Particle Fluxes in the North-East Atlantic and Southern Ocean

Concerns regarding the climatic implications of the increase in atmospheric CO2 concentrations<br/>throughout the anthropocene have provided the impetus to obtain a mechanistic understanding<br/>of oceanic processes and their role in regulating atmospheric pCO2. One important mechanism<br/>is the functioning of the biological pump which partitions carbon between the atmosphere and<br/>ocean reservoirs over relevant time scales. Current uncertainties revolve around the accuracy of<br/>upper ocean particle flux measurements, and the effect of iron and ballast minerals on the<br/>strength and efficiency of the biological carbon pump. This study documents the design and<br/>deployment of a neutrally buoyant sediment trap (PELAGRA). In the north-east Atlantic<br/>organic carbon fluxes were measured using this new technology and compared to indirect<br/>estimates of export based on 234Th and nutrient budgets. The vertical fluxes of 234Th into the<br/>traps were less than those estimated from the 234Th water column budget, which is interpreted to<br/>be the result of previous export events removing 234Th from the water column and the lateral<br/>advection of gradients of total 234Th/238U disequilibria confounding the Eulerian budgeting<br/>approach adopted. Successful simultaneous deployments in July 2006 at different depths<br/>provided a direct measurement of the attenuation of flux with depth, which at 1.8 is<br/>substantially greater than the canonical value of 0.856. PELAGRA deployments in the<br/>Southern Ocean were conducted as part of the CROZEX project, which examined the role of<br/>iron supply on bloom dynamics and subsequent export. Using a mass balance approach to<br/>account for the seasonal depletion of dissolved silica acid in surface waters and Si fluxes from<br/>the euphotic zone, potential surface export(100m) of organic carbon from +Fe bloom area was<br/>estimated to be in the order of 11-15 g C m-2, which is higher than previous estimates obtained<br/>from artificial fertilisation experiments. The issue of temporal decoupling between production<br/>and export processes was addressed by employing retrospective estimates of production.<br/>Particle export efficiency in the +Fe region to the north of the plateau (25-70%) was higher than<br/>similar estimates in the –Fe region (11-20%). Diatom size was well correlated with a range of<br/>calculated export ratios(100m). The main diatoms involved in the export from the surface were E.<br/>Antarctica in the +Fe region and F. kerguelensis in the –Fe region. E. Antarctica fluxes also<br/>dominated deep-water (3000m) diatom fluxes in the +Fe region, and its importance is attributed<br/>to the regions proximity to the Crozet Islands, where resting spores and dissolved iron are<br/>advected into the bloom area during the winter. Deep-water carbon fluxes measured to the<br/>south of the plateau. Deep-water carbon fluxes measured south of the plateau (0.09 g C m-2 yr-1)<br/>are consistent with previous measurements in a similar environment. In the +Fe region to the<br/>north, deep water fluxes were 0.4 g C m-2 yr-1 indicating that natural iron fertilisation can<br/>increase the strength of the biological carbon pump by a factor of 4. Comparison of fluxes with<br/>satellite-derived productivity also suggests that the efficiency of the biological pump in<br/>transferring organic carbon to the deep-ocean is increased by a factor of 3 in the presence of<br/>iron. The flux and composition of amino acids, in relation to the dominant mineral phases that<br/>comprised the particulate flux in the NE Atlantic and the Southern Ocean was also examined.<br/>The fraction of carbon that could be accounted for by the total hydrolysable amino acids varied<br/>very little (20-30%) with sample composition. Protein amino acids were used to quantify the<br/>degradation state of the settling particulate material. Specific amino acids seem to infer<br/>diatomaceous rather than calcareous as the dominant organic matter source. Multiple linear<br/>regression analysis reveals that mineral fluxes can only explain a very small amount of the<br/>variability in amino acid composition, which does not support previous hypotheses that relate<br/>mineral fluxes and organic carbon fluxes through the differential protective capacity of various<br/>mineral phases.

Name dc:type.qualificationname

Ph.D.

Level dc:type.qualificationlevel

doctoral

Grantor dc:publisher.institution

University of Southampton

Year dc:date.issued

2007

Author dc:creator

• Salter, Ian