OS3.1

Cyanobacteria of the genus Trichodesmium provide about 80 Tg of fixed nitrogen to the surface ocean per year and contribute to marine biogeochemistry, including the sequestration of carbon dioxide and mobilization of particulate iron. Trichodesmium fixes nitrogen in the daylight, despite the incompatibility of the nitrogenase enzyme with oxygen produced during photosynthesis. While the mechanisms protecting nitrogenase remain unclear, all proposed strategies require considerable resource investment. Here we describe a crucial benefit of daytime nitrogen fixation in Trichodesmium spp. that may counteract these costs. We analysed diel proteomes of cultured and field populations of Trichodesmium in comparison with the marine diazotroph Crocosphaera watsonii WH8501, which fixes nitrogen at night. Trichodesmium’s proteome is extraordinarily dynamic and demonstrates simultaneous photosynthesis and nitrogen fixation, resulting in balanced particulate organic carbon and particulate organic nitrogen production. Unlike Crocosphaera, which produces large quantities of glycogen as an energy store for nitrogenase, proteomic evidence is consistent with the idea that Trichodesmium reduces the need to produce glycogen by supplying energy directly to nitrogenase via soluble ferredoxin charged by the photosynthesis protein PsaC. This minimizes ballast associated with glycogen, reducing cell density and decreasing sinking velocity, thus supporting Trichodesmium’s niche as a buoyant, high-light-adapted colony forming cyanobacterium. To occupy its niche of simultaneous nitrogen fixation and photosynthesis, Trichodesmium appears to be a conspicuous consumer of iron, and has therefore developed unique iron-acquisition strategies, including the use of iron-rich dust. Particle capture by buoyant Trichodesmium colonies may therefore increase the residence time and degradation of mineral iron in the euphotic zone. These findings describe how cellular biochemistry defines and reinforces the ecological and biogeochemical function of these keystone marine diazotrophs, particularly their role as a microbial link in the nitrogen, carbon, and iron cycles. 

How to cite: Held, N., Waterbury, J., Webb, E., Kellogg, R., McIlvin, M., Jakuba, M., Valois, F., Moran, D., Sutherland, K., and Saito, M.: Why does Trichodesmium fix nitrogen during the day? Special biochemistry linking biogeochemical cycles., EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-9136, https://doi.org/10.5194/egusphere-egu22-9136, 2022.

Biological nitrogen fixation (BNF) is a critical process for the N budget and productivity of marine ecosystems. Nitrogen-fixing organisms typically turn off BNF when less metabolically costly N sources, like ammonium (NH₄⁺), are available. Yet, several studies have reported BNF in benthic marine sediments despite high porewater NH₄⁺ concentrations (10-1,500 µM). These activities were generally linked to anaerobic sulfate-reducing bacteria (SRB) and fermenting firmicutes.

To better understand the regulation and importance of benthic marine BNF, we evaluate the sensitivity of BNF to NH₄⁺ in benthic diazotrophs using incubations of increasing complexity. We conduct our experiment with cultures of model anaerobic diazotrophs (sulfate-reducer Desulfovibrio vulgaris var. Hildenborough, fermenter Clostridium pasteurianum strain W5), sulfate-reducing sediment enrichment cultures, and slurry incubations of sediments from three Northeastern salt marshes (USA).

All our samples demonstrate high sensitivity to external NH₄⁺. BNF is inhibited by NH₄⁺ beyond an apparent threshold [NH₄⁺] of 2 µM in liquid cultures and 9 µM in sediment slurries. Consistent with other studies, we find SRB-like nitrogenase (nifH) gene and transcripts are prevalent in sediments. We compare our inhibition threshold value with a survey of porewater NH₄⁺ data from diverse sediments, suggesting the confinement of benthic BNF to surficial sediments.

Variations in the timing to onset BNF inhibition, NH₄⁺ uptake rate, and sediment composition and biophysics could affect measurements of the apparent sensitivity of benthic BNF to NH₄⁺. We propose a simple model based on NH₄⁺ transporter affinity as a fundamental mechanistic constraint on NH₄⁺ control of BNF to improve biogeochemical models of N cycling.

How to cite: Darnajoux, R., Reji, L., Zhang, X., Luxem, K., and Zhang, X.: Ammonium sensitivity of biological nitrogen fixation by anaerobic diazotrophs in cultures and benthic marine sediments., EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-10967, https://doi.org/10.5194/egusphere-egu22-10967, 2022.

The factors that control the distribution of marine diazotrophs and their ability to fix N₂ are not fully elucidated. We discuss insights that can be gained from the emerging picture of a wide geographical distribution of marine diazotrophs and provide a critical assessment of environmental (bottom-up) versus trophic (top-down) controls. Expanding a simplified theoretical framework, we find that selective mortality on non-fixing phytoplankton is identified as a critical process that can broaden the ability of diazotrophs to compete for resources in top-down controlled systems and explain an expanded ecological niche for diazotrophs. Our simplified analysis predicts a larger importance of top-down control on competition patterns as resource levels increase. As selective mortality can control the faster growing phytoplankton, coexistence of the slower growing diazotrophs can be established. However, these predictions require corroboration by experimental and field data, together with the identification of specific traits of organisms and associated trade-offs related to selective top-down control. Elucidation of these factors could greatly improve our predictive capability for patterns and rates of marine N₂fixation.

How to cite: Landolfi, A., Prowe, A. E. F., Pahlow, M., Somes, C. J., Chien, C.-T., Schartau, M., Koeve, W., and Oschlies, A.: Top-down controls on the ecological niche of marine N2 fixers, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-7176, https://doi.org/10.5194/egusphere-egu22-7176, 2022.

The stoichiometric coupling of carbon to limiting nutrients in marine phytoplankton determines that of the main biogeochemical cycles through the process of biomass production by the phytoplankton. While clear links between phytoplankton stoichiometry and environmental drivers have been identified, the nature and direction of these links, as well as the underlying physiological and ecological mechanisms, remain uncertain. Here we compare the predictions of a well-constrained mechanistic model of plankton ecophysiology to multiple observational data sets to investigate the specific case of the C:N phytoplankton stoichiometry in the North-Atlantic. We show that N availability and temperature emerge as the main drivers of phytoplankton stoichiometry. The biological mechanisms involved however vary depending on the spatiotemporal scale and region considered, leading to opposite predictions regarding the evolution of phytoplankton primary productivity in response to environmental changes. At low to intermediate latitudes phytoplankton stoichiometry is predominantly driven by ecoevolutionary shifts in the functional composition of the phytoplankton communities while phytoplankton stoichiometric plasticity in response to dropping temperatures and increased grazing pressure dominates at higher latitudes. Those results shine a new light on what currently influences the circulation of elements through marine ecosystems but also have great implications regarding the evolution of oceans’ primary productivity and of the main biogeochemical cycles under a regime of climate change.

How to cite: Sauterey, B. and Ward, B.: Environmental control of marine phytoplankton C:N stoichiometry in the North Atlantic Ocean today and under climate change, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-2930, https://doi.org/10.5194/egusphere-egu22-2930, 2022.

Nitrous oxide (N₂O) is a greenhouse gas contributing to global warming. Estuaries are a potential source for N₂O.  We aimed to identify seasonal and spatial variations of N₂O production and emission along the Elbe estuary in Germany.

Between 2015 and 2021, we performed nine research cruises along the Elbe estuary. Most of the cruises took place in growing seasons (April – September), while one cruise was conducted in winter (early March). We continuously measured the dissolved N₂O dry mole fraction 2 m below the surface using a laser-based analyzer coupled with an equilibrator. Based on these profiles, we calculated N₂O concentration, saturation and emissions.

During all cruises, the Elbe estuary was supersaturated in N₂O. Highest N₂O concentration occurred in the Hamburg port region, a hotspot of N₂O production by nitrification in the water column and denitrification in the sediments. The maximum concentration in this region was 158 nmol L^‑1 in March 2021. Nitrification in the maximum turbidity zone (MTZ) produced a second local N₂O maximum.  Average N₂O emissions were 0.19 Gg a^‑1(0.52 Mg d^-1­) during the growing season. The N₂O emission was highest in winter with 0.64 Gg a^-1 (1.76 Mg d^-1).

During growing seasons emissions were strongly correlated with pH (R² = 0.73) and suspended particulate matter concentration (R² = 0.55). A trend toward higher N₂O saturations and emissions during cruises in summer is evident. We presume that N₂O saturation and emission were likely driven by temperature-dependent turnover processes in high turbidity areas of the Elbe estuary, such as nitrification and denitrification.  

However, the maximum N₂O concentrations in winter (March 2021) cannot be explained that way, because water temperature was low. N₂O production may be driven by the dissolved inorganic nutrient (DIN) load, which is more than doubled in comparison to all other cruises. Two other possible explanations come to mind: First, N₂O production in this case may be less sensitive to water temperature, possibly due to sedimentary sources. Second, a sink for N₂O in the water column may exist, which is more active during higher temperatures. These two scenarios may both apply and might interact over the course of the year.  

Overall, seasonality affects N₂O production in the Hamburg port region more than in the maximum turbidity zone. In late spring/summer, N₂O production is driven mainly by enhanced microbial productivity. High N₂O concentrations in colder seasons may result from high DIN concentration, but further research on the controls on N₂O production, and possibly consumption, is clearly needed.

How to cite: Schulz, G., Sanders, T., and Dähnke, K.: Spatial and seasonal variation of dissolved nitrous oxide along the Elbe estuary, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-1846, https://doi.org/10.5194/egusphere-egu22-1846, 2022.

Denitrification and anammox are the main nitrogen (N) removal pathways in seawaters. Both processes are important in regions, such as the Baltic Sea, which receive high nutrient loads, that enhance primary production and eutrophication. The Baltic Sea is also characterized by a strong vertical salinity gradient and the presence of a permanent halocline hampering mixing in the water column and ventilation of the deep water layers. Rare events of deep water renewal, together with high oxygen consumption, lead to suboxic and anoxic conditions in the Baltic Sea, which are favorable for denitrification and anammox – processes for which the end product is a non-reactive N₂. In seawater, the concentration of dissolved gases is controlled by biological and physical processes. The latter can be traced by measuring inert gases such as argon (Ar). Hence, the N₂/Ar ratio can be used to separate physical and biological effects influencing N₂ fields. This approach may suit especially to the stratified water bodies, where deep waters are separated from the surface water layer influenced by the gas exchange with the atmosphere.

The study aimed at investigating the potential use of the supersaturation ratio – ΔN₂ as a tracer of denitrification and anammox processes in the water column of the Baltic Proper. The ΔN₂ ratio was derived as an anomaly from the N₂/Ar ratio in seawater being at equilibrium with the atmosphere. The used technique was Membrane Inlet Mass Spectrometry, which allows performing high-precision measurements of dissolved N₂ and Ar in water (masses 28 and 40 were detected, respectively). Seawater samples were collected between 2017 and 2021 from nineteen stations, including Gdańsk, Gotland, and Bornholm Deeps.

The ΔN₂ indicated N₂ accumulation in the oxygen minimum zones below the halocline with the highest values found​​ in the bottom layers. This can be explained by both denitrification and possibly anammox in the water column and with N₂ release from sediments. ΔN₂ values ranged from 1.0 to 32.6 µmol L^-1. In autumn 2021 a significant difference in ΔN₂ (p = 0.0008) between the studied sites was observed. For example on station located in Gotland Deep ΔN₂ values were in the range from 17.6 to 32.6 µmol L^-1, while on station located in the Central Baltic Proper the maximum was 6.1 µmol L^-1. The seasonal ΔN₂ changes (autumn, spring, and winter) were investigated for two stations located in the Gdańsk Deep and indicated statistically significant variability (p=0.0077) with the highest ΔN₂ observed in winter. Additionally, ΔN₂ was negatively correlated with nitrate (R²=0.5469) and oxygen (R²=0.6382), positively with phosphate (R²=0.4382) and ammonium (R²=0.2898), while no clear dependency was observed for nitrite (R²=0.0388).

The presented study was the first attempt performed on such a large scale in the Baltic Proper. It demonstrates a high potential in the use of supersaturation ratio for identification of the active sites for denitrification and anammox processes.

The results were obtained within the framework of the statutory activities of the Institute of Oceanology Polish Academy of Sciences and the research project: 2019/34/E/ST10/00217 funded by the Polish National Science Centre.

How to cite: Borecka, M., Winogradow, A., Koziorowska-Makuch, K., Makuch, P., Diak, M., Kuliński, K., Pempkowiak, J., and Szymczycha, B.: Investigation of the potential use of the supersaturation ratio of N2 (ΔN2) for the estimation of the seasonal and spatial variability of denitrification and anammox in the water column of the Baltic Proper, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-11261, https://doi.org/10.5194/egusphere-egu22-11261, 2022.

Climate change and anthropogenic impact drastically affect the biogeochemical regime of the Black Sea, the contains the largest in the world volume of sulfidic water. The volume of the oxic layer of the sea depends on vertical mixing, that transports dissolved oxygen (DO) from the upper euphotic layer in the deeper layers and dissolved oxygen consumption for oxidation organic matter (OM). Changes in the Sea hydrodynamic properties due to warmer winters restricts renovation of the Black Sea Cold Intermediate Layer and therefore the DO flux to the deeper layers. The main goal of this study was to model upper 350 m with emphasis on redox layer.

In the study we use the benthic-pelagic biogeochemical BROM model combined with 2DBP model for vertical and horizontal transport via FABM. BROM combines a relatively simple ecosystem model with a detailed biogeochemical model considering interconnected transformations of chemical species (N, P, Si, C, O, S, Mn, Fe). OM dynamics include parameterizations of production (via photosynthesis and chemosynthesis) and decay via oxic mineralization, denitrification, metal reduction, sulfate reduction and methanogenesis.

Hydrophysical forcing (temperature, salinity and northward and eastward sea water velocity) was hourly data for year 2010 for a point with coordinates 43.5 °N. 37.75 °E from the E.U. Copernicus Marine Service Information “Copernicus”. The data is the result of a reanalysis calculation based on the NEMO hydrodynamic numerical model.

The 2DBP/BROM model was applied for the 350 m water column with bottom boundary positioned in sulfidic layer. A steady-state solution was reached after 50 calculation years. The results of calculations were compared with the data of field observations of the expedition on the RV "Knorr" in March 2003. The obtained vertical distributions of hydrochemical characteristics are consistent with the existing understanding of the hydrochemical structure of the Black Sea. The dissolved oxygen has the similar structure in the model and observations occupying the upper 70 m layer, its onset was positioned shallower than the appearance of hydrogen sulfide at appr. 80. In the limits of the redox layer there were reproduced maxima of nitrite, Mn(IV), Mn(III), Fe(III), elemental sulfur and phosphate minimum. Below the redox layer the model reproduced maxima of Mn(II) and Fe(II).

Calculated seasonal variability in the upper layer shows seasonality in development of phytoplankton and corresponding changes dissolved oxygen and nutrients. Organic matter distributions changes in accordance with seasonality of its production and destruction. At the same time, the redox layer remains practically unchanged during the year.

At the present the BROM model is applied for the Black Sea and satisfactory validated against the data of observations. The calculated seasonal dynamics of the biogeochemical properties of the Black Sea will be used as an initial condition for studying of effects of changing in mixing (considering modeling interannual changes with difference in hydrodynamical scenarios) and allochthonous OM delivery on the biogeochemical structure of the Sea.

This study was funded by the Ministry of Science and Higher Education of the Russian Federation, theme 13.2251.21.0008.

How to cite: Novikov, M., Berezina, A., Pakhomova, S., and Yakushev, E.: An application of biogeochemical model BROM with 1-D transport model for studying of the vertical biogeochemical structure of the Black Sea, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-68, https://doi.org/10.5194/egusphere-egu22-68, 2022.

Reconstructing the oxygen content at the base of the ocean provides insight into ocean circulation, carbon storage in the deep ocean and hence, the global carbon cycle. The microbial breakdown of organic carbon within marine sediment through aerobic respiration consumes oxygen in the pore fluid and releases dissolved inorganic carbon. The offset in the carbon isotopic composition of epifaunal and infaunal foraminifera is related to the respiration of the organic carbon and can be used to reconstruct the oxygen content of the bottom water. Previous work has provided a data-derived calibration which is valid for oxygen reconstructions between 55–235 micromolar. In this study, we apply a biogeochemical reactive transport model (RTM) to extend and update the calibration, which allows for the reconstruction of oxygen concentrations up to ~400 micromolar. Using the RTM and new data from the Iberian Margin, we also demonstrate that the calibration between the carbon isotope gradient and bottom water oxygen concentrations must account for the coupled changes in all aspects of the carbon system due to the respiration of organic carbon. We apply the improved calibration to reconstruct the changes in oxygen content in the North Atlantic over the past 1.4 Myr.

How to cite: Bradbury, H., Thomas, N., and Hodell, D.: Revisiting the relationship between the pore water carbon isotope gradient and bottom water oxygen concentrations, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-7896, https://doi.org/10.5194/egusphere-egu22-7896, 2022.

The ocean absorbs around a quarter of the annual man-made CO₂ emissions, with the Southern Ocean being responsible for the lion-share of this uptake. Despite the disproportional role of the Southern Ocean in taking up anthropogenic CO2, it still remains one of the most sparsely observed ocean regions. While autonomous measurement devices have started to fill this void, high quality shipboard measurements remain limited. One fleet with the potential to fill this gap has thus far received little attention: sailboats. There is growing willingness among skippers to help science, providing a opportunity to collect valuable measurements of the sea surface partial pressure of CO2 (pCO2) during round-the-world racing events. Using the latest membrane sensor technology, we have thus far – together with professional racing teams - collected high frequency measurements of the sea surface pCO2 from nearly all ocean basins and most notably in remote southern hemisphere ocean regions where no shipboard pCO2 data were collected in the past 70 years (based on the SOCAT database). Our results highlight the potential for equipping sail yachts as a low-cost solution to fill data gaps and provide a new constraint for high resolution model studies.

How to cite: Landschützer, P., Tanhua, T., and Behncke, J.: Round-the-globe racing events to fill the pCO2 data void in the Southern Ocean, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-3562, https://doi.org/10.5194/egusphere-egu22-3562, 2022.

The ocean moderates the world’s climate through absorption of heat and carbon, but how much carbon the ocean will continue to absorb remains unknown. The North Atlantic Ocean west (Baffin Bay/Labrador Sea) and east (Fram Strait/Greenland Sea) of Greenland features the most intense absorption of anthropogenic carbon globally; the biological carbon pump (BCP) contributes substantially. As Arctic sea-ice melts, the BCP changes, impacting global climate and other critical ocean attributes (e.g. biodiversity). Full understanding requires year-round observations across a range of ice conditions. Here we present such observations: autonomously collected Eulerian continuous 24-month time-series in Fram Strait. We show that, compared to ice-unaffected conditions, sea-ice derived meltwater stratification slows the BCP by 4 months, a shift from an export to a retention system, with measurable impacts on benthic communities. This has implications for ecosystem dynamics in the future warmer Arctic where the seasonal ice zone is expected to expand. (Published in Nature Communications, December 2021, https://www.nature.com/articles/s41467-021-26943-z )

How to cite: von Appen, W.-J., Waite, A., and Boetius, A. and the FRAM team: Sea-ice derived meltwater stratification slows thebiological carbon pump: results from continuousobservations, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-2322, https://doi.org/10.5194/egusphere-egu22-2322, 2022.

Ocean Acidification (OA) is considered a major threat and is projected to impact all areas of global ocean, therefore understanding its ecological impacts remains a priority for science and management. The Mediterranean Sea is considered a highly vulnerable region, so we analyzed material coming from Planier sediment trap in order to characterize the seasonal variability of weight and calcification of planktic foraminifera. This sediment trap is located in the Gulf of Lions (GoL), in the northwestern part of the Mediterranean Sea, one of the few non-oligotrophic regions in the Mediterranean (high productivity period from January to May). We performed planktic foraminifera picking focusing on 3 different species: Globigerina bulloides, Neogloboquadrina incompta and Globorotalia truncatulinoides. A mean of 13 to 27 specimens per sample were picked. These foraminifera samples were then cleaned with the ultrasonication in methanol technique and then weighted using a Sartorius ME5 balance (precision= 0.001mg) in the micropaleontology laboratory of the University of Salamanca. A total of 126 samples and 2077 individuals were weighted. SBW (Sieve Based Weight) results showed that traditional used sieved size fractions do not provide enough control on the effect of morphometric parameters on the weight/calcification data, highlighting the need of a size-normalization. Area and diameter measurements were carried using a Nikon SMZ18 and a DS-Fi3 through the NIS Elements. MBW (Measured Based Weights) results showed that both of these parameters (area and diameter) have no influence on MBW values, indicating these values are good index for calcification intensity. Seasonal MBW variations differ according to the species: G.bulloides showed a maximum MBW values during the high productivity period, N.incompta reached its maximum values slightly after the high productivity period while G.truncatulinoides displayed a maximum calcification value during the low productivity period. Finally, we compared these results with “Optimum Growth Conditions” (Chlorophyll-a and species relative abundance) data.

How to cite: Béjard, T. M., Rigual-Hernández, A. S., Sierro, F. J., and Pérez-Tarruella, J.: Planktic foraminifera seasonal calcification variations in the northwestern Mediterranean Sea, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-9409, https://doi.org/10.5194/egusphere-egu22-9409, 2022.

A fraction of the carbon fixed in the surface ocean by phytoplankton is isolated away from the atmosphere in the ocean interior with the respiration of sinking detritus (Particles of Organic Carbon: POC) - a process known as the "Biological Carbon Pump'' (BCP). The BCP sequesters ~1700 Pg of dissolved inorganic carbon (DIC) in the ocean beyond the concentration expected solely with physio-chemical drivers, effectively lowering the base-line atmospheric CO2 concentration by ~150-250 ppm. The components that make up the BCP (export production, sinking and remineralisation of POC, ocean ventilation timescales) are all expected to change in response to a changing climate but there is currently low confidence in how these changes will influence the magnitude and direction of the ocean carbon feedback.

Here we quantify the predicted historical and future changes in the Biological Carbon Pump in the latest CMIP6 projections as fully as possible. We find that all models consistently predict that the BCP will accumulate carbon in the ocean interior by 2100, i.e., acting as a sink for atmospheric CO2, albeit contributing only a small fraction (~10%) of the net carbon sink. The accumulation of carbon along with a concurrent decrease in globally integrated export production at 100m is associated with warming-driven stratification. In contrast there is significant disagreement in both the magnitude and direction of global mean trends and spatial patterns of transfer efficiency of POC at 1000m. This uncertainty arises because of the range of processes resolved across the biogeochemical models that influence the sinking and remineralisation rate of POC such as: temperature and oxygen-dependent remineralisation, ballasting, and dependence of sinking velocities on cell size. We demonstrate that these changes in transfer efficiency could likely determine the larger long-term impact of the BCP on atmospheric CO2 beyond 2100. Our results have wider implications for the biogeochemical cycling of nutrients and oxygen as well as implications for future impacts on twilight zone ecology.

How to cite: Wilson, J., Andrews, O., and Katavouta, A.: Future trends and uncertainties in the Biological Carbon Pump predicted by CMIP6 models, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-13378, https://doi.org/10.5194/egusphere-egu22-13378, 2022.

The low-latitude ocean regions spanning 30°S-30°N are thought to account for more than 50% of the global export production. However, previous analyses of paleo-proxy records and modeling studies strongly suggest contradictory evidence as to whether low latitude nutrient cycling and export production is locally or non-locally controlled. Here we address this question through the new application of observational (PACIFICA) and modeling (NEMO-PISCES) tools and show that low latitude recycling of nutrients within the thermocline overturning structures is largely responsible for sustaining low latitude export production (60%) for the mean state, with only second-order controls from the injection of new (preformed) nutrients from the Southern (16%) and northern (9%) oceans.  The implications for understanding controls on long-term changes under sustained anthropogenic climate perturbations is investigated using CMIP6 Earth system models under idealized 4xCO2 forcing, where significant reductions in low-latitude export production and net primary production over 30°S-30°N are investigated.

How to cite: Rodgers, K., Aumont, O., Toyama, K., Resplandy, L., Ishii, M., Lindsay, K., Yamaguchi, R., Sasano, D., and Nakano, T.: A weak role for Southern Ocean nutrient drawdown in low latitude marine export production, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-10868, https://doi.org/10.5194/egusphere-egu22-10868, 2022.

The ocean plays an instrumental role in regulating the Earth’s climate through the buffering of the anthropogenic-induced excess carbon. Our capacity to predict long-term future oceanic carbon uptake depends on highly sophisticated numerical Earth system models, whose simulations of future climate have a wide inter-model dispersion. Inter-model spread in projections arises from three distinct sources: 1) internal variability of the climate system, 2) model uncertainty, and 3) scenario uncertainty. The spread related to (1) and (2) is even greater when predicting changes at regional scales. In order to elucidate the main origins of present and future internal variability and model uncertainty in oceanic carbon uptake, it is important to identify the uncertainty and sensitivity of the major underlying mechanisms in different ocean regions and across models. A limitation of this approach is the high costof computational and manpower required to systematically assess all mechanisms and identify processes that are important in a consistent way, especially across a large ensemble of model sets. Machine learning methods can be applied to simultaneously estimate the sensitivity of variable sets and explore them automatically across the ensemble of models. Here, we use the Kernel non-linear regression approach to reconstruct the inter-annual carbon uptake variability using monthly outputs of surface temperature, salinity, nutrient, dissolved inorganic carbon,alkalinity, atmospheric CO2 concentration, surface wind speed, and sea-ice cover. The exercise was performed on preindustrial, historical, and future scenario simulation outputs. The algorithm was optimized with a subset of ‘training’ data and evaluated with ‘test’ data. We applied bootstrapping method to delineate the main drivers for the projected inter-annual sea-air carbon fluxes variability in different ocean domains.

How to cite: Couespel, D., Tjiputra, J., and Johannsen, K.: Identifying the underlying mechanisms of present and future inter-annual variability of oceanic carbon uptake using a machine learning approach with CMIP6 simulations., EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-7709, https://doi.org/10.5194/egusphere-egu22-7709, 2022.