The global carbon cycle comprises two parts: firstly, the human derived (anthropogenic) flux of about 8 PgCy^-1of which the oceans take up about 2 PgCy^-1, and the Southern Ocean contributes about 1PgCy^-1.  Secondly, a very much larger but up to now balanced ocean – atmosphere exchange of natural CO₂ of approximately 90GtCy^-1.

The steady-state magnitudes of the anthropogenic and natural CO2 fluxes are robustly constrained by the global and regional decadal mean data sets.  The science gap lies in the non-steady-state part of these fluxes: both in terms of variability and long-term trends.  SOCCO research is focused on understanding and quantifying the non-steady state part of the CO2 variability and trends.

Century-scale trends in atmospheric and ocean CO2 face a dual problem which impacts on global and regional scale mitigation of CO₂ emissions to limit climate change risk: not only will the uptake rate of anthropogenic CO₂ decrease because of changing CO₂ chemistry (Ocean Acidification) and warming but as importantly, climate forcing will begin to alter the ocean physics that controls the much larger natural CO₂ flux in the Southern Ocean.

Carbon research in SOCCO approaches these questions along three lines:

• It has established a long-term ship –based CO2 observations system making underway observations in the southeast Atlantic Ocean and the southwest Indian Ocean (Fig. 1).  These data are made available to the national (SADCO) and global (CDIAC) databases from where they are then integrated after 2^nd level QC into the SOCAT and later ocean acidification databases.  Through this SOCCO builds up a data set, which includes a wide range of ancillary variables, to support its own research as well as global community initiatives.
• As part of its scientific focus on fine scale ocean dynamics the CO2 research makes a strong contribution to dedicated experiments which aim to understand the sensitivity of the carbon cycle in and CO2 fluxes to the seasonal and intraseasonal dynamics of upper ocean physics (meso and sub-mesoscale).  In this domain we are exploring the use of robotics (surface wave gliders and ocean interior buoyancy gliders) to make observations within these spatial and temporal scale constraints.  The Southern Ocean Seasonal Cycle Experiment (SOSCEx) is our platform for these experiments, which target the core hypothesis of the programme: fine scale (carbon) – large scale (climate) links.
• Modelling: in SOCCO we undertake CO2 modelling research in two main areas: firstly, we explore the use of different empirical numerical / machine learning methods to address the need for high precision (< 0.1PgCy^-1) CO2 air – sea exchange fluxes in a data sparse system.  This part of a global effort to reduce the global uncertainty of CO2 fluxes to ~10% of the mean annual flux necessary to resolve inter-annual variability and long-term trends.   Secondly, we use a hierarchy of global and regional coarse (200km) to very high resolution (2km) model runs to test both scale sensitivity research questions for CO2 and the carbon cycle as well as explore the understanding and use of the seasonal cycle as a mode that provides a rigorous test to model outputs.

The Southern Ocean is one of the largest high-nutrient low-chlorophyll region in the Global Ocean, where primary productivity is primarily limited by iron availability. The spatio-temporal observations of iron in the Southern Ocean are restricted compared to other oceanic regions, primarily due to inaccessibility during winter, limiting our understanding of the full seasonal cycle. Seasonality plays an important role in determining the variability of the internal and external sources of iron in the Southern Ocean, which ultimately constrains phytoplankton biomass and primary productivity.

The primary focus of SOCCO’s iron biogeochemical research is as follows:

• Expand our measurements across the full seasonal cycle
• Increase our observations in naturally fertilised regions downstream of sub-Antarctic Islands such as Gough Island and Prince Edward Islands
• Measure the physical speciation of iron across the dissolved, soluble, colloidal and particulate size fractions
• Measure both iron and copper organic speciation
• Measure Fe(II) oxidation-kinetics which has yet to be performed in the Southern Ocean

Ocean colour remote sensing can provide routine, synoptic and highly cost-effective observations of biological and biogeochemical response to physical drivers across oceanic ecosystems, over decadal time scales and at high frequency. In many cases, remotely sensed data are the only systematic observations available for chronically under-sampled marine systems (e.g. the polar oceans), and there is thus a need to maximise the value of these observations by developing ecosystem-appropriate, well characterised products.

A primary focus of SOCCO’s bio-optical research is on gathering the necessary bio-optical and physiological data to develop and validate appropriate regional ocean colour algorithms for the Southern Ocean. This includes bio-optical data in the form of Inherent Optical Properties (IOP’s) (scattering, beam attenuation and absorption) and Apparent Optical Properties (AOP) (radiance, irradiance, reflectance, diffuse attenuation coefficient) and biogeochemical data that characterises the phytoplankton community (e.g. carbon content, size structure and dominant functional type). This information in conjunction with radiative transfer models and reflectance inversion algorithms will allow us to use satellite derived ocean colour data to investigate biological responses (through changes in biomass, community structure and physiology) to event, seasonal and inter-annual variability in ecosystem physical drivers at the required spatial and temporal scales. Given the important relationship between community size and carbon export these approaches will allow us to assess the potential for carbon cycling and carbon sequestration at the regional scale.

The biogeochemistry laboratory in the Oceanography department at UCT performs routine hydrographic measurements such as nutrient concentrations (NO₃, NO₂, NH₄, SiO₄, PO₄, urea), dissolved oxygen, chlorophyll and pH. An automated system (flow injection auto-analyser) is available for NO₃ and SiO₄ analyses while PO₄ and NH₄ is determined spectrophotometrically. In collaboration with the Marine Biogeochemistry Laboratory, nutrient cycling experiments are performed to look at primary production, nutrient uptake and nutrient regeneration. These experiments are performed using both stable isotopes (¹⁵N, ¹³C) and radioactive isotopes (¹⁴C, ⁵⁵Fe) in a custom certified radioisotope laboratory container.

In order to understand the spatial scales in the ocean that contribute to the atmosphere-ocean carbon exchange in the Southern Ocean, SOCCO uses a hierarchy of model configurations and model classes.

Processes and exchange in the upper ocean, are examined using the NEMO ocean modelling platform which includes coupled ocean, ice and biogeochemical models. A reference suite of varying resolution configurations that can address spatial scales have been produced: ranging from global coarse-resolution (2º) through regional mesoscale-resolving  (1/2º, 1/4º, 1/12º) to submesoscale-permitting localised regions (< 1/36º). The latter configuration corresponds to the domain of the high-resolution in situ sampling campaign of SOSCEX. This set of configurations form the baseline for sensitivity studies, further experimentation and model development.

Additionally, SOCCO uses the current generation of IPCC earth system model outputs for analysis.

Using these models as tools, researchers and students are able to understand processes and compare them to observations. These may help account for the differences between observations and the models that are used in long-term climate prediction.

Ocean and atmospheric physical characteristics and dynamics are crucial to understanding key SOCCO related scientific focus areas where a multi-disciplinary approach is taken to understanding the links between climate, biogeochemistry and ecosystems. SOCCO researchers pursue research focused predominantly on submesoscale (<10km) to mesoscale (10-200km) oceanographic processes that have an impact on upper ocean mixing and stratification dynamics and variability. Core to this science includes understanding the link between the ocean and the atmosphere through air-sea exchange and interaction. Our work also extends to the deeper ocean processes and ventilation, while laterally to the larger scale circulation of the Southern Ocean from the Subtropical to Antarctic sea-ice domains. These approaches make South Africa a leading contributor to Southern Hemisphere ocean and climate science.

SOCCO’s physics-related research is underpinned by an integrated approach, combining the use of numerical modelling simulations, ship-based observations and high-resolution measurements collected by autonomous ocean gliders and floats. Recent emphasis has been placed on resolving the seasonal cycle of upper ocean physical processes in the Southern Ocean and relating this to biogeochemical responses. This was undertaken through unique experimental design by deploying marine robotic instruments in the Southern Ocean that continuously observe the ocean and air-sea exchange processes for extended periods of time (6 months) and resolving the temporal and spatial scales of variability at unprecedentedly high resolution.

Our research contributes to South Africa’s developmental needs by using novel approaches in advanced observations, numerical modelling and analysis to train undergraduate and post graduate students.