Abstract. The Kerguelen Plateau region in the Indian sector of the Southern Ocean supports annually a large-scale phytoplankton bloom which is naturally fertilized with iron. As part of the second KErguelen Ocean and Plateau compared Study expedition (KEOPS2) in austral spring (October–November 2011), we examined upper-ocean Particulate Organic Carbon (POC) export using the 234Th approach. We aimed at characterizing the spatial and the temporal variability of POC export production at high productivity sites over and downstream the Kerguelen plateau. Export production is compared to a High Nutrient Low Chlorophyll area upstream of the plateau in order to assess the impact of iron-induced productivity on the vertical export of carbon. Deficits in 234Th activities relative to its parent nuclide 238U were observed at all stations in surface waters, indicating that scavenging by particles occurred during the early stages of the phytoplankton bloom. 234Th export was lowest at reference station R-2 (412 ± 134 dpm m–2 d–1) and highest inside a~permanent meander of the Polar Front (PF) at stations E (1995 ± 176 dpm m–2 d–1, second visit E-3) where a detailed time series was obtained as part of a~pseudo-lagrangian study. 234Th export over the central plateau was relatively limited at station A3 early (776 ± 171 dpm m–2 d–1, first visit A3-1) and late in the survey (993 ± 223 dpm m–2 d–1, second visit A3-2), but it was higher at high biomass stations TNS-8 (1372 ± 255 dpm m–2 d–1) and E-4W (1068 ± 208 dpm m–2 d–1) in waters which could be considered as derived from plateau. Limited 234Th export of 973 ± 207 dpm m–2 d–1 was also found in the northern branch of the Kerguelen bloom located downstream of the island, north of the PF (station F-L). The 234Th results support that Fe fertilization increased particle export in all iron fertilized waters. The impact was greatest in the recirculation feature (3–4 fold at 200 m depth), but more moderate over the central Kerguelen plateau and in the northern plume of the Kerguelen bloom (∼2-fold at 200 m depth). The C : Th ratio of large (> 53 μm) potentially sinking particles collected via sequential filtration using in situ pumping (ISP) systems were used to convert the 234Th flux into a POC export flux. The C : Th ratios of sinking particles were highly variable (range: 3.1 ± 0.1–10.5 ± 0.2 μmol dpm–1) with no clear site related trend, despite the variety of ecosystem responses in the fertilized regions. C : Th ratios showed a decreasing trend between 100 and 200 m depth suggesting preferential loss of carbon relative to 234Th possibly due to heterotrophic degradation and/or grazing activity. Comparison of the C : Th ratios within sinking particles obtained with the drifting sediment traps showed in most cases very good agreement to those collected via ISP deployments (> 53 μm particles). Carbon export production varied between 3.5 ± 0.9 mmol m–2 d–1 and 11.8 ± 1.3 mmol m–2 d–1 from the upper 100 m and between 1.8 ± 0.9 mmol m–2 d–1 and 8.2 ± 0.9 mmol m–2 d–1 from the upper 200 m. Highest export production was found inside the PF meander with a range of 5.4 ± 0.7 mmol m–2 d–1 to 11.8 ± 1.1 mmol m–2 d–1 at 100 m depth decreasing to 5.3 ± 1.0 mmol m–2 d–1 to 8.2 ± 0.8 mmol m–2 d–1 at 200 m depth over the 19 day survey period. The impact of Fe fertilization is highest inside the PF meander with 2.9- up to 4.5-fold higher carbon flux at 200 m depth in comparison to the HNLC control station. The impact of Fe fertilization was significantly less over the central plateau (stations A3 and E-4W) and in the northern branch of the bloom (station F-L) with 1.6- up to 2.0-fold higher carbon flux compared to the reference station R. Export efficiencies (ratio of export to primary production) were particularly variable with relatively high values in the recirculation feature (6–27%) and low values (1–5%) over the central plateau (station A3) and north of the PF (station F-L) indicating spring biomass accumulation. Comparison with KEOPS1 results indicated that carbon export production is much lower during the onset of the bloom in austral spring in comparison to the peak and declining phase in late summer.

International audienc

Abstract. Dissolved Fe (dFe) concentrations were measured in the upper 1300 m of the water column in the vicinity of Kerguelen Island as part of the second Kerguelen Ocean Plateau compared Study (KEOPS2). Concentrations ranged from 0.06 nmol L−1 in offshore, Southern Ocean waters, to 3.82 nmol L−1 within Hillsborough Bay, on the north-eastern coast of Kerguelen Island. Direct island runoff, glacial melting and resuspended sediments were identified as important inputs of dFe that could potentially fertilize the northern part of the plateau. A significant deep dFe enrichment was observed over the plateau with dFe concentrations increasing up to 1.30 nmol L−1 close to the seafloor, probably due to sediment resuspension and pore water release. Biological uptake was identified as a likely explanation for the decrease in dFe concentrations between two visits (28 days apart) at a station above the plateau. Our results allowed studying other processes and sources, such as atmospheric inputs, lateral advection of enriched seawater, remineralization processes and the influence of the Polar Front (PF) as a vector for Fe transport. Overall, heterogeneous sources of Fe over and off the Kerguelen Plateau, in addition to strong variability in Fe supply by vertical or horizontal transport, may explain the high variability in dFe concentrations observed during this study.

Hydrothermal iron supply contributes to the Southern Ocean carbon cycle via the regulation of regional export production. However, as hydrothermal iron input estimates are coupled to helium, which are uncertain depending on whether helium inputs are based on ridge spreading rates or inverse modelling, questions remain regarding the magnitude of the export production impacts. A particular challenge is the limited observations of dissolved iron (dFe) supply from the abyssal Southern Ocean ridge system to directly assess different hydrothermal iron supply scenarios. We combine ocean biogeochemical modelling with new observations of dFe from the abyssal Southern Ocean to assess the impact of hydrothermal iron supply estimated from either ridge spreading rate or inverse helium modelling on Southern Ocean export production. The hydrothermal contribution to dFe in the upper 250 m reduces 4–5 fold when supply is based on inverse modelling, relative to those based on spreading rate, translating into a 36–73% reduction in the impact of hydrothermal iron on export production. However, only the spreading rate input scheme reproduces observed dFe anomalies >1 nM around the circum-Antarctic ridge. The model correlation with observations drops 3 fold under the inverse modelling input scheme. The best dFe scenario has a residence time for hydrothermal iron that is between 21 and 34 years, highlighting the importance of rapid physical mixing to surface waters. Overall, because of its short residence time, hydrothermal Fe supplied locally by circum-Antarctic ridges is most important to the Southern Ocean carbon cycle and our results highlight decoupling between hydrothermal iron and helium supply.