsi et al. 1999) of AABW. One of the important predecessors of AABW is WSBW. It results from the interaction of HSSW, locally formed during sea ice formation (Foster and Carmack 1976) in front of FRIS and LIS (Foldvik et al. 1985; Weppernig et al. 1996; Foldvik et al. 2004; Huhn et al. 2008), and the mixing with WDW. Further entrainment of WDW forms WSDW found above the WSBW. Additionally, deep water from easterly sources with comparable properties as WSDW enters the Weddell Basin in the deep southern boundary current, probably formed off Cape Darnley west of Prydz Bay (Mantisi et al. 1991; Hoppema et al. 2001; Klatt et al. 2002; Lo Monaco et al. 2005; Oshima et al. 2013). The shallower fractions of WSBW and WSDW exit the Weddell Basin through gaps in the South Scotia Ridge (Muench et al. 1990; Whitworth et al. 1994; Gordon et al. 2001; Naveira Garabato et al. 2002; Schröder et al. 2002; van Caspel et al. 2015), underride the ACC, and power the lower limb of the meridional overturning circulation. Deeper fractions of WSBW and WSDW that do not spill over the South Scotia Ridge follow the Weddell Gyre circulation eastward (Fig. 18).
From transient tracer data taken between 1984 and 1995, Orsi et al. (1999, 2002) estimated a total AABW formation rate of 8.1 ± 2.6 Sv (1 Sv = 106 m3/s), with AABW defined by neutral densities γ n > 28.27 kg/m3. In contrast, Broecker et al. (1997) using radiocarbon (14C) and nutrient observations deduced a total AABW (here defined as ‘ventilated water’, deeper than 1000 m and silicate >90 μmol/kg) formation rate of 15 Sv. The discrepancy is thought to be caused by a reduction of AABW formation in more recent decades (see also Broecker et al. 1999).
The AABW formation rate in the Weddell Sea of 4.9 Sv determined by Orsi et al. (1999) is compatible with results inferred from current meter observations (3–4 Sv, Fahrbach et al. 1991), inverse modelling approaches (5–6 Sv, Sloyan and Rintoul 2001; Lumpkin and Speer 2007), and other hydrography or tracer-based estimates (4–5 Sv, Gordon et al. 1993; Gordon 1998; Mensch et al. 1998; Huhn et al. 2008). Recently, indications for changing water mass properties and a possible reduction of deep and bottom ventilation in the Weddell Sea have emerged. Hellmer et al. (2011) observed a substantial freshening of shelf water masses in the northwestern Weddell Sea, with possible implications for the subsequent bottom water formation. Conductivity-temperature-depth (CTD) sections from 1984 to 2008 revealed a significant warming in the deep water masses (Fig. 21), while WSDW became saltier and WSBW fresher (Fahrbach et al. 2011). Moreover, the cross section area occupied by WSBW at the Prime Meridian declined by 25 %. Purkey and Johnson (2012) estimated a reduction of WSBW in the Weddell Basin of roughly 3 Sv between the 1980s and 2000s. Based on long-term CFC time series, all deep water masses within the Weddell Sea have been continually growing older and less ventilated during the last 27 years. The decline of the ventilation rate of WSDW and WSBW by 18–23 % seems to be mainly caused by mixing with WDW, which aged much faster by 35 %. Increased entrainment of WDW into WSBW and WSDW or a deceleration of the Weddell Gyre circulation may also play a role. The Weddell Sea is not the only region where ageing of water masses was observed. Waugh et al. (2013) estimated that CDW aged in all Southern Ocean sectors between 30 and 60 % in the period of the early 1990s to the late 2000s (~60 % in the Atlantic sector) and attributed this age increase or ventilation decrease to an intensified westerly wind field over the ACC. A new concept of validity areas for tracer couples may help to identify and validate future changes in ventilation in these regions (Stöven et al. 2015).
Due to the involvement of surface water in the formation process, the WSBW and subsequently the AABW carry relatively high loads of atmospheric gases such as oxygen, transient tracers like the anthropogenic CFCs and C ant, and therefore contribute to the ventilation of the global abyss. Despite their relevance, the variability of ventilation and formation rates of deep and bottom water and the related storage of C ant in the Southern Ocean is currently not well constrained (e.g. Bullister et al. 2013). Classical oceanographic measurements (e.g. Fahrbach et al. 2011) combined with observations of helium and neon isotopes (e.g. Schlosser et al. 1990; Huhn et al. 2008) and CFCs (e.g. Orsi et al. 1999; Klatt et al. 2002; Huhn et al. 2013) provided insight into the relevant physical processes. Numerical models with very high resolution are promising and necessary to account for the complex atmosphere-ocean-sea ice-interaction and for irregularities in the bottom topography. The latter influences the exchange of WDW across the continental shelf break as well as the processes under the ice shelf (Timmermann et al. 2012).
3.3 Changes in storage of anthropogenic CO2 in the Southern Ocean: the Weddell Sea
A decrease of the CO2 uptake since the 1980s was also suggested by some large-scale data-based studies (e.g. Le Quéré et al. 2007). Changes in the Southern Annular Mode intensified the westerly winds and shifted their tracks southwards with the consequence of elevated upwelling of subsurface water enriched in (natural) CO2. However, recently, the uptake of CO2 in the Southern Ocean appears to have been increasing again (Landschützer et al. 2015). A more asymmetric atmospheric circulation around Antarctica may be causing this, possibly regulated by the La Nina conditions in the tropical Pacific (Landschützer et al. 2015). With a certain time lag, more C ant will also be transferred into the deeper water masses and, thus, changing future trends should be expected there. These alleged climate changes are much debated and continuing research is necessary.
In the future, SAM-related climate changes are thought to continue (Thompson et al. 2011). Thus, the upwelling of subsurface water (like CDW or WDW) will also continue at an enhanced rate, and the rise of the C ant concentration in the Weddell Sea surface layer will further be reduced in spite of the incessant C ant undersaturation. On longer time scales (decades to centuries), the C ant content of the CDW and thus of WDW will increase due to the invasion of C ant in the source regions of the CDW; in addition, upwelling of CDW would transport C ant to the surface layer of the Weddell Sea. This contemplation as yet excludes possible changes in the sea ice cover. Due to increased global warming, sea ice-free periods might get longer, thus enforcing the exchange of CO2 with the atmosphere. The changes occurring in the Weddell Sea cannot be regarded as isolated, since upwelling is always compensated by enhanced downwelling in other regions. Waugh et al. (2013) suggested that this will occur in northern regions of the ACC where mode waters are formed. This counteracting process was shown to occur in models (Hauck et al. 2013b) and represents a negative feedback for decreasing C ant uptake.
At present, uptake of C ant occurs year-round at a relatively constant rate as it is mainly driven by the increase in atmospheric pCO2. However, there is a strong seasonality of (natural) CO2 uptake, with the largest part occurring in summer during phytoplankton blooms. In the future, the same level of biological production will lead to a larger uptake of CO2 because the carbonate system will be less well buffered at higher Revelle factors, which then will be predominating (Hauck and Völker 2015). Most C ant uptake will occur in austral spring and summer (Fig. 22b), due to the increased importance of the biological carbon pump with rising atmospheric CO2.
In addition to the sequestration of inorganic carbon via the thermohaline advection, the downwelling of dissolved organic matter (DOM) also contributes to the vertical carbon flux in the ocean (Carlson et al. 2010). Marine dissolved organic carbon (DOC) forms a large reservoir in the global carbon cycle (662 Pg C, Hansell 2013), ~90 % of which is not bioavailable and chemically unreactive (refractory). Based on DOC concentration gradients, an export flux of ~86 Tg C year−1 was calculated for the deep Atlantic (Hansell et al. 2009). Since the ultimate source of marine DOC is primary production and therefore CO2, this export also contributes in parts to C ant sequestration. Although the extent of this flux is 2–3 orders of magnitude smaller than the inorganic carbon pump, the residence time for refractory DOC is much longer than for inorganic carbon: DOC in the Weddell Sea and Southern Ocean has an average 14C age of ~5500 years and deviates from the general global correlation of DOC concentration and age (Lechtenfeld et al. 2014; Druffel and Bauer, 2000). Individual organic fractions can reside for much longer-time scales than the average DOC (>24,000 year; Lechtenfeld et al. 2014).
The microbial process, which forms unreactive DOM is critical for an efficient sequestration of DOC: Incubation experiments demonstrated that surface bacteria in the Weddell Sea produce 5–9 % of non-labile DOC from simple organic substrates (Koch et al. 2014). Thus, DOC downwelling in the Weddell Sea contributes to the buffer in the ocean carbon cycle. It is not clear how this carbon buffer will evolve in the future global biogeochemical cycle and how it affects the climate system (Denman et al. 2007). Further removal of DOC depends primarily on microbial degradation and light availability (photo degradation). It was suggested that during the Eocene, a stratified and anaerobic ocean stored large amounts of DOC, which was quickly released after circulation restarted and ventilated the ocean, resulting in a rapid global warming event (Sexton et al. 2011). Although we know that DOC mineralization primarily depends on the molecular chemical composition, structure, concentration, and biochemical formation, the degradation mechanisms are still unresolved.
3.4 Acidification in the polar ocean
The rapid increase of the partial pressure of carbon dioxide (pCO2) in the atmosphere has an impact on the carbonate system in the oceanic surface layer and, with delay, also in the deeper ocean. The uptake of CO2 by the oceanic surface layer leads to a rearrangement in the carbonate system where carbonate ions are lost and protons (H+) gained, i.e. the pH decreases. This process is commonly known as ocean acidification (OA; note that despite large changes the oceanic pH remains in the alkaline realm, i.e. pH >7). The substantial pH-changes that have occurred in the Anthropocene (Rhein et al. 2013) are caused by CO2 emissions at a much larger rate than the ocean-wide overturning and response of carbonate-rich sediments (the latter of which would be able to counter pH changes). Another predominant effect of OA is the decrease in the saturation state of the two main forms of carbonate minerals, namely aragonite and calcite. It must be accounted for that pH and the carbonate ion concentration are subject to a seasonal cycle with high values in summer, due to drawdown of CO2 by photosynthesis and low pH and carbonate in winter (Fig. 22b). Uncertainty in the seasonal signal is still prevailing, which is further demonstrated by Roden et al. (2013), who compared two annual cycles of pH at the same station in the Indian sector of the Southern Ocean and found a factor-of-two difference in amplitude.
It is obvious that OA is driven by the atmospheric burden of anthropogenic CO2, which brings CO2 into the surface layer and changes the carbonate equilibrium, diminishing pH and the carbonate ion concentration. This mechanism will continue in the future, possibly at an even higher rate due to the expected faster rise in atmospheric CO2. The future increase of the Revelle factor as a consequence of a higher CO2 level of the surface water will slow down the CO2 uptake and hence the acidification rate to some extent, although this might not be the case in the southern Weddell Sea where the CO2 uptake rate might continue to grow with a higher Revelle factor (Hauck and Völker, 2015). At the same time, the rate of acidification per CO2 uptake might increase as the carbonate equilibria shift to higher CO2 concentrations. Altogether, these processes will likely lead to higher acidification rates in the future. Via the formation of ventilated deep and bottom waters, the acidification is passed on to the deeper realms of the Weddell Sea and from there to the global ocean; the typical time scale is hundreds to thousands of years. An ice cover tends to slow down OA because the sea ice prevents a full equilibration of ocean and atmosphere. A further effect of sea-ice formation in winter is a reduction of alkalinity of the underlying water, i.e. within the ice, calcium carbonate precipitates in the brine channels. This deteriorates the level of acidification seasonally. It is not clear where the balance of these counteracting processes by sea ice lies, but it is likely to be spatially variable. In addition, decreasing ventilation of the deep Weddell Sea (see Section 3.2) is already slowing down the uptake of anthropogenic CO2 (Huhn et al. 2013) and, thus, the acidification. Even so, the shallow sediments of the Southern Ocean do not contain sufficient carbonate to buffer the enhanced acidification (Hauck et al. 2013a).
The consequences of ocean acidification are likely to be numerous, ranging from changes in chemical speciation of metals (Millero et al. 2009) to changes in absorption of sound waves (Ilyina et al. 2009), and major or minor impact on marine ecosystems (Wittmann and Portner, 2013) with consequences for the structure and functioning of ecosystems and biogeochemical cycles (Gattuso and Hansson 2011; Doney et al. 2012).
3.5 Trace elements and the nutrient cycle
In addition to the carbon sequestration mechanisms highlighted in Section 3.3, C ant is transported from the surface layer into the ocean interior via the biological carbon pump: carbon is taken up by primary production and sinks down the water column as particulate organic matter and calcareous particles and ultimately gets stored in the ocean sediment. Microbial respiration, heterotrophic bacteria (Obernosterer et al. 2008; Jiao and Azam 2011) and photochemical degradation affect the efficacy of the biological pump and the conversion between dissolved and particulate carbon pools. The fluxes depend on the availability of macronutrients (nitrate, phosphate, ammonium and silicate); trace elements (such as iron, manganese, zinc, cadmium); oxygen; temperature; and light availability. Anthropogenic changes in CO2 can also stimulate primary productivity and change the structure of phytoplankton assemblages, therewith affecting the potential for carbon export (Tortell et al. 2008; Hoppe et al. 2013; Trimborn et al. 2013). In the following, we review recent studies that give insight into primary productivity, micronutrient utilization (Fe, Zn Cd, Th, and Pa). The final paragraph infers oceanic mixing and continental weathering sources based on Nd and Hf isotopes.
Iron availability is considered the primary control for biological carbon fixation in the Southern Ocean, contributing about half of the annual carbon fixation from atmospheric CO2. According to the iron hypothesis raised by Martin (1990), stimulation of the biological pump by increased atmospheric supply of iron-containing dust to the Southern Ocean during the drier cold climate periods explains the lowering of CO2 in the atmosphere during glacial periods, which is documented in ice cores. While many ocean iron fertilization experiments provided support for the first part of the iron hypothesis, the stimulation of phytoplankton primary production, a recent study also supported the second part, namely the carbon export from an iron-stimulated phytoplankton bloom to the deep ocean (Smetacek et al. 2012). However, extrapolating the results of such mesoscale experiments to larger space and time scales is hardly possible. In part, it is hampered by the complexity of interactions between the species, which builds the marine food web. Complexity begins already at the level of the primary producers, even within the genus of the diatoms, of which some act as carbon sinkers while others drain predominately silicate (Assmy et al. 2013). The species distribution in turn is influenced by the physical environment—the variables of ocean state, currents, and turbulence regime.
230Th and 231Pa have been used to estimate the export flux of particulate organic carbon as well as trace-metal scavenging processes. 230Th and 231Pa data (Venchiarutti et al. 2011) indicate that distributions in the Atlantic sector of the Southern Ocean are mainly driven by reversible-scavenging mechanisms and the southward upwelling of radionuclide-rich deep water masses. High resolution mapping of particulate and dissolved 234Th in surface water in comparison with satellite chlorophyll-a and trace metal data showed the time lag of 1–2 months between bloom development and the export of adsorbed trace elements to depth (Rutgers van der Loeff et al. 2011). However, bottom scavenging or scavenging in opal rich areas locally create significant variations in these radionuclide distributions. Near the Antarctic Peninsula, thorium isotopes (234Th and 232Th) combined with trace metals (Al, Mn, Fe; Middag et al. (2011a, b); Klunder et al. (2011)) showed signatures highly influenced by terrigenous inputs and effects of sediment resuspension. Settling diatoms in the ACC form an efficient sink for 231Pa that accumulates at mid-depth, while bottom waters are again depleted in 231Pa by contact with opal-rich sediments.
Dissolved hafnium (Hf) and neodymium (Nd) in seawater are of lithogenic origin and are introduced via weathering inputs, dissolution of particles, or seawater-sediment interactions. Their respective radiogenic isotope compositions (143Nd/144Nd expressed as εNd and 176Hf/177Hf expressed as εHf in parts per 10,000) in sea water allow the evaluation of the continental source rocks and weathering regimes (e.g. Piepgras and Wasserburg 1982; van de Flierdt et al. 2002; Bayon et al. 2006). In the first detailed study of the Hf and Nd distributions in the Atlantic sector of the Southern Ocean, only one surface sample had a significantly higher εHf (+6.1) and εNd (−4.0) as a consequence of local weathering inputs from the volcanic King George Island (Stichel et al. 2012b). Markedly negative εNd values coinciding with high Nd concentrations and continental rock-like rare earth elements pattern in surface waters were found 200 km off South Africa, indicating weathering inputs from old continental crust (Stichel et al. 2012b).
Hf consistently shows the lowest concentrations in the surface waters within the Polar Frontal Zone and systematically increases south of the ACC (Stichel et al. 2012b). The concentrations of both elements increase with water depth. Within the ACC, the increase in Nd is nearly linear and correlates well with dissolved Si, indicative of a vertical flux controlled by desorption of scavenged Nd from diatom frustules (Stichel et al. 2012a). The Hf concentration shows a less pronounced increase with depth and remains essentially constant below 1500 m suggesting shallower remineralisation. Within the Weddell Gyre Nd concentrations follow a pattern similar to Hf with increasing values only in the uppermost 1 km of the water column (Stichel et al. 2012a). The limited increase coincides with low particle fluxes and a shallow remineralisation depth documented by the vertical distribution of excess 234Th, which has been detectable in the upper few hundred meters in the Weddell Gyre (Usbeck et al. 2002). Additionally, the high concentrations in the Weddell Gyre are promoted by upwelling of the densest modes of CDW (Stichel et al. 2012a,b).
The Nd isotope composition in CDW can be almost solely explained by mixing of deep water masses coming from the North Atlantic and North Pacific (Stichel et al. 2012b), while Nd concentrations in CDW are governed by a non-conservative behaviour (Stichel et al. 2012a). Antarctic Bottom Water concentrations at the Prime Meridian indicate interactions with the continental shelf off East Antarctica. The observed isotope signature suggests a source region of AABW at about 60° E, east of the Weddell Gyre (Meijers et al. 2010; Stichel et al. 2012a).
4 Summary and conclusions
The main challenges that were addressed by the DFG Antarctic program funded research were to improve the simulation of Antarctic climate and understanding the interaction between atmosphere, ice and ocean. The related research showed significant progress towards more realistic simulations and towards a deeper understanding and improved quantification of the involved processes.
Germany contributed to the long-term monitoring of the ozone layer by continuous measurements starting in 1985. It was found that Antarctic springtime ozone depletion shows large interannual variations and is strongly correlated with a cooling of the stratosphere. So far, the observations are inconclusive regarding a recent recovery of the Antarctic ozone.
The simulation of ozone-related changes in atmospheric circulation and sea ice extent using the EMAC chemistry-climate model reveals a positive correlation between the Southern Annular Mode index and sea ice extent on interannual time scales in agreement with observations. Due to these results, for the second half of the twenty-first century, when total column ozone is projected to recover, the GHG induced climate change will dominate and Antarctic sea ice concentration will decrease. Multi-model ensemble coupled atmosphere-ocean general circulation model simulations show that climate change impacts on extra-tropical cyclones (ETCs) lead to a decrease of the total number of cyclones at the end of the twenty-first century, but the number of extremely strong ETCs increases.
The regional climate model simulations (HIRHAM4) with a horizontal resolution of 50 km made significant progress with regard to the water cycle over Antarctica. Covering the years 1958 to 1998, the mean atmospheric circulation, synoptic weather systems, and the spatial distribution of precipitation minus sublimation structures are realistically simulated. The observed increase in surface mass accumulation at the West Antarctic coasts and reductions in parts of East Antarctica could be explained by the modelled P-E trends.
The parameterization of sub grid-scale processes is still a great challenge for climate research. Work funded by the DFG concentrated on the parameterization of turbulence with a focus on convection studies over leads and during cold-air outbreaks. Using the Large Eddy Simulation model PALM, cases with and without organized roll convection were investigated with the focus on vertical turbulent fluxes. A striking finding was that these fluxes do not differ between roll and non-roll cases. This implies that a development of additional parameterization schemes accounting for specific characteristics of roll convection is not necessary in numerical weather prediction and climate models. On the other hand, the impact of leads on the surface energy balance has been found to be substantial. Progress has been obtained with the parameterizations of the lead-generated turbulence in microscale models. The necessary studies helped to better understand the complex processes over leads. However, parameterizations of the lead impact on energy and momentum fluxes and their non-linear dependence on the sea ice concentration need still improvements in forthcoming climate models. In this respect, the strategy of involving LES and micro/mesoscale modelling in the parameterization development has proven to be promising.
In the marginal seas of the Southern Ocean, katabatic winds and polynyas are features important for the air-sea fluxes. At Luitpold Coast, the link between katabatic winds and polynya formation was studied with the high-resolution (5 km) non-hydrostatic atmospheric model COSMO for the winter season. Changes in polynya area are mainly controlled by the downslope component of the surface wind, which offshore component is mainly steered by a pressure gradient due to katabatic force. The study was supplemented with regard to polynya and dense shelf water formation using the finite element sea ice ocean model FESOM for the southern Weddell Sea. A large sensitivity of coastal polynya formation to the atmospheric forcing is found. Major differences occur in mountainous areas where the wind is strongly influenced by surface topography. Thus, high-resolution atmospheric models are needed to provide realistic forcing for sea ice-ocean models in the Weddell Sea area. The model efforts were complemented by a long-term study for coastal polynyas using MODIS thermal-infrared imagery. Thin-ice thickness and sea-ice production were determined on a daily basis with a high spatial resolution of 2 km. The most efficient sea ice production areas are the polynyas off Ronne and Brunt Ice Shelves. All regions show a high interannual variability and, except for Coats Land, a negative trend for sea ice production. The high-resolution (2 km) data set represents a new benchmark for the validation of model estimates of sea ice production in Weddell Sea polynyas.
Combining observations and state-of-the-art numerical models, the German efforts in the ocean focused on the changes in circulation and water mass ventilation, and the impact of these changes on ice shelf melting as well as the distribution of trace elements and climate relevant trace gases such as CO2.
Starting in 1984, the monitoring at a high-resolution hydrographic section along the Prime Meridian represents the longest time series in the Southern Ocean. Together with a section across the central Weddell Sea, it constitutes the major portion of hydrographic data from the Atlantic sector of the Southern Ocean. The AABW formation rate inferred from the current meter observations (3–4 Sv) corresponds to inverse modelling approaches (5–6 Sv) and other hydrography or tracer-based estimates (4–5 Sv). Recently, indications for changing water mass properties and a possible reduction of deep and bottom ventilation in the Weddell Sea have emerged. The CTD sections (1984 to 2008) reveal a significant warming in the deep waters, while WSDW became saltier and WSBW fresher. Moreover, the cross section area occupied by WSBW at the Prime Meridian declined by 25 %. Based on long-term CFC time series, all deep water masses within the Weddell Sea have been continually getting older and less ventilated during the last 27 years. The decline of the ventilation rate of WSDW and WSBW by 18–23 % seems to be mainly caused by mixing with WDW, which aged much faster by 35 %.
Despite their relevance, the variability of ventilation and formation rates of deep and bottom waters and the related storage of C ant in the Southern Ocean is currently not well constrained. Numerical models with very high resolution are promising and necessary to account for the complex atmosphere-ocean-sea ice-interaction and for irregularities in the bottom topography. In union with the reduced ventilation of the deep and bottom waters, C ant uptake and storage slowed down by 14–21 % as compared to the steady state situation with level ventilation. The mechanism behind the reduction of C ant uptake in the Weddell Sea may well be an elevated upwelling of subsurface water, which occurred at least during the 1990s and 2000s. Downwelling of dissolved organic carbon (DOC) in the Weddell Sea contributes to the buffer in the ocean carbon cycle. It is not clear how this carbon buffer will evolve in the future global biogeochemical cycle and how it affects the climate system. Further removal of DOC depends primarily on microbial degradation and light availability (photo degradation). Via the formation of ventilated deep and bottom waters, the acidification is passed on to the deeper realms of the Weddell Sea and from there to the global ocean with a typical time scale of hundreds to thousands of years. In the bottom water of the Weddell Sea significant reductions—slightly less than in the surface layer—of pH were found. However, the onset of an aragonite undersaturation in the Weddell Sea is still under debate—under steady conditions it might not occur before year 2100 (for calcite even later) or as early as the 2030s in a coastal region east of the Weddell Gyre.
Using He and Ne isotope observations, including the hydrographic data along the aforementioned sections, the contribution of glacial melt to the Weddell Sea water masses is estimated to be 35 ± 19 Gt year−1 (1 Gt = 109 tons) for the Larsen (LIS) and 123 ± 53 Gt year−1 for the Filchner-Ronne Ice Shelves (FRIS). These numbers are a mean over several decades. Melt water more directly fed into the world ocean originates from the LIS fringing the western Weddell Sea. Part of this freshwater remains on the northwestern Weddell Sea continental shelf causing a freshening of the winter shelf water column and a salinity decrease of the deep waters in the central basin of Bransfield Strait. Projections, based on IPCC AR4 climate scenarios, show that the density structure on the southern Weddell Sea continental shelf may allow the slope current to penetrate into the Filchner Trough transporting warm water of open ocean origin to the deep FRIS grounding line. This might boost the basal mass loss by roughly a factor of 10. The associated thickness decrease has severe consequences for the buttressing potential of the ice shelf and the dynamics of the ice streams draining East Antarctica. The increase in glacial meltwater also freshens the western Weddell Sea continental shelf, reducing the basal melting of LIS. This indicates a link between the different ice shelves via the continental shelf and slope circulations and calls for circumpolar rather than regional approaches if the interaction of the Southern Ocean with Antarctic ice shelves is investigated.
Many ocean iron fertilization experiments provided support for the stimulation of phytoplankton primary production, but a recent study also supports the carbon export from an iron-stimulated phytoplankton bloom to the deep ocean. Extrapolating the results of such mesoscale experiments to larger space and time scales is hardly possible, in part, due to the complexity of interactions between the species building the marine food web. Complexity begins already at the level of the primary producers, even within the genus of the diatoms of which some act as carbon sinkers while others drain predominately silicate. The dependence of Cd/P and Zn/P uptake on the availability of other trace metals, most importantly Fe, has clear implications for their use as paleo-tracers of past deep water circulation and nutrient utilization. Culture studies are needed to fully understand the physiological mechanisms governing the uptake and resulting isotope fractionation of Cd and Zn in the oceans, and they will be paramount for establishing a link with the global carbon cycle.
High-resolution mapping of particulate and dissolved 234Th in surface water in comparison with satellite chlorophyll-a and trace metal data shows a time lag of 1–2 months between bloom development and the export of absorbed trace elements to depth. However, bottom scavenging or scavenging in opal-rich areas create locally significant variations in these radionuclide distributions. Near the Antarctic Peninsula, thorium isotopes (234Th and 232Th) combined with trace metals (Al, Mn, Fe) show signatures highly influenced by terrigenous inputs and effects of sediment resuspension while Antarctic Bottom Water concentrations at the Prime Meridian indicate interactions with the continental shelf off East Antarctica. The observed isotope signature suggests a source region of AABW at about 60° E, east of the Weddell Gyre.
In summary, German Antarctic research of the last decade in collaboration with international partners fostered significantly our understanding of a variety of climate relevant processes in the atmosphere and the ocean as well as at their interface.
Much of the research presented here would have not been possible without the funding through the DFG SPP-1158 ‘Antarctic Research’. We also thank the captains and crews of the icebreaker POLARSTERN, the aircraft crews, the personnel of the Antarctic stations, and the logistic department of the Alfred Wegener Institute.
- Abouchami W, Galer SJG, Horner TJ, Rehkämper M, Wombacher F, Xue Z, Lambelet M, Gault-Ringold M, Stirling CH, Schönbächler M, Shiel AE, Weis D, Holdship PF (2012) A common reference material for cadmium isotope studies—NIST SRM 3108. Geost Geoanal Res. doi: 10.1111/j.1751-908X.2012.00175 Google Scholar
- Arndt JE (2013) The International Bathymetric Chart of the Southern Ocean (IBCSO) – Digital Bathymetric Model. Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research, Bremerhaven. doi: 10.1594/PANGAEA.805734
- Assmy P, Smetacek V, Montresor M, Klaas C, Henjes J, Strass V, Arrieta JM, Bathmann UV, Berg GM, Breitbarth E, Cisewski B, Friedrichs L, Fuchs N, Herndl GJ, Jansen S, Krägefsky S, Latasa M, Peeken I, Röttgers R, Scharek R, Schüller SE, Steigenberger S, Webb A, Wolf-Gladrow D (2013) Thick-shelled, grazer-protected diatoms decouple ocean carbon and silicon cycles in the iron-limited Antarctic circumpolar current. Proc Natl Acad Sci U S A. doi: 10.1073/pnas.1309345110 Google Scholar
- Bayon G, Vigier N, Burton KW, Jean Carignan AB, Etoubleau J, Chu N-C (2006) The control of weathering processes on riverine and seawater hafnium isotope ratios. Geology 34(433). doi: 10.1130/G22130.1
- Bromwich DH, Monaghan AJ, Manning KW, Powers JG (2005) Real-time forecasting for the Antarctic: an evaluation of the Antarctic mesoscale prediction system (AMPS). Month Weath Rev 133:581–603Google Scholar
- Bullister J, Rhein M, Mauritzen C (2013) Deep water formation. In: Siedler G, Church J, Gould J, Griffies S (eds) Ocean circulation and climate—observing and modelling the global ocean, 2nd edn. Academic Press, Oxford ISBN 978-0-12-391851-2 Google Scholar
- Comiso JC, Parkinson CL, Gersten R, Stock L (2008) Accelerated decline in the Arctic sea ice cover. Geophys Res Lett 35(L01703). doi: 10.1029/2007GL031972
- Denman KL, Brasseur G, Chidthaisong A, Ciais P, Cox PM, Dickinson RE, Hauglustaine D, Heinze C, Holland E, Jacob D, Lohmann U, Ramachandran S, Da Silva Dias PL, Wofsy SC, Zhang X (2007) Couplings between changes in the climate system and biogeochemistry. In: Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, Tignor M, Miller HL (eds) Climate change 2007: the physical science basis. Contribution of working group I to the fourth assessment report of the intergovernmental panel on climate change. Cambridge University Press, CambridgeGoogle Scholar
- Dethloff K, Glushak K, Rinke A, Handorf D (2010) Antarctic 20th century accumulation changes based on regional climate model simulations. Adv Meteorol ID 327172 14 pp. doi: 10.1155/2010/327172.
- Drucker R, Martin S, Kwok R (2011) Sea ice production and export from coastal polynyas in the Weddell and Ross seas. Geophys Res Lett 38(L17502). doi: 10.1029/2011GL048668
- Esau IN (2007) Amplification of turbulent exchange over wide arctic leads: Large-eddy simulation study. J Geophys Res 112(D08109). doi: 10.1029/2006JD007225
- Foldvik A, Gammelsrød T, Tørresen T (1985) Circulation and water masses on the southern Weddell Sea shelf. In: Jacobs SS (ed) Oceanology of the Antarctic continental shelf, Ant Res Seri 43:5–20 AGU WashingtonGoogle Scholar
- Foldvik A, Gammelsrød T, Østerhus S, Fahrbach E, Rohardt G, Schröder M, Nicholls KW, Padman L, Woodgate RA (2004) Ice Shelf Water overflow and bottom water formation in the southern Weddell Sea. J Geophys Res 109(C02015). doi: 10.1029/2003JC002008
- Foster TD, Carmack EC (1976) Frontal zone mixing and Antarctic bottom water formation in the southern Weddell Sea. Deep-Sea Res 23:301–317Google Scholar
- Garcia CAE, Mata MM (2005) Deep and bottom water variability in the central basin of Bransfield Strait (Antarctica) over the 1980–2005 period. CLIVAR Exchanges 10:48–50Google Scholar
- Gattuso JP, Hansson L (eds) (2011) Ocean acidification. Oxford University PressGoogle Scholar
- Gordon AL (1998) Western Weddell Sea thermohaline stratification. In: Jacobs SS, Weiss RF (eds) Ocean, ice and atmosphere: interactions at Antarctic continental margins. Ant Res Seri 75:215–240 AGU WashingtonGoogle Scholar
- Gryschka M, Drüe C, Etling D, Raasch S (2008) On the influence of sea-ice inhomogeneities onto roll convection in cold-air outbreaks. Geophys Res Lett 35(L23804). doi: 10.1029/2008GL035845
- Hauck, J (2013) Processes in the Southern Ocean carbon cycle: Dissolution of carbonate sediments and inter-annual variability of carbon fluxes. Rep Pol Mar Res 669. 122 pp AWI Bremerhaven GermanyGoogle Scholar
- Hauck J, Hoppema M, Bellerby RGJ, Völker C, Wolf-Gladrow D (2010) Data-based estimation of anthropogenic carbon and acidification in the Weddell Sea on a decadal timescale. J Geophys Res 115(C03004). doi: 10.1029/2009jc005479
- Hellmer HH (2004) Impact of Antarctic ice shelf melting on sea ice and deep ocean properties. Geophys Res Lett 31(L10307). doi: 10.1029/2004GL19506
- Hoppel K, Nedoluha G, Fromm M, Allen D, Bevilacqua R, Alfred J, Johnson B, König-Langlo G (2005) Changes in ozone loss at the upper edge of the Antarctic ozone hole during 1994–2005. Proc AGU Fall Meeting 2005 San FranciscoGoogle Scholar
- IPCC 2013 Climate change 2013: the physical science basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. In: Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, Midgley PM (eds) Cambridge University Press, Cambridge. 1535 pp. doi: 10.1017/CBO9781107415324
- Jacobs SS, Hellmer HH, Doake CSM, Jenkins A, Frolich RM (1992) Melting of ice shelves and the mass balance of Antarctica. J Glaciol 38:375–387Google Scholar
- Jiao N, Azam, F (2011) Microbial carbon pump and its significance for carbon sequestration in the Ocean. Science Suppl 43–45 Google Scholar
- Jöckel P, Tost H, Pozzer A, Brühl C, Buchholz J, Ganzeveld L, Hoor P, Kerkweg A, Lawrence MG, Sander R, Steil B, Stiller G, Tanarhte M, Taraborelli D, Van Aardenne J, Lelieveld J (2006) The atmospheric chemistry general circulation model ECHAM5/MESSy: consistent simulation of ozone from the surface to the mesosphere. Atmos Chem Phys 6:5067–5104CrossRefGoogle Scholar
- Jonassen, M. O., Tisler, P., Altstädter, B., Scholtz, A., Vihma, T., Lampert, A., König-Langlo, G. and Lüpkes, C. (2015) Application of remotely piloted aircraft systems in observing the atmospheric boundary layer over Antarctic sea ice in winter. Polar Res 34(25651). doi: 10.3402/polar.v34.25651
- Kern S (2009) Wintertime Antarctic coastal polynya area: 1992–2008. Geophys Res Lett 36(L14501). doi: 10.1029/2009GL038062
- King JC, Turner J (1997) Antarctic meteorology and climatology. Cambridge Univ. Press 409 ppGoogle Scholar
- Klatt O, Roether W, Hoppema M, Bulsiewicz K, Fleischmann U, Rodehacke C, Fahrbach E, Weiss RF, Bullister JL (2002) Repeated CFC sections at the Greenwich meridian in the Weddell Sea. J Geophys Res 107. doi: 10.1029/2000JC000731
- König-Langlo G, Loose B (2007) The meteorological observatory at Neumayer Stations (GvN and NM-II) Antarctica. Polarforschung 76:25–38Google Scholar
- Lenaerts JTM, van den Broeke MR, van de Berg WJ, van Meijgaard E, Kuipers Munneke P (2012) A new, high-resolution surface mass balance map of Antarctica (1979–2010) based on regional atmospheric climate modeling. Geophys Res Lett 39(L04501). doi: 10.1029/2011GL050713
- Lo Monaco C, Metzl N, Poisson A, Brunet C, Schauer B (2005) Anhropogenic CO2 in the Southern Ocean: distribution and inventory at the Indian-Atlantic boundary (World Ocean Circulation Experiment line I6). J Geophys Res 110(C06010). doi: 10.1029/2004JC002643
- Lüpkes C, Vihma T, Birnbaum G, Wacker U (2008a) Influence of leads in sea ice on the temperature of the atmospheric boundary layer during polar night. Geophys Res Lett 35(L03805). doi: 10.1029/2007GL032461
- Lüpkes C, Gryanik VM, Witha B, Gryschka M, Raasch S, Gollnik T (2008b) Modeling convection over arctic leads with LES and a non-eddy-resolving microscale model. J Geophys Res 113(C09028). doi: 10.1029/2007JC004099
- Maronga B, Gryschka M, Heinze R, Hoffmann F, Kanani-Sühring F, Keck M, Ketelsen K, Letzel MO, Sühring S, Raasch S (2015) The parallelized large-Eddy simulation model (PALM) version 4.0 for atmospheric and oceanic flows: model formulation, recent developments, and future perspectives. Geosci Model Dev 8:2515–2551. doi: 10.5194/gmd-8-2515-2015 CrossRefGoogle Scholar
- Mensch M, Smethie WM, Schlosser P, Weppernig R, Bayer R (1998) Transient tracer observations from the western Weddell Sea during the drift and recovery of Ice Station Weddell. In: Jacobs SS, Weiss RF (eds) Ocean, ice and atmosphere: interactions at Antarctic continental margins, Ant Res Seri 75:241–256 AGU WashingtonGoogle Scholar
- Mironov DV (2009) Turbulence in the lower troposphere: second-order closure and mass-flux modelling frameworks. Lecture note in. Physics 756:161–221Google Scholar
- Nakicenovi N, Alcamo J, Davis G, De Vries B, Fenhann J, Gaffin S, Gregory K, Grübler A, Yong Jung T, Kram T, La Rovere EL, Michaelis L, Mori S, Morita T, Pepper W, Pitcher H, Price L, Riahi K, Roehrl A, Rogner H-H, Sankovski A, Schlesinger M, Shukla P, Smith S, Swart R, van Rooijen S, Victor N, Dadi Z (2000) Special report on emissions scenarios: a special report of working group III of the intergovernmental panel on climate change. Cambridge University Press, Cambridge, U.K. 599ppGoogle Scholar
- Oshima KI, Fukachami Y, Williams GD, Nihashi S, Roquet F, Kitade Y, Tamura T, Hirano D, Herraiz-Borreguero L, Field I, Hindell M, Aoki S, Wakatusucchi M (2013) Antarctic bottom water production by intense sea-ice formation in the cape Darnley polynya. Nature Geosci 6:235–240. doi: 10.1038/ngeo1738 CrossRefGoogle Scholar
- Rhein M, Rintoul SR, Aoki S, Campos E, Chambers D, Feely RA, Gulev S, Johnson GC, Josey SA, Kostianoy A, Mauritzen C, Roemmich D, Talley LD, Wang F (2013) Observations: Ocean. In: Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, Midgley PM (eds) Climate change 2013: the physical science basis. Contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change. Cambridge University Press, CambridgeGoogle Scholar
- Rintoul SR (2007) Rapid freshening of Antarctic bottom water formed in the Indian and Pacific Oceans. Geophys Res Lett 34(L06606). doi: 10.1029/2006GL028550
- Rodehacke C, Hellmer HH, Beckmann A, Roether W (2007) Formation and spreading of Antarctic deep and bottom waters inferred from a chlorofluorocarbon (CFC) simulation. J Geophys Res 112(C09001). doi: 10.1029/2006JC003884
- Rutgers van der Loeff MM, Cai P, Stimac I, Bracher A, Middag R, Klunder M, Van Heuven S (2011) 234Th in surface waters: distribution of particle export flux across the Antarctic circumpolar current and in the Weddell Sea during the GEOTRACES expedition ZERO and DRAKE. Deep-Sea Res II 58:2749–2766CrossRefGoogle Scholar
- Schlosser E, Duda MD, Powers JG, Manning KW (2008) Precipitation regime of Dronning Maud Land, Antarctica, derived from Antarctic Prediction System (AMPS) archive data. J Geophys Res 113(D24108). doi: 10.1029/2008JD009968
- Seckmeyer G, Bais A, Bernhard G, Blumthaler M, Lantz K, Mckenzie RL, Kiedron P, Drüke S, Riechelmann S (2010) Instruments to measure solar ultraviolet radiation, part 4: Array Spectroradiometers. 43 pp, WMO-GAW report 191, TD 5038Google Scholar
- Smetacek V, Klaas C, Strass V, Assmy P, Montresor M, Cisewski B, Savoye N, Webb A, D’ovidio F, Arrieta JM, Bathmann U, Bellerby R, Berg GM, Croot PL, Gonzalez S, Henjes J, Herndl GJ, Hoffmann LJ, Leach H, Losch M, Mills MM, Neill C, Peeken I, Roettgers R, Sachs O, Sauter E, Schmidt M, Schwarz JN, Terbrüggen A, Wolf-Gladrow D (2012) Deep carbon export from a Southern Ocean iron-fertilized diatom bloom. Nature 487(7407):313–319. doi: 10.1038/Nature11229 CrossRefGoogle Scholar
- Takahashi T, Sutherland SC, Wanninkhof R, Sweeney C, Feely RA, Chipman DW, Hales B, Friederich G, Chavez F, Sabine C, Watson A, Bakker DCE, Schuster U, Metzl N, Yoshikawa-Inoue H, Ishii M, Midorikawa T, Nojiri Y, Kortzinger A, Steinhoff T, Hoppema M, Olafsson J, Arnarson TS, Tilbrook B, Johannessen T, Olsen A, Bellerby R, Wong CS, Delille B, Bates NR, De Baar HJW (2009) Climatological mean and decadal change in surface ocean pCO2, and net sea-air CO2 flux over the global oceans. Deep-Sea Res II 56:554–577CrossRefGoogle Scholar
- Tamura T, Ohshima KI, Nihashi S (2008) Mapping of sea ice production for Antarctic coastal polynyas. Geophys Res Lett 35(L07606). doi: 10.1029/2007GL032903
- Thoma M, Determann J, Grosfeld K, Göller S, Hellmer HH (2015) Additional sea-level rise due to projected ocean warming beneath the Filchner Rønne ice shelf: a coupled model study. Earth Planet Sci Lett 431(217–224). doi: 10.1016/j.epsl.2015.09.013
- Tortell PD, Payne CD, Li Y, Trimborn S, Rost B, Smith WO, Risselman C, Dunbar R, Sedwick P, Di Tullio GR (2008) The CO2 sensitivity of Southern Ocean phytoplankton. Geophys Res Lett 35(L04605). doi: 10.1029/2007GL032583
- Turner J, Comiso JC, Marshall GJ, Lachlan-Cope TA, Bracegirdle T, Maksym T, Meredith MP, Wang Z, Orr A (2009) Non-annular atmospheric circulation change induced by stratospheric ozone depletion and its role in the recent increase of Antarctic sea ice extent. Geophys Res Lett 36(L08502). doi: 10.1029/2009GL037524
- Van de Berg WJ, van den Broeke MR, van Meijgaard E, Reijmer CH (2006) Reassessment of the Antarctic surface mass balance using calibrated output of a regional atmospheric climate model. J Geophys Res 111(D11104). doi: 10.1029/2005JD006495
- Van Heuven SMAC, Hoppema M, Jones EM, De Baar HJW (2014) Rapid invasion of anthropogenic CO2 into the deep circulation of the Weddell gyre. Phil Trans Royal Soc Lond A 372(20130056). doi: 10.1098/rsta.2013.0056
- Van Lipzig NPM, King JC, Lachlan-Cope TA, van den Broeke MR (2004) Precipitation, sublimation, and snow drift in the Antarctic Peninsula region from a regional atmospheric model. J Geophys Res 109(D24106). doi: 10.1029/2004JD004701
- Vihma T, Pirazzini R, Fer I, Renfrew IA, Sedlar J, Tjernström M, Lüpkes C, Nygard T, Notz D, Weiss J, Marsan D, Cheng B, Birnbaum G, Gerland S, Chechin D, Gascard JC (2014) Advances in understanding and parameterization of small-scale physical processes in the marine Arctic climate system: a review. Atmos Chem Phys 14:9403–9450. doi: 10.5194/acp-14-9403-2014 CrossRefGoogle Scholar
- WMO (World Meteorological Organization) (2014) Scientific assessment of ozone depletion: 2014, Global Ozone Research and Monitoring Project – Report No. 55 416 pp Geneva SwitzerlandGoogle Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.