Transport and budget of carbon, nutrients and oxygen in the North Atlantic

Resumen

The current biogeochemical status of the eastern subpolar North Atlantic (eSPNA) was evaluated under an integrated approach based on quantifications of transports and budgets of carbon, nutrients and oxygen. The first component of the carbon cycle analyzed was the organic component. In the ocean carbon cycle, the dissolved organic fraction (DOC) represents practically the whole amount in the total organic carbon (TOC) pool because the particulate fraction (POC) is minority and assumed in balance. The scarcity of DOC data throughout the oceans, and the North Atlantic in particular, hamper a correct evaluation of the relative importance of DOC in the total carbon cycle [Hansell et al., 2004]. In order to solve this, here we show an assessment coupling well-solved water mass transports along transoceanic sections with high-quality [DOC] measurements. With [DOC] samples measured in the OVIDE 2002 cruise and a recent water mass mixing model developed for the subpolar North Atlantic gyre [García-Ibañez et al., 2015], a characterization of the [DOC] source water type for each water mass existent at the OVIDE section was resolved by an inversion system. With the [DOC] defined for each water mass, the combination of the water mass structure resolved by means of an extended optimum multiparameter (eOMP) analysis [García-Ibañez et al., 2015] with absolute velocity fields for the OVIDE section [Lherminier et al., 2007, 2010], the transport of DOC could be evaluated for cruises when it was not measured. Combining transports at OVIDE section with transports of water mass and DOC available in the literature at the northern limit of the eSPNA (G-I-S sills) [Hansen & Osterhus, 2000; Hansen & Osterhus, 2007; Jeansson et al., 2011; Jochumsen et al., 2012], the carbon organic budget in the eSPNA was elaborated. The combination of own data at OVIDE section with public available estimates at the boundaries of the eSPNA is a common characteristic followed along this thesis for the budget assembly. We separated the contribution of the upper and lower limb of the AMOC to the total transport of DOC, and concluded that when DOC advection in the eastern subpolar North Atlantic is considered, the DOC budget is in balance (8±77 kmol·s-1), suggesting that DOC production in the eSPNA is balanced by its removal. With regard to the [DOC] water mass characterization, our results confirm that the Nordic overflows (DSOW and ISOW) are relatively rich in DOC content, adding important quantities of DOC to North Atlantic deep waters. We also found that a significant characteristic of the DOC cycle in the eSPNA is its export from the upper to the lower limbs of the AMOC. The relatively fast vertical transport of DOC contributes to carbon sequestration, analogous to the vertical transport of DIC. DOC injected to the deep layers is more labile than the majority of the deep DOC observed in the South Atlantic and Pacific Oceans, so it is more susceptible to remineralization. Following a similar strategy, and inspired by the closely similar mean velocity-weighted [DOC] found in both limbs of the AMOC at the eSPNA, we also reconstructed the DOC transports at subtropical latitudes in the North Atlantic (24ºN). Here we show that in the North Atlantic, 62 Tg-C·yr–1 of DOC are consumed in the lower limb of the AMOC between subpolar and subtropical latitudes, representing 72% of the DOC exported by the whole Atlantic Ocean. This outcome implies that much of the net DOC exported with the overturning circulation in the eastern-SPNA, the major source of new DOC in the deep global ocean, is remineralized within decades, thus impacting deep microbial and dissolved organic matter compositional dynamics. DOC downward export due to overturning circulation acts as a carbon sink and represents a considerable contribution to CO2 sequestration. Given the atmospheric CO2 uptake of 0.20 Pg-C·yr–1 in the area [Takahashi et al., 2009], the carbon sequestration mediated by DOC would represent ~30% of the total North Atlantic CO2 sink. Once evaluated the organic component in a separate way, we focus on the main core of the thesis, the elaboration of an integrated biogeochemical budget balanced around carbon and analyzing the longest time-serie available for the OVIDE section (2002-2016). This study-case is based in the combination of eight repetitions of the high-quality hydrographic section OVIDE with public available data sources. Using an optimization method, a linear inverse box model was developed. Stablishing relationships between biogeochemical tracers based in stoichiometric ratios, sources and sinks of carbon, nutrients and oxygen in the eSPNA are estimated. This approach allows the assessment of biogeochemical cycles in an integrated budget model using well-known elemental relations as conversion factors. The evaluation relies on a robust circulation scheme of the mean circulation in the northern North Atlantic [Daniault et al., 2016] and quantitative tracer transports estimates, allowing an integrated evaluation that is independent of assumptions about biogeochemical behavior and with an output consistent between tracers. On decadal time scale, the biogeochemical budget model confirms that the eSPNA is a significant atmospheric carbon sink. The eSPNA is a net autotrophic region where biological primary production exceeds respiration, and where an organic carbon export to sediment of 38±10 Tg-Corg·yr-1 occurs. The hydrographical conditions and the specific location of the Irminger Sea, located just at the confluence of the two limbs of the AMOC, where around 10 Sv of deep-water formation occurs, make it an ideal hotspot for tracers to be long-time sequestered in the deep ocean. The Irminger Sea, representing less than 20% in area, is the place where the 48% of the organic and 46% of the CaCO3 sediment flux in the eSPNA takes place. We suggest that in a similar way, light-to-dense water formation processes are the reason for the oxygen uptake (856±74 kmol·s-1) and the heat release that occurs in the eSPNA, and that the transport of recently ventilated waters, with high-oxygen concentration, in the southward flowing lower limb of the AMOC is the principal mechanism for the equatorward flux of oxygen observed in the North Atlantic. The predicted ocean deoxygenation due global warming is not detected with our dataset. There are two complementary explanations: in the heterogeneous ocean, deoxygenation at global scale can happen without alteration in the oxygen levels of the eSPNA; and/or a longer time series is needed in this highly dynamic location with strong natural variability, in order to detect trends in oxygen deviations. One of the main conclusions in the biogeochemical budget, is that within the variability observed in the time period 2002-2016, there are not important changes in the surface-to-bottom biological carbon cycle in the eSPNA. Finally, the sink detected by the water-column balance approach was contrasted against sediment fluxes. With newly retrieved superficial sediment cores attained at the OVIDE/BOCATS 2016 cruise and a novel technique to stablish absolute chronologies, a quantitative first-estimate of carbon fluxes into the deep-sea sediment of the subpolar basins Irminger and Iceland was reached. As a way to link the time-scale of biogeochemical processes potentially affected by anthropogenic climate change with geological processes that usually acts at longer time-scales, we focus in the contemporary times, in the recently proposed geological epoch called Anthropocene. The geochronology based on high-resolution low background gamma-ray spectrometry with two simultaneous hyper-pure germanium (HPGe) detectors is shown as an accurate technique and sensitive enough for dating deep-sea pelagic sediments. The chronological framework based in the natural radionuclide 210Pb reveals that the average sedimentation rate is 0.95±0.74 mm·yr-1 in the central Irminger Basin; and 0.79±0.58 mm·yr-1 in the east flank of the Reykjanes Ridge in the Iceland Basin. Both rates are quite high for deep-sea sediments. Applying the dating model Constant rate of Supply (CRS), the time-evolution of sedimentation rates could also be evaluated. With the exception of short-time deposition events, both cores show quite constant sedimentation rates, implying that the deposition framework was uniform during the Anthropocene. Combining accurate chronologies with geochemical characterizations, allows the quantification of elemental fluxes into the sediment. Inter-annual variability in the sedimentation rate and in the terrigenous and biogenic proxies suggests short time-scale changes of bottom currents and/or biological productivity at the subpolar North Atlantic during the Anthropocene. The proxies support an enhanced biological productivity since the latest 50 years in both basins, but it is too soon to infer any reliable trend in the flux of carbon since the onset of the Anthropocene. The flux analysis concluded that during the Anthropocene 37.9±15 g Cinorg·m-2·yr-1 and 5.5±1.1 g Corg·m-2·yr-1 are deposited in the Irminger Sea, while 55.8±26.5 g Cinorg·m-2·yr-1 and 6.5±2 g Corg·m-2·yr-1 are deposited in the Iceland Basin. This burial sedimentation estimates supposes that a substantial carbon sink in the area exists, quantified in more than 26 Tg-C·yr-1 for the Irminger and almost 60 Tg-C·yr-1 for the Iceland basin. Taking together this two important basins of the subpolar North Atlantic, the annual quantity of organic carbon preserved in the sediments is almost 10 Tg-Corg·yr-1. In summary, an integrated approach it is a requirement not only to comprehend the carbon cycle, but also to quantify it. The combination of observations coming from different methodologies and the utilization of open access databases along with modelling efforts, allowed to analyze carbon fluxes in a dynamical complex region. The relevance of the North Atlantic as a carbon sink area in the global carbon cycle was once again confirmed under a broad set of approaches. Following the order of the research chapters, we conclude that: (i) the DOC downward export mediated by the AMOC supposes a significant fraction (~30%) of the total carbon sink and has to be accounted for in current ocean carbon models and in studies carried out by the biogeochemical and microbiological communities. (ii) Transports and budgets of carbon, nutrients and oxygen in the subpolar North Atlantic can be taken as a baseline in future evaluations of the Atlantic carbon cycle, and should be used to test and/or constrain global ocean models outputs. And (iii), it is possible to evaluate contemporary fluxes of carbon into deep-sea marine sediments with the combination of high-resolution dating techniques and geochemical analysis. The unabated increase in atmospheric CO2 levels, together with the whole set of anthropogenic perturbations in the natural cycle of elements, will require further evaluations on the assumption that the biological carbon cycle it is in steady state. The results of this memory can be seen as a try to integrate the organic/inorganic and natural/anthropogenic components of the carbon cycle in the North Atlantic on its fluxes between the pools of atmosphere-ocean-sediments. We are aware that this study-case can be improved with the addition of more constraining information, specially about water column seasonality in circulation and tracer concentrations, and that some estimates like the external additions to the system or the area scaling of sediment fluxes were deduced in an oversimplified way. Even so, the combination of different sources of data along with a set of well stablished biogeochemical relationships in the elemental composition of organic matter, allows the assessment of marine biogeochemical budgets in an integrated way. We suggest that this kind of comprehending approach is the one required in the evaluation of potentially changing biogeochemical cycles.