Search

Quantifying the strength and efficiency of the Southern Ocean biological carbon pump (BCP) and its response to predicted changes in the Earth's climate is fundamental to our ability to predict long-term changes in the global carbon cycle and, by extension, the impact of continued anthropogenic perturbation of atmospheric CO2. There is little agreement, however, in climate model projections of the sensitivity of the Southern Ocean BCP to climate change, with a lack of consensus in even the direction of predicted change, highlighting a gap in our understanding of a major planetary carbon flux. In this review, we summarize relevant research that highlights the important role of fine-scale dynamics (both temporal and spatial) that link physical forcing mechanisms to biogeochemical responses that impact the characteristics of the seasonal cycle of phytoplankton and by extension the BCP. This approach highlights the potential for integrating autonomous and remote sensing observations of fine scale dynamics to derive regionally optimized biogeochemical parameterizations for Southern Ocean models. Ongoing development in both the observational and modelling fields will generate new insights into Southern Ocean ecosystem function for improved predictions of the sensitivity of the Southern Ocean BCP to climate change.This article is part of the discussion meeting issue 'Heat and carbon uptake in the Southern Ocean: the state of the art and future priorities'.

Resolving and understanding the drivers of variability of CO2 in the Southern Ocean and its potential climate feedback is one of the major scientific challenges of the ocean-climate community. Here we use a regional approach on empirical estimates of pCO2 to understand the role that seasonal variability has in long-term CO2 changes in the Southern Ocean. Machine learning has become the preferred empirical modelling tool to interpolate time- and location-restricted ship measurements of pCO2. In this study we use an ensemble of three machine-learning products: support vector regression (SVR) and random forest regression (RFR) from Gregor et al. (2017), and the self-organising-map feed-forward neural network (SOM-FFN) method from Landschützer et al. (2016). The interpolated estimates of ΔpCO2 are separated into nine regions in the Southern Ocean defined by basin (Indian, Pacific, and Atlantic) and biomes (as defined by Fay and McKinley, 2014a). The regional approach shows that, while there is good agreement in the overall trend of the products, there are periods and regions where the confidence in estimated ΔpCO2 is low due to disagreement between the products. The regional breakdown of the data highlighted the seasonal decoupling of the modes for summer and winter interannual variability. Winter interannual variability had a longer mode of variability compared to summer, which varied on a 4–6-year timescale. We separate the analysis of the ΔpCO2 and its drivers into summer and winter. We find that understanding the variability of ΔpCO2 and its drivers on shorter timescales is critical to resolving the long-term variability of ΔpCO2. Results show that ΔpCO2 is rarely driven by thermodynamics during winter, but rather by mixing and stratification due to the stronger correlation of ΔpCO2 variability with mixed layer depth. Summer pCO2 variability is consistent with chlorophyll a variability, where higher concentrations of chlorophyll a correspond with lower pCO2 concentrations. In regions of low chlorophyll a concentrations, wind stress and sea surface temperature emerged as stronger drivers of ΔpCO2. In summary we propose that sub-decadal variability is explained by summer drivers, while winter variability contributes to the long-term changes associated with the SAM. This approach is a useful framework to assess the drivers of ΔpCO2 but would greatly benefit from improved estimates of ΔpCO2 and a longer time series.

International audience; Accurate assessment of anthropogenic carbon dioxide (CO2) emissions and their redistribution among the atmosphere, ocean, and terrestrial biosphere – the "global carbon budget" – is important to better understand the global carbon cycle, support the development of climate policies, and project future climate change. Here we describe data sets and methodology to quantify the five major components of the global carbon budget and their uncertainties. CO2 emissions from fossil fuels and industry (EFF) are based on energy statistics and cement production data, respectively, while emissions from land-use change (ELUC), mainly deforestation, are based on land-cover change data and bookkeeping models. The global atmospheric CO2 concentration is measured directly and its rate of growth (GATM) is computed from the annual changes in concentration. The ocean CO2 sink (SOCEAN) and terrestrial CO2 sink (SLAND) are estimated with global process models constrained by observations. The resulting carbon budget imbalance (BIM), the difference between the estimated total emissions and the estimated changes in the atmosphere, ocean, and terrestrial biosphere, is a measure of imperfect data and understanding of the contemporary carbon cycle. All uncertainties are reported as ±1σ. For the last decade available (2007–2016), EFF was 9.4 ± 0.5 GtC yr−1, ELUC 1.3 ± 0.7 GtC yr−1, GATM 4.7 ± 0.1 GtC yr−1, SOCEAN 2.4 ± 0.5 GtC yr−1, and SLAND 3.0 ± 0.8 GtC yr−1, with a budget imbalance BIM of 0.6 GtC yr−1 indicating overestimated emissions and/or underestimated sinks. For year 2016 alone, the growth in EFF was approximately zero and emissions remained at 9.9 ± 0.5 GtC yr−1. Also for 2016, ELUC was 1.3 ± 0.7 GtC yr−1, GATM was 6.1 ± 0.2 GtC yr−1, SOCEAN was 2.6 ± 0.5 GtC yr−1, and SLAND was 2.7 ± 1.0 GtC yr−1, with a small BIM of −0.3 GtC. GATM continued to be higher in 2016 compared to the past decade (2007–2016), reflecting in part the high fossil emissions and the small SLAND consistent with El Niño conditions. The global atmospheric CO2 concentration reached 402.8 ± 0.1 ppm averaged over 2016. For 2017, preliminary data for the first 6–9 months indicate a renewed growth in EFF of +2.0 % (range of 0.8 to 3.0 %) based on national emissions projections for China, USA, and India, and projections of gross domestic product (GDP) corrected for recent changes in the carbon intensity of the economy for the rest of the world. This living data update documents changes in the methods and data sets used in this new global carbon budget compared with previous publications of this data set (Le Quéré et al., 2016, 2015b, a, 2014, 2013). All results presented here can be downloaded from https://doi.org/10.18160/GCP-2017 (GCP, 2017).

The Southern Ocean accounts for 40% of oceanic CO2 uptake, but the estimates are bound by large uncertainties due to a paucity in observations. Gap-filling empirical methods have been used to good effect to approximate pCO(2) from satellite observable variables in other parts of the ocean, but many of these methods are not in agreement in the Southern Ocean. In this study we propose two additional methods that perform well in the Southern Ocean: support vector regression (SVR) and random forest regression (RFR). The methods are used to estimate Delta pCO(2) in the Southern Ocean based on SOCAT v3, achieving similar trends to the SOM-FFN method by Landschitzer et al. (2014). Results show that the SOM-FFN and RFR approaches have RMSEs of similar magnitude (14.84 and 16.45 mu atm, where 1 atm D 101 325 Pa) where the SVR method has a larger RMSE (24.40 mu atm). However, the larger errors for SVR and RFR are, in part, due to an increase in coastal observations from SOCAT v2 to v3, where the SOM-FFN method used v2 data. The success of both SOM-FFN and RFR depends on the ability to adapt to different modes of variability. The SOM-FFN achieves this by having independent regression models for each cluster, while this flexibility is intrinsic to the RFR method. Analyses of the estimates shows that the SVR and RFR's respective sensitivity and robustness to out-liers define the outcome significantly. Further analyses on the methods were performed by using a synthetic dataset to assess the following: which method (RFR or SVR) has the best performance? What is the effect of using time, latitude and longitude as proxy variables on Delta pCO(2)? What is the impact of the sampling bias in the SOCAT v3 dataset on the estimates? We find that while RFR is indeed better than SVR, the ensemble of the two methods outperforms either one, due to complementary strengths and weaknesses of the methods. Results also show that for the RFR and SVR implementations, it is better to include coordinates as proxy variables as RMSE scores are lowered and the phasing of the seasonal cycle is more accurate. Lastly, we show that there is only a weak bias due to undersampling. The synthetic data provide a useful framework to test methods in regions of sparse data coverage and show potential as a useful tool to evaluate methods in future studies.

International audience

The Southern Ocean accounts for 40 % of oceanic CO2 uptake, but the estimates are bound by large uncertainties due to a paucity in observations. Gap-filling empirical methods have been used to good effect to approximate pCO2 from satellite observable variables in other parts of the ocean, but many of these methods are not in agreement in the Southern Ocean. In this study we propose two additional methods that perform well in the Southern Ocean: support vector regression (SVR) and random forest regression (RFR). The methods are used to estimate ΔpCO2 in the Southern Ocean based on SOCAT v3, achieving similar trends to the SOM-FFN method by LandschYtzer et al. (2014). Results show that the SOM-FFN and RFR approaches have RMSEs of similar magnitude (14.84 and 16.45 µatm, where 1 atm = 101 325 Pa) where the SVR method has a larger RMSE (24.40 µatm). However, the larger errors for SVR and RFR are, in part, due to an increase in coastal observations from SOCAT v2 to v3, where the SOM-FFN method used v2 data. The success of both SOM-FFN and RFR depends on the ability to adapt to different modes of variability. The SOM-FFN achieves this by having independent regression models for each cluster, while this flexibility is intrinsic to the RFR method. Analyses of the estimates shows that the SVR and RFR's respective sensitivity and robustness to outliers define the outcome significantly. Further analyses on the methods were performed by using a synthetic dataset to assess the following: which method (RFR or SVR) has the best performance? What is the effect of using time, latitude and longitude as proxy variables on ΔpCO2? What is the impact of the sampling bias in the SOCAT v3 dataset on the estimates? We find that while RFR is indeed better than SVR, the ensemble of the two methods outperforms either one, due to complementary strengths and weaknesses of the methods. Results also show that for the RFR and SVR implementations, it is better to include coordinates as proxy variables as RMSE scores are lowered and the phasing of the seasonal cycle is more accurate. Lastly, we show that there is only a weak bias due to undersampling. The synthetic data provide a useful framework to test methods in regions of sparse data coverage and show potential as a useful tool to evaluate methods in future studies.

The Surface Ocean CO2 Atlas (SOCAT) is a synthesis of quality-controlled fCO2 (fugacity of carbon dioxide) values for the global surface oceans and coastal seas with regular updates. Version 3 of SOCAT has 14.7 million fCO2 values from 3646 data sets covering the years 1957 to 2014. This latest version has an additional 4.6 million fCO2 values relative to version 2 and extends the record from 2011 to 2014. Version 3 also significantly increases the data availability for 2005 to 2013. SOCAT has an average of approximately 1.2 million surface water fCO2 values per year for the years 2006 to 2012. Quality and documentation of the data has improved. A new feature is the data set quality control (QC) flag of E for data from alternative sensors and platforms. The accuracy of surface water fCO2 has been defined for all data set QC flags. Automated range checking has been carried out for all data sets during their upload into SOCAT. The upgrade of the interactive Data Set Viewer (previously known as the Cruise Data Viewer) allows better interrogation of the SOCAT data collection and rapid creation of high-quality figures for scientific presentations. Automated data upload has been launched for version 4 and will enable more frequent SOCAT releases in the future. High-profile scientific applications of SOCAT include quantification of the ocean sink for atmospheric carbon dioxide and its long-term variation, detection of ocean acidification, as well as evaluation of coupled-climate and ocean-only biogeochemical models. Users of SOCAT data products are urged to acknowledge the contribution of data providers, as stated in the SOCAT Fair Data Use Statement. This ESSD (Earth System Science Data) "living data" publication documents the methods and data sets used for the assembly of this new version of the SOCAT data collection and compares these with those used for earlier versions of the data collection (Pfeil et al., 2013; Sabine et al., 2013; Bakker et al., 2014). Individual data set files, included in the synthesis product, can be downloaded here: doi:10.1594/PANGAEA.849770. The gridded products are available here: doi:10.3334/CDIAC/OTG.SOCAT_V3_GRID.

International audience

Here we study the effect of La doping in EuO thin films using SQUID magnetometry, muon spin rotation (µSR), polarized neutron reflectivity (PNR), and density functional theory (DFT). The µSR data shows that the La0.15Eu0.85O is homogeneously magnetically ordered up to its elevated TC. It is concluded that bound magnetic polaron behavior does not explain the increase in TC and an RKKY-like interaction is consistent with the µSR data. The estimation of the magnetic moment by DFT simulations concurs with the results obtained by PNR, showing a reduction of the magnetic moment per LaxEu1−xO for increasing lanthanum doping. This reduction of the magnetic moment is explained by the reduction of the number of Eu-4f electrons present in all the magnetic interactions in EuO films. Finally, we show that an upwards shift of the Fermi energy with La or Gd doping gives rise to half metallicity for doping levels as high as 3.2%.

Here we study the effect of La doping in EuO thin films using SQUID magnetometry, muon spin rotation ($\mu$SR), polarized neutron reflectivity (PNR), and density functional theory (DFT). The $\mu$SR data shows that the La$_{0.15}$Eu$_{0.85}$O is homogeneously magnetically ordered up to its elevated $T_{\rm C}$. It is concluded that bound magnetic polaron behavior does not explain the increase in $T_{\rm C}$ and an RKKY-like interaction is consistent with the $\mu$SR data. The estimation of the magnetic moment by DFT simulations concurs with the results obtained by PNR, showing a reduction of the magnetic moment per La$_{x}$Eu$_{1-x}$O for increasing lanthanum doping. This reduction of the magnetic moment is explained by the reduction of the number of Eu-4$f$ electrons present in all the magnetic interactions in EuO films. Finally, we show that an upwards shift of the Fermi energy with La or Gd doping gives rise to half-metallicity for doping levels as high as 3.2 %., Comment: 7 pages, 11 figures

We have used low energy implanted muons as a volume sensitive probe of the magnetic properties of EuO_{1-x} thin films. We find that static and homogeneous magnetic order persists up to the elevated T_C in the doped samples and the muon signal displays the double dome feature also observed in the sample magnetization. Our results appear incompatible with either the magnetic phase separation or bound magnetic polaron descriptions previously suggested to explain the elevated T_C, but are compatible with an RKKY-like interaction mediating magnetic interactions above 69 K., 5 pages, 10 pages of supplementary information, 11 figures

This study was undertaken to characterise the seasonal cycle of air–sea fluxes of carbon dioxide (CO2 ) in the southern Benguela upwelling system off the South African west coast. Samples were collected from six monthly cross-shelf cruises in the St. Helena Bay region during 2010. CO2 fluxes were calculated from pCO2 derived from total alkalinity and dissolved inorganic carbon and scatterometer-based winds. Notwithstanding that it is one of the most biologically productive eastern boundary upwelling systems in the global ocean, the southern Benguela was found to be a very small net annual CO2 sink of -1.4 ± 0.6 mol C/m2 per year (1.7 Mt C/year). Regional primary productivity was offset by nearly equal rates of sediment and sub-thermocline remineralisation flux of CO2 , which is recirculated to surface waters by upwelling. The juxtaposition of the strong, narrow near-shore out-gassing region and the larger, weaker offshore sink resulted in the shelf area being a weak CO2 sink in all seasons but autumn (-5.8, 1.4 and -3.4 mmol C/m2 per day for summer, autumn and winter, respectively).

Bibliography: leaf 78-83., Analysis and interpretation was carried out on trace metal samples (Cu,Zn,Cd) collected in the Southern Ocean south of South Africa. The results tentatively show that there is a significant copper flux from interstitial waters of the Weddell - Enderby basins. There is little evidence that these trace metals follow the systematics of nutrients throughout the water column; advection appears to play an important role in determining the vertical distribution of these chemical components.

Accurate assessment of anthropogenic carbon dioxide (CO2) emissions and their redistribution among the atmosphere, ocean, and terrestrial biosphere – the “global carbon budget” – is important to better understand the global carbon cycle, support the development of climate policies, and project future climate change. Here we describe data sets and methodology to quantify all major components of the global carbon budget, including their uncertainties, based on the combination of a range of data, algorithms, statistics, and model estimates and their interpretation by a broad scientific community. We discuss changes compared to previous estimates and consistency within and among components, alongside methodology and data limitations. CO2 emissions from fossil fuels and industry (EFF) are based on energy statistics and cement production data, respectively, while emissions from land-use change (ELUC), mainly deforestation, are based on combined evidence from land-cover change data, fire activity associated with deforestation, and models. The global atmospheric CO2 concentration is measured directly and its rate of growth (GATM) is computed from the annual changes in concentration. The mean ocean CO2 sink (SOCEAN) is based on observations from the 1990s, while the annual anomalies and trends are estimated with ocean models. The variability in SOCEAN is evaluated with data products based on surveys of ocean CO2 measurements. The global residual terrestrial CO2 sink (SLAND) is estimated by the difference of the other terms of the global carbon budget and compared to results of independent dynamic global vegetation models. We compare the mean land and ocean fluxes and their variability to estimates from three atmospheric inverse methods for three broad latitude bands. All uncertainties are reported as ±1σ, reflecting the current capacity to characterise the annual estimates of each component of the global carbon budget. For the last decade available (2006–2015), EFF was 9.3 ± 0.5 GtC yr−1, ELUC 1.0 ± 0.5 GtC yr−1, GATM 4.5 ± 0.1 GtC yr−1, SOCEAN 2.6 ± 0.5 GtC yr−1, and SLAND 3.1 ± 0.9 GtC yr−1. For year 2015 alone, the growth in EFF was approximately zero and emissions remained at 9.9 ± 0.5 GtC yr−1, showing a slowdown in growth of these emissions compared to the average growth of 1.8 % yr−1 that took place during 2006–2015. Also, for 2015, ELUC was 1.3 ± 0.5 GtC yr−1, GATM was 6.3 ± 0.2 GtC yr−1, SOCEAN was 3.0 ± 0.5 GtC yr−1, and SLAND was 1.9 ± 0.9 GtC yr−1. GATM was higher in 2015 compared to the past decade (2006–2015), reflecting a smaller SLAND for that year. The global atmospheric CO2 concentration reached 399.4 ± 0.1 ppm averaged over 2015. For 2016, preliminary data indicate the continuation of low growth in EFF with +0.2 % (range of −1.0 to +1.8 %) based on national emissions projections for China and USA, and projections of gross domestic product corrected for recent changes in the carbon intensity of the economy for the rest of the world. In spite of the low growth of EFF in 2016, the growth rate in atmospheric CO2 concentration is expected to be relatively high because of the persistence of the smaller residual terrestrial sink (SLAND) in response to El Niño conditions of 2015–2016. From this projection of EFF and assumed constant ELUC for 2016, cumulative emissions of CO2 will reach 565 ± 55 GtC (2075 ± 205 GtCO2) for 1870–2016, about 75 % from EFF and 25 % from ELUC. This living data update documents changes in the methods and data sets used in this new carbon budget compared with previous publications of this data set (Le Quéré et al., 2015b, a, 2014, 2013). All observations presented here can be downloaded from the Carbon Dioxide Information Analysis Center (doi:10.3334/CDIAC/GCP_2016).

Accurate assessmen of anthropogenic carbon dioxide (CO₂) emissions and their redistribution among the atmosphere, ocean, and terrestrial biosphere – the “global carbon budget” – is important to betterunderstand the global carbon cycle, support the development of climate policies, and project future climate change. Here we describe data sets and methodology to quantify all major components of the global carbon budget, including their uncertainties, based on the combination of a range of data, algorithms, statistics, and model estimates and their interpretation by a broad scientific community. We discuss changes compared to previous estimates and consistency within and among components, alongside methodology and data limitations. CO₂ emissions from fossil fuels and industry (EFF) are based on energy statistics and cement production data, respec- tively, while emissions from land-use change (ELUC), mainly deforestation, are based on combined evidence from land-cover change data, fire activity associated with deforestation, and models. The global atmospheric CO₂ concentration is measured directly and its rate of growth (GATM) is computed from the annual changes in concentration. The mean ocean CO₂ sink (SOCEAN) is based on observations from the 1990s, while the an- nual anomalies and trends are estimated with ocean models. The variability in SOCEAN is evaluated with data products based on surveys of ocean CO₂ measurements. The global residual terrestrial CO₂ sink (SLAND) is estimated by the difference of the other terms of the global carbon budget and compared to results of independent dynamic global vegetation models. We compare the mean land and ocean fluxes and their variability to estimates from three atmospheric inverse methods for three broad latitude bands. All uncertainties are reported as ± 1σ, reflecting the current capacity to characterise the annual estimates of each component of the global carbon budget. For the last decade available (2006–2015), EFF was 9.3± 0.5 GtC yr⁻¹, ELUC 1.0± 0.5 GtC yr⁻¹,GATM 4.5± 0.1 GtC yr⁻¹, SOCEAN 2.6± 0.5 GtC yr⁻¹, and SLAND 3.1± 0.9 GtC yr⁻¹. For year 2015 alone, the growth in EFF was approximately zero and emissions remained at 9.9± 0.5 GtC yr⁻¹, showing a slowdown in growth of these emissions compared to the average growth of 1.8 % yr⁻¹ that took place during 2006–2015. Also, for 2015, ELUC was 1.3± 0.5 GtC yr⁻¹,GATM was 6.3± 0.2 GtC yr⁻¹, SOCEAN was 3.0± 0.5 GtC yr⁻¹, and SLAND was 1.9± 0.9 GtC yr⁻¹ GATM was higher in 2015 compared to the past decade (2006–2015), reflecting a smaller SLAND for that year. The global atmospheric CO₂ concentration reached 399.4± 0.1 ppm averaged over 2015. For 2016, preliminary data indicate the continuation of low growth in EFF with +0.2 % (range of −1.0 to + 1.8 %) based on national emissions projections for China and USA, and projections of gross domestic product corrected for recent changes in the carbon intensity of the economy for the rest of the world. In spite of the low growth of EFF in 2016, the growth rate in atmospheric CO₂ concentration is expected to be relatively high because of the persistence of the smaller residual terrestrial sink (SLAND) in response to El Niño conditions of 2015–2016. From this projection of EFF and assumed constant ELUC for 2016, cumulative emissions of CO₂ will reach 565± 55 GtC (2075± 205 GtCO₂) for 1870–2016, about 75 % from EFF and 25 % from ELUC. This living data update documents changes in the methods and data sets used in this new carbon budget compared with previous publications of this data set (Le Quéré et al., 2015b, a, 2014, 2013).

Accurate assessment of anthropogenic carbon dioxide (CO₂) emissions and their redistribution among the atmosphere, ocean, and terrestrial biosphere – the “global carbon budget” – is important to better understand the global carbon cycle, support the development of climate policies, and project future climate change. Here we describe data sets and methodology to quantify the five major components of the global carbon budget and their uncertainties. CO₂ emissions from fossil fuels and industry (EFF) are based on energy statistics and cement production data, respectively, while emissions from land-use change (ELUC), mainly deforestation, are based on land-cover change data and bookkeeping models. The global atmospheric CO₂ concentration is measured directly and its rate of growth (GATM) is computed from the annual changes in concentration. The ocean CO₂ sink (SOCEAN) and terrestrial CO₂ sink (SLAND) are estimated with global process models constrained by observations. The resulting carbon budget imbalance (BIM), the difference between the estimated total emissions and the estimated changes in the atmosphere, ocean, and terrestrial biosphere, is a measure of imperfect data and understanding of the contemporary carbon cycle. All uncertainties are reported as ±1σ. For the last decade available (2007–2016), EFF was 9.4 ± 0.5 GtC yr⁻¹, ELUC 1.3 ± 0.7 GtC yr⁻¹, GATM 4.7 ± 0.1 GtC yr⁻¹, SOCEAN 2.4 ± 0.5 GtC yr⁻¹, and SLAND 3.0 ± 0.8 GtC yr⁻¹, with a budget imbalance BIM of 0.6 GtC yr⁻¹ indicating overestimated emissions and/or underestimated sinks. For year 2016 alone, the growth in EFF was approximately zero and emissions remained at 9.9 ± 0.5 GtC yr⁻¹. Also for 2016, ELUC was 1.3 ± 0.7 GtC yr⁻¹, GATM was 6.1 ± 0.2 GtC yr⁻¹, SOCEAN was 2.6 ± 0.5 GtC yr⁻¹, and SLAND was 2.7 ± 1.0 GtC yr⁻¹, with a small BIM of −0.3 GtC. GATM continued to be higher in 2016 compared to the past decade (2007–2016), reflecting in part the high fossil emissions and the small SLAND consistent with El Niño conditions. The global atmospheric CO₂ concentration reached 402.8 ± 0.1 ppm averaged over 2016. For 2017, preliminary data for the first 6–9 months indicate a renewed growth in EFF of +2.0 % (range of 0.8 to 3.0 %) based on national emissions projections for China, USA, and India, and projections of gross domestic product (GDP) corrected for recent changes in the carbon intensity of the economy for the rest of the world. This living data update documents changes in the methods and data sets used in this new global carbon budget compared with previous publications of this data set (Le Quéré et al., 2016, 2015b, a, 2014, 2013). All results presented here can be downloaded from https://doi.org/10.18160/GCP-2017 (GCP, 2017).

The Southern Ocean forms a vital component of the earth system as a sink of CO2 and heat, taking over 40% of the annual oceanic CO2 uptake (75% of global heat uptake), slowing down the accumulation of CO2 in the atmosphere and thus the rate of climate change. However, recent studies based on the Coupled Model Intercomparison Project version 5 (CMIP5) Earth System Models (ESMs) show that CMIP5 ESMs disagree on the phasing of the seasonal cycle of the CO2 flux (FCO2) and compare poorly with available observation estimates in the Southern Ocean. Notwithstanding these differences, the seasonal cycle is a dominant mode of CO2 variability in the Southern Ocean, and hence this is an important bias. Previous studies suggest that these biases of FCO2 in ESMs might be a significant limitation to the long-term simulation of CO2 characteristics in the Southern Ocean. Consequently, this study has three primary objectives: first, to develop a process-based diagnostic method to analyze and isolate key biases and their underlaying mechanisms in the model-observations seasonal cycle of FCO2 differences for forced ocean models and ESMs. Second, to use this framework to examine sources of biases responsible for the limited skill of CMIP5 models in simulating the seasonal cycle of FCO2 with respect to observed estimates. Thirdly, to investigate how these present-day biases in the seasonality and drivers of CO2 in CMIP5 ESMs affect modelled longterm changes in the mechanisms of CO2 uptake in the Southern Ocean. In the first part of the dissertation, an objective diagnostic framework was established to analyze model-observation biases in the seasonal scale of FCO2 using the NEMO PISCES ORCA2LP model output, and Takahashi et al. (2009) observed estimates. The diagnostic framework focuses on examining the relative contributions of the competing drivers (SST and DIC) and related processes (solubility, biological and mixing) to instantaneous monthly changes in surface pCO2 (and FCO2) at the seasonal scale. In the second part of the dissertation, this approach is applied to 10 CMIP5 models in the Southern Ocean, to investigate the mechanistic basis for the seasonal cycle of FCO2 biases. It was found that FCO2 biases in CMIP5 models can be grouped into two main categories, i.e. group-SST and group-DIC. Group-SST models are characterized by an exaggeration of the seasonal rates of change of Sea Surface Temperature (SST) in autumn and spring during the cooling and warming peaks, respectively. These faster-than-observed rates of change of SST tip the control of the seasonal cycle of pCO2 and FCO2 towards SST and result in divergence between the observed and modelled seasonal cycles, particularly in the Sub-Antarctic Zone. While almost all analyzed models show these SST-driven biases, 3 out of 10 (namely NorESM1-ME, HadGEM2-ES and MPI-ESM, collectively the group-DIC models) compensate the solubility bias because of their exaggerated primary production, such that biologically-driven DIC changes become the regulators of the seasonal cycle of FCO2. It was also found that despite significant differences in the spatial characteristics of the mean annual fluxes, CMIP5 models show a zonal homogeneity in the seasonal cycle of FCO2 at the basin-scale in contrast to observed estimates. In the final third of the dissertation, using five CMIP5 ESMs from the RCP8.5 scenario, it was found that CMIP5 models present climate biases in the seasonality and drivers of FCO2 are fundamental to how models simulate long-term changes in the mechanisms of CO2 uptake in the Southern Ocean. Although all five analyzed models show an increased annual mean CO2 uptake by the end of the century, they show significant differences in the mechanisms. The present-day temperature biased models (group-SST) generally maintain the dominance of the temperature driver in the seasonal variability of FCO2 to end of the century. But show enhanced CO2 uptake due to increased anthropogenic atmospheric CO2 and decreased surface CO2 buffering capacity but they display a weak to null role of biological activity in the increased CO2 sink. On the other hand, the increased CO2 uptake at the end of the century in group-DIC models is explained increased biological driven CO2 uptake in spring, linked to increased Revelle factor and solubility driven CO2 uptake in winter. Increased Revelle factor at the end of the century enhance pCO2 changes for even smaller DIC changes.

The Southern Ocean plays an important role in mitigating the effects of anthropogenically driven climate change. The region accounts for 43% of oceanic uptake of anthropogenic carbon dioxide (CO₂). This is foreseen to change with increasing greenhouse gas emissions due to ocean chemistry and climate feedbacks that regulate the carbon cycle in the Southern Ocean. Studies have already shown that Southern Ocean CO₂ is subject to interannual variability. Measuring and understanding this change has been difficult due to sparse observational data that is biased toward summer. This leaves a crucial gap in our understanding of the Southern Ocean CO₂ seasonal cycle, which needs to be resolved to adequately monitor change and gain insight into the drivers of interannual variability. Machine learning has been successful in estimating CO₂ in may parts of the ocean by extrapolating existing data with satellite measurements of proxy variables of CO₂. However, in the Southern Ocean machine learning has proven less successful. Large differences between machine learning estimates stem from the paucity of data and complexity of the mechanisms that drive CO₂. In this study the aim is to reduce the uncertainty of estimates, advance our understanding of the interannual drivers, and optimise sampling of CO₂ in the Southern Ocean. Improving the estimates of CO₂ was achieved by investigating: the impact of increasing the gridding resolution of input data and proxy variables, and Support vector regression (SVR) and Random Forest Regression (RFR) as alternate machine learning methods. It was found that the improvement gained by increasing gridding resolution was minimal and only RFR was able to improve on existing error estimates. Yet, there was good agreement of the seasonal cycle and interannual trends between RFR, SVR and estimates from the literature. The ensemble mean of these methods was used to investigate the variability and interannual trends of CO₂ in the Southern Ocean. The interannual trends of the ensemble confirmed trends reported in the literature. A weakening of the sink in the early 2000's, followed by a strengthening a strengthening of the sink into the early 2010's. Wind was the overall driver of dominant decadal interannual trends, being more important during winter due to the increased efficacy of entrainment processes. Summer interannual variability of CO₂ was driven primarily by chlorophyll, which responded to basin scale changes in drivers by the complex interaction with underlying physics and possibly sub-mesoscale processes. Lastly CO₂ sampling platforms, namely ships, profiling floats and moorings, were tested in an idealised simulated model environment using a machine learning approach. Ships, simulated from existing cruise tracks, failed to adequately resolve CO₂ below the uncertainty threshold that is required to resolve the seasonal cycle of Southern Ocean CO₂. Eight high frequency sampling moorings narrowly outperformed 200 profiling floats, which were both able to adequately resolve the seasonal cycle. Though, a combination of ships and profiling floats achieved the smallest error.

Includes bibliographical references., The two principal bottom-up drivers of the High Nutrient Low Chlorophyll (HNLC) characteristics of the Southern Ocean are light and nutrient (mainly dissolved iron) limitation ( Boyd , 2002; Mitchell et al., 1991), which have varying limiting roles over the growing season ( Boyd, 2002; Swart et al., 2014). This research commenced with an investigation of the meridional characteristics of primary productivity in the Atlantic Southern Ocean during austral summer 2008.

Includes bibliographical references, A recent study by Lenton et al., 2013, compared the mean seasonal cycle of air-sea CO₂ flux in the Southern Ocean(SO) over 1990 – 2009 period using five ocean biogeochemical models(BGMs) and atmospheric and ocean inversion models with monthly mean observations for the year 2000. This was done using a set of geographic boundaries to defined sub-domains of the SO consistent with the Regional Carbon Cycle and Assessment and Processes (RECCAP) protocol. Lenton et al., 2013 found that the seasonal cycle anomaly of the five BGMs better resolved observations of the air-sea CO₂ flux seasonal cycle in the SAZ, but was generally out phase with observations in the polar zone. In this study two setups of the ocean biogeochemical model NEMO PISCES was used to investigate the characteristics of the air-sea CO₂ flux seasonal cycle in the Southern Ocean in the period 1993- 2006. The study focused on two aspects i.e. (i) the sensitivity of air-sea CO₂ flux seasonal cycle to model resolution: comparing the ORCA2-LIM-PISCES (2° x 2° cos Ø) and PERIANT05 (NEMO-PISCES) (0.5° x 0.5° cos Ø) model configurations relative to climatological mean observations for the year 2000 (Takahashi et al., 2009) , and (ii) the sensitivity of air-sea CO₂ flux seasonal cycle to zonal boundary definition: comparing the air-sea CO₂ flux seasonal cycle and annual fluxes for three different boundaries i.e. Lenton 2013 RECCAP boundaries (44°S – 58°S and south of 58°S), geographic boundaries (40°S -50°S and south of 50°S) and dynamic boundaries (Sub-Antarctic Zone and Antarctic Zone, defined using climatological frontal positions). The seasonal cycle of the air-sea CO₂ flux in ORCA2 was found to be out of phase and overestimated the CO₂ flux compared to observations in almost all the sub-regions considered. The use of dynamic boundaries was found not to improve resolving observations seasonal cycle of air-sea CO₂ flux in both ORCA2 and PERIANT05. Boundary definition was found to affect the magnitude of ORCA2 annual air-sea CO₂ fluxes surface area based, where sub-regions of larger surface area gave larger annual CO₂ uptake and vice versa. This was mainly because ORCA2 air-sea CO₂ fluxes were found to show a general CO₂ in-gassing bias and spatially uniform in most parts of the SO and hence integration over a larger surface area gave larger annual fluxes. On the contrary PERIANT05 air-sea CO₂ fluxes spatial variability was not uniform in most parts of the SO however influenced by regional processes and hence annual fluxes were found not surface area based. The poor spatial representation and seasonal cycle sensitivity of ORCA2 air-sea CO₂ fluxes was found to be primarily due to lack or weak winter CO₂ entrainment and biological CO₂ draw down during the summer season. PERIANT05 on the contrary showed the effect of winter CO₂ entrainment, however maintains lack of or weak biological CO₂ draw down in the seasonal cycle. PERIANT05 was also found to show major weakness in the spatial representation of air-sea CO₂ fluxes north of the polar front with relative to T09 observations.

Includes bibliographical references., A model study was undertaken to investigate the optimization of sampling strategies for returning low-uncertainty sea-air CO₂ flux measurements in the Southern Ocean. Replicating Lenton et al. (2006) using the ORCA2/PISCES ocean biological model shows that sampling 4 times a year, every 2⁰ in latitude and every 40⁰ in longitude reduces the uncertainty of estimating annual CO₂ flux estimates such that sampling at a higher frequency does not reduce the total uncertainty in proportion to the increase in sampling effort.

Includes bibliographical references., An observational study was undertaken to determine the seasonality of the marine carbonate system of the southern Benguela focusing on three key points: the processes driving bulk stoichiometry, alkalinity production on the continental shelf, and the air-sea flux of CO2. Monthly samples were taken along the St. Helena Bay Monitoring Line in the southern Benguela for ten of the months in 2010. Samples were analysed for dissolved inorganic carbon (DIC) and total alkalinity (TA). Temperature, salinity, oxygen and nutrients were also measured.

Includes bibliographical references., Ocean Acidification through the uptake of anthropogenic CO₂ is resulting in a decrease in surface water carbonate ion concentration, a critical compound for marine calcifying organisms (Fabry et al., 2008; Orr et al., 2005). Natural seasonal variability is predicted to hasten the effects of Ocean Acidification in the Southern Ocean, resulting in possible surface water wintertime aragonite (the more soluble form of calcium carbonate) undersaturation (Ωarag< 1) south of the Antarctic Polar Front by the year 2030 (McNeil and Matear, 2008). An Ocean Acidification study was conducted to determine the seasonal and interannual variability in aragonite saturation state (Ωarag) at the Antarctic ice shelf between 4°E and 14°W and in the Eastern Weddell Gyre, during the Austral summers of 2008/2009, 2009/2010, 2010/2011 and 2011/2012. This study shows that at the Antarctic ice shelf andin the Eastern Weddell Gyre (EWG), seasonal summertime phytoplankton blooms were a critical factor in the observed decrease in summer surface water CO 2 and the subsequent increase in summer surface water Ωarag.

Includes bibliographical references (leaves 82-89)., The Central Benguela inner shelf is characterised by hypoxic to anoxic sub-thermocline water conditions on a semi-permanent basis. Historical understanding of the incidence of hypoxia is that the inner shelf area off Namibia is one of the main areas of formation of low oxygen water (LOW) in the Benguela upwelling system. Local biogeochemical remineralization of organic matter mostly related to the primary production in surface waters (Chapman and Shannon, 1985) is thought to drive seasonal variability in sediment oxygen demand. New understanding of the system suggests that shelf hypoxia is driven by a remotely forced equatorial hypoxic boundary condition thought to trigger anaerobic remineralization and increased sediment fluxes of reduced metabolites (Monteiro et al., in press). The study focused on seasonal variability of sediment oxygen demand on the Namibian inner shelf and its relation to particulate organic carbon and reduced metabolite fluxes.

Bibliography: leaves 33-34., An ADCP and water-column temperature study was conducted to determine the circulation aspect of subtidal-scale, density-driven exchange in Saldanha Bay, South Africa. Density-driven exchange conditions develop in response to synoptic-scale wind events in the southern Benguela region, even under light (