Laurent Oziel

Global oceans store large quantities of human-made persistent hydrophobic organic contaminants (HOCs), driven by the organic carbon (OC) cycle. Despite diverse organic matter composition, bulk OC is typically treated as a uniform sorbent when assessing HOC fate. Significant uncertainty remains regarding the factors controlling variability in the sorption of HOCs by organic matter. Using surface sediments across the Eurasian Arctic shelves, spanning a land-to-ocean continuum of OC, we show that polychlorinated biphenyl concentrations per unit marine‑derived OC are up to an order of magnitude higher than those per unit terrestrial OC. Coupling these empirical relationships with a state‑of‑the‑art ocean biogeochemistry model shows that climate‑driven increases in marine primary production in the Eurasian Arctic shelves lead to 1.3 times more HOC removal from the water column. These findings reveal that both OC quantity and quality control the sediment HOC sink and that climate‑induced shifts in OC cycles may restructure HOC fate in the Arctic Ocean.

Carbon-rich habitats in polar regions are globally important, as uptake of CO₂ by marine systems is shifting toward the poles. Changing conditions are exposing these regions to anthropogenic exploitation. Timely policy intervention can protect these habitats to prevent further emissions and further integration of polar blue carbon in management frameworks.

The Arctic Ocean is experiencing sea-ice losses. The increase in light availability has increased net primary production. However, recent studies postulate that nutrients (and not light) are now the dominant control. We present observations from the Fram Strait (1998–2023) showing a transition around 2009, consisting of a sharp decline in fixed-nitrogen concentrations in Polar Surface Waters and an accompanying increase in silicon-to-nitrogen ratios. We suggest this represents a regime shift where nitrate has emerged as the main limiting factor for primary production in the Arctic Ocean. This reduction in nitrate may have resulted from increased benthic denitrification on the shelves. By combining modelled benthic denitrification rates and Lagrangian trajectories, we identify a marked increase in nitrogen loss after 2009, with increasing denitrification in the Chukchi and East Siberian shelves. We attribute this response to reduced sea ice and circulation changes resulting in a regime shift toward stronger nitrogen limitation.

La mer de Barents possède un écosystème particulièrement riche. Elle est affectée par le changement climatique actuel, comme le reste de l'Arctique. L'effet le plus visible et le plus connu est la réduction spectaculaire de la banquise. On examine dans cette thèse les répercussions de ces changements sur l'hydrologie et le phytoplancton en mer de Barents. Pour cela, on s'appuie sur la création d'une base de données historiques comprenant les paramètres hydrologiques et bio-géochimiques. Un modèle 3D bio-géochimique spécifique à l'écosystème arctique est aussi utilisé quand les observations sont manquantes. Enfin, les données satellites fournissent des séries temporelles de concentration de glace, de Chlorophyle-a...La mer de Barents est caractérisée par un front polaire séparant les eaux atlantiques issues des mers Nordiques des eaux arctiques dont la position est connue à l'ouest de 35°E. Nous avons pu montrer que ce front se sépare en deux branches dans l'Est du bassin: le front du Nord et le front du Sud. Ces fronts enferment les eaux de la mer de Barents qui se forment en hiver. Un doublement du volume des eaux atlantiques (une " Atlantification " de la mer de Barents), a aussi été mise en évidence. Elle accompagne un déplacement des fronts Nord et Sud vers le Nord-est. Le volume des eaux de la mer de Barents reste inchangé.Ces changements, affectant l'hydrologie et la glace de Mer, ont un impact significatif sur le phytoplancton. Les deux efflorescences qui le caractérisent ont lieu plus au Nord et à l'Est. La biomasse totale annuelle a augmenté de 40% lors des deux dernières décennies. Ce travail montre que les conditions de glace de mer et la structure frontale sont les paramètres clefs dirigeant la variabilitéinter-annuelle du phytoplancton.

The Arctic Ocean (AO) is changing very rapidly. Retreating sea-ice and the subsequent increase in light availability has significantly increased AO Net Primary Production (NPP). However, recent studies postulate that nutrients (and not light) now control NPP dynamics. We present observations from the Fram Strait (1998-2023) where this transition is revealed around 2009 as a sharp decline in fixed-nitrogen concentrations in the Polar Surface Water and an accompanying increase in Si:N ratios. We suggest that this represents a regime shift where fixed Nitrogen (N) has emerged as the main limiting factor for NPP in the contemporary AO. This reduction of N levels in the last decade may have resulted from increased benthic denitrification (BD) on the shelves. We investigate this by combining modelled BD rates and Lagrangian trajectories, which show a sharp increase in BD around 2009, with increasing contributions from the Chukchi and East Siberian shelves. We attribute this biogeochemical response to a drastic reduction in sea ice and circulation shift around this time. We suggest that Arctic climate change has led to a regime shift where low N levels resulting from increased loss of fixed N on the shelves now exert a tighter control on Arctic NPP.

The fate of hydrophobic organic pollutants in the marine environment is largely controlled by organic carbon (OC) cycling processes. The Arctic is warming three times as fast as the global average, resulting in a profound alteration in OC fluxes to the Arctic Ocean. Consequently, remote Arctic shelf sediments serve as an ideal receptor for assessments of the impact of OC quantity and quality on pollutant fate.Here, we compiled a database of congener-specific polychlorinated biphenyl (PCB) concentrations (510 entries, 84 sites) in surface sediment of four Eurasian Arctic shelves via integration of new measurements and literature data. Total organic carbon content and isotopic data were retrevied from CASCADE (The Circum-Arctic Sediment CArbon DatabasE) to examine the PCB storage in sediment as a function of OC source (marine versus terrestrial). In order to reduce the impact of variability in water-phase concentrations caused by region, latitude and depositional year, we controlled for these factors. The adjusted concentrations (in ng-PCB/g-OC) in sediment with high fractions of marine OC are 0.82-1.22 log units higher in Barents and Kara Seas and 0.092-1.49 log units higher in Laptev and East Siberian Seas, compared to those with high fractions of terrestrial OC. Albeit uncertainties in current estimations due to wide geographical coverage and correction assumptions, these values are comparable to previously reported differences of partition coefficients between marine and terrestrial OC in other regions (e.g., 0.2-1 log units higher in marine OC sites in Baltic Sea, compared to terrestral OC sites).Based on these results, PCB accumulation in Arctic shelf sediment was predicted for future climate change scenarios using an observational dataset of terrestrial inputs and marine OC fluxes derived from the global state-of-the-art ocean- and sea ice biogeochemistry model FESOM2.1-REcoM3. The accumulated amount of PCBs in marine OC in Arctic shelf sediments from 2000-2100 is estimated to be about 26 tonnes, which is more than 8 times higher than the accumulated amount in terrestral OC from both coastal erosion and riverine inputs. These results demonstrate that shifts in OC fluxes as a consequence of climate change can impact storage capacity of hydrophobic organic pollutants in aquatic systems.

We present an assessment of the coupled ocean–sea ice–biogeochemistry model FESOM2.1–REcoM3, in which we integrated state equations for dissolved acidic polysaccharides (PCHO) and transparent exopolymer particles (TEP), as proposed by Engel et al. (2004), to explicitly describe these two organic carbon pools in the Arctic Ocean. PCHO is simulated as one fraction of the phytoplankton exudates, which can then aggregate to form larger particles, TEP. Since observational datasets on TEP are rare in time and space, we systematically assess the novel model implementation by stepwise discussing the essential components of the organic carbon cycle. Firstly, the simulated phytoplankton biomass yields good results when compared to in situ and remote-sensing products of total Chlorophyll a and particulate organic carbon. Secondly, we compare PCHO to observations in the Fram Strait, as an exemplary data-rich region, and to datasets in other regions of the Arctic Ocean. The model realistically reproduces a high phytoplankton exudation rate of PCHO under nutrient-depleted conditions. Thirdly, we assess simulated TEP concentrations by comparing them to in situ measurements from several campaigns to the Arctic Ocean. The simulation provides a first estimate of mean TEP concentrations of 200–400 µg C L⁻¹ on the continental shelves and 10–50 µg C L⁻¹ in the central basins (0–30 m depth range). Lastly, we put the model performance into a global context for TEP concentrations in the upper ocean layer. As such, the implementation of PCHO exudation, aggregation to TEP, and their remineralization processes into FESOM2.1–REcoM3 offers a reasonably good agreement with observations, on which further modeling work can build upon.

We present an assessment of the coupled ocean–sea ice–biogeochemistry model FESOM2.1–REcoM3, in which we integrated state equations for dissolved acidic polysaccharides (PCHO) and transparent exopolymer particles (TEP), as proposed by Engel et al. (2004), to explicitly describe these two organic carbon pools in the Arctic Ocean. PCHO is simulated as one fraction of the phytoplankton exudates, which can then aggregate to form larger particles, TEP. Since observational datasets on TEP are rare in time and space, we systematically assess the novel model implementation by stepwise discussing the essential components of the organic carbon cycle. Firstly, the simulated phytoplankton biomass yields good results when compared to in situ and remote-sensing products of total Chlorophyll a and particulate organic carbon. Secondly, we compare PCHO to observations in the Fram Strait, as an exemplary data-rich region, and to datasets in other regions of the Arctic Ocean. The model realistically reproduces a high phytoplankton exudation rate of PCHO under nutrient-depleted conditions. Thirdly, we assess simulated TEP concentrations by comparing them to in situ measurements from several campaigns to the Arctic Ocean. The simulation provides a first estimate of mean TEP concentrations of 200–400 μg C L⁻¹ on the continental shelves and 10–50 μg C L⁻¹ in the central basins (0–30 m depth range). Lastly, we put the model performance into a global context for TEP concentrations in the upper ocean layer. As such, the implementation of PCHO exudation, aggregation to TEP, and their remineralization processes into FESOM2.1–REcoM3 offers a reasonably good agreement with observations, on which further modeling work can build upon.

We present an assessment of the coupled ocean–sea ice–biogeochemistry model FESOM2.1–REcoM3, in which we integrated state equations for dissolved acidic polysaccharides (PCHO) and transparent exopolymer particles (TEP), as proposed by Engel et al. (2004), to explicitly describe these two organic carbon pools in the Arctic Ocean. PCHO is simulated as one fraction of the phytoplankton exudates, which can then aggregate to form larger particles, TEP. Since observational datasets on TEP are rare in time and space, we systematically assess the novel model implementation by stepwise discussing the essential components of the organic carbon cycle. Firstly, the simulated phytoplankton biomass yields good results when compared to in situ and remote-sensing products of total Chlorophyll a and particulate organic carbon. Secondly, we compare PCHO to observations in the Fram Strait, as an exemplary data-rich region, and to datasets in other regions of the Arctic Ocean. The model realistically reproduces a high phytoplankton exudation rate of PCHO under nutrient-depleted conditions. Thirdly, we assess simulated TEP concentrations by comparing them to in situ measurements from several campaigns to the Arctic Ocean. The simulation provides a first estimate of mean TEP concentrations of 200–400 μg C L⁻¹ on the continental shelves and 10–50 μg C L⁻¹ in the central basins (0–30 m depth range). Lastly, we put the model performance into a global context for TEP concentrations in the upper ocean layer. As such, the implementation of PCHO exudation, aggregation to TEP, and their remineralization processes into FESOM2.1–REcoM3 offers a reasonably good agreement with observations, on which further modeling work can build upon.

Changing Arctic climate conditions have accelerated sea ice retreat, altering ocean–atmosphere interactions and marine ecosystems. Reduced sea ice cover likely enhances emissions of primary marine organic aerosol (PMOA) through bubble bursting, with implications for aerosol–cloud interactions. This study examines the emission patterns, seasonality, and historical trends of key PMOA species (dissolved carboxylic acidic containing polysaccharides, PCHO; dissolved combined amino acids, DCAA; polar lipids, PL) within the Arctic from 1990 to 2019. Surface ocean concentrations of marine biomolecules, derived from a biogeochemistry model used in the ECHAM-HAM aerosol–climate model, exhibit pronounced seasonal cycles. PMOA emissions show strong variability, driven by marine productivity and sea-salt emissions, with maxima from May to September. Total PMOA emissions increased by about 12 %, and the burden rose by 4 % between 1990–2004 and 2005–2019. A 30 year summer trend (July–September) indicates a rapid decline in sea ice, accompanied by increasing concentrations of organic groups in inner-Arctic waters. Positive PMOA emission anomalies have become more frequent over the past 15 years, and total PMOA production has increased by 0.8 % yr⁻¹ since 1990. Differences among biomolecular types persist, with PCHO showing the strongest increases in both emissions (1.3 % yr⁻¹) and aerosol concentrations (0.8 % yr⁻¹).