Carbon dioxide dynamics and air-seaCO2fluxes in several Southern European Seas
A.V. Borges (1), B. Delille (1), J.-M. Beckers (2), and M. Grégoire (3)
(1) Chemical Oceanography Unit, University of Liège, Liège, Belgium (firstname.lastname@example.org, +32-(0)4-3663367), (2) Physical Oceanography Laboratory, University of Liège, Liège, Belgium, (3) Oceanology Laboratory, University of Liège, Liège, Belgium
Seasonal and inter-annual variability of air-seaCO2fluxes and seawater carbonate chemistry in the Southern North Sea
Nathalie Gypens (1), Geneviève Lacroix (2), Christiane Lancelot (1), and Alberto V. Borges (3)
(1) Université Libre de Bruxelles, Faculté des Sciences, Ecologie des Systèmes Aquatiques, CP-221, Bd du Triomphe, B-1050, Belgium (email@example.com), (2) Management Unit of the North Sea Mathematical Models, Royal Belgian Institute of Natural Sciences, 100 Gulledelle, B-1200 Brussels, Belgium, (3) Université de Liège, Unité d’Océanographie Chimique, Institut de Physique (B5), B-4000, Belgium
We carried out measurements of pCO2 within brines and bulk ice, and related air-ice CO2fluxes (chamber method) in Antarctic first year pack ice (“Sea Ice Mass Balance in Antarctica –SIMBA” drifting station experi- ment September – October 2007) and in Arctic first year land fast ice (“Circumpolar Flaw Lead” – CFL, April – June 2008). These 2 experiments were carried out in contrasted sites. SIMBA was carried out on sea ice in early spring while CFL was carried out in from the middle of the winter to the late spring while sea ice was melting. Both in Arctic and Antarctic, no air-ice CO2fluxes were detected when sea ice interface was below -10°C. Slightly above -10°C, fluxes toward the atmosphere were observed. In contrast, at -7°C fluxes from the atmosphere to the ice were significant. The pCO2 of the brine exhibits a same trend in both hemispheres with a strong decrease of the pCO2 anti-correlated with the increase of sea ice temperature. The pCO2 shifted from a large over-saturation at low temperature to a marked under-saturation at high temperature. These air-ice CO2fluxes are partly controlled by the permeability of the air-ice interface, which depends of the temperature of this one. Moreover, air-ice CO2fluxes are driven by the air-ice pCO2 gradient. Hence, while the temperature is a leading factor in controlling magnitude of air-ice CO2fluxes, pCO2 of the ice controls both magnitude and direction of fluxes. However, pCO2 in Arctic is significantly higher than in Antarctica. This difference could be due to a higher level of organic matter in Arctic. The degradation of this organic matter fuel CO2 efflux from the ice to the atmosphere in early spring. We observed evidence of CaCO3 precipitation, but only at the top of the ice. Implications in term of air-ice CO2 transfer of such CaCO3 precipitation will be discussed. In addition, salt-rich snow appears to strongly affect air-ice CO2fluxes in the arctic.
A 3D coupled biogeochemical–hydrodynamic model (MIRO-CO 2 &CO) is implemented in the English Channel (ECH) and the Southern Bight of the North Sea (SBNS) to estimate the present-day spatio-tem- poral distribution of air–sea CO 2 ﬂuxes, surface water partial pressure of CO 2 (pCO 2 ) and other compo- nents of the carbonate system (pH, saturation state of calcite ( X ca ) and of aragonite ( X ar )), and the main drivers of their variability. Over the 1994–2004 period, air–sea CO 2 ﬂuxes show signiﬁcant inter- annual variability, with oscillations between net annual CO 2 sinks and sources. The inter-annual variabil- ity of air–sea CO 2 ﬂuxes simulated in the SBNS is controlled primarily by river loads and changes of biological activities (net autotrophy in spring and early summer, and net heterotrophy in winter and autumn), while in areas less inﬂuenced by river inputs such as the ECH, the inter-annual variations of air–sea CO 2 ﬂuxes are mainly due to changes in sea surface temperature and in near-surface wind strength and direction. In the ECH, the decrease of pH, of X ca and of X ar follows the one expected from the increase of atmospheric CO 2 (ocean acidiﬁcation), but the decrease of these quantities in the SBNS during the considered time period is faster than the one expected from ocean acidiﬁcation alone. This seems to be related to a general pattern of decreasing nutrient river loads and net ecosystem production (NEP) in the SBNS. Annually, the combined effect of carbon and nutrient loads leads to an increase of the sink of CO 2 in the ECH and the SBNS, but the impact of the river loads varies spatially and is stronger in river plumes and nearshore waters than in offshore waters. The impact of organic and inorganic carbon (C) inputs is mainly conﬁned to the coast and generates a source of CO 2 to the atmosphere and low pH, of X ca and of X ar values in estuarine plumes, while the impact of nutrient loads, highest than the effect of C inputs in coastal nearshore waters, also propagates offshore and, by stimulating primary production, drives a sink of atmospheric CO 2 and higher values of pH, of X ca and of X ar .
and 10-day river (Seine and Scheldt) loads calculated by the RIVERSTRAHLER model. Because of the lack of data, a climatology corresponding to the 1989–1999 period (Lancelot et al., 2005) was used as global solar radiation. Land-based fluxes of DIC and TA were esti- mated based on a compilation of DIC and TA data in the Seine and the Scheldt rivers (Frankignoulle et al., 1996, 1998; Frankignoulle & Borges, 2001; A. V. Borges un- published data; G. Abril, personal communication) and river discharges, making use of the ‘apparent zero end- member’ method (Kaul & Froelich, 1984). The ‘apparent zero end-member’ concentration of DIC and TA was computed as the y-intercept of the linear regression of these quantities as a function of salinity for values between 15 and 30. Such an approach allows computing realistic river fluxes to the coastal zone, by removing the effect of nonconservative behaviour of chemicals in upper estuaries (e.g. TA in the Scheldt estuary; Fran- kignoulle et al., 1996) that leads to erroneous estimates if the real concentration at zero salinity is used. Because of the scarcity of TA and DIC data in the Seine river, constant concentrations of these quantities were used to compute the river fluxes. For the Scheldt river, linear regressions of DIC or of TA as a function of the logarithm of freshwater discharge (Q) were used to determine the concentration of DIC or TA for a given Q value:
4.3 How did the food web respond and feed back to
organic and inorganic C cycling?
There is no other variable that more effectively connects bio- geochemical carbon cycling with food web activities than NCP. It accounts for the balance between PP and respira- tion, and thus of other metabolic processes in between such as biomass build-up and reproduction. Since NCP is tightly linked with both CO 2 and O 2 cycling, NCP is typically based on an assertion of O 2 dynamics using bottle incuba- tions or geochemical tracers. At the heart of such NCP es- timates currently lies a very active debate on the inconsis- tency between the results of in vitro O 2 bottle incubations and in situ O 2 / Ar-based mixed-layer NCP, as exhaustively reviewed by Ducklow and Doney (2013). Here, we adopted another approach by summing up respiration rates of food web components, which were subsequently subtracted from gross PP estimates that were also assessed empirically. While our approach has its own uncertainties (e.g., conversion fac- tors, plankton biomass estimates), we are confident that the overall picture of ecosystem metabolism across the complex Arctic shelf system that we sampled was better estimated this way. In fact, assuming a steady state between mixed- layer NCP and sea-to-airfluxes (as the O 2 / Ar methodol- ogy requires) would have yielded a wrong portrait of ecosys- tem metabolism, since the temporal decoupling between au- totrophic and heterotrophic processes was obvious. In addi- tion, O 2 bottle incubations would have provided only snap- shot bulk measurements without having the possibility to ver- ify which food web components were influencing ecosystem metabolism.
Black Sea in fall to +2.5 mmolC m -2 d -1 in the Cilician
bassin in fall. In Spring, all the SES regions were sinks for
atmospheric CO 2 . In fall, the more « open » SES areas were
sources of atmospheric CO 2 (Alboran, Levantine, Cilician &
We report the first two year of results from a 10m deep mooring over a Posidonia Oceanica seagrass meadow (Corsica, France) where we deployed from August 2006 to August 2008 an array of 3 optodes, a fluorometer and a sensor for measurements of the partial pressure of CO2 (pCO2). The oxygen data are used to compute by mass balance ecosystem metabolic performance rates (gross primary production, community respiration, net community production). The comparison with rates derived from discrete benthic incubations (every 2 months) is very satisfactory. The pCO2 data are used to assess the sink or source of atmospheric CO2 of the Posidonia Oceanica seagrass meadow. An application of such a mooring is to detect changes in the productivity of the Posidonia meadow that can be used as indicators of overall ecosystem “health” or degradation by human activities. Such a mooring can be used as an affordable and simple tool for management and sustainable development of coastal areas in the Mediterranean.
Flat first-year land-fast sea-ice near Barrow (Alaska), 1 km off the coast.
The source area for EC measurements at 2.8 m was well within the boundaries of the floe. Duration: from the end of January 2009 to the beginning of June 2009, before ice break-up.
Brine and underlying seawater pCO 2 were measured in situ using a custom-made equilibration system (Geilfus et al., 2012a). The system consisted of a membrane contac- tor equilibrator (Membrana, Liqui-cell) connected to a non- dispersive infrared gas analyzer (IRGA, Li-Cor 6262) via a closed air loop. Brine and airflow rates from the equili- brator and IRGA were approximately 2 and 3 L min − 1 , re- spectively. Temperature was measured within the sackholes or under-ice water and at the equilibrator outlet simultane- ously using Li-Cor temperature sensors. The pCO 2 values were temperature-corrected assuming that the Copin Mon- tégut (1988) relation is valid at low temperatures and high salinities. The IRGA was calibrated immediately upon re- turning to the ship while the analyzer was still cold. All de- vices, except the peristaltic pump, were enclosed in an in- sulated box that contained a 12 V power source providing enough heat to keep the inside temperature just above 0 ◦ C.
Sea ice dynamics and related air-sea CO 2 fluxes during a flood-freeze cycle (Bellingshaussen Sea, Antarctica)
N.X. Geilfus 1,2 , J.-L. Tison 2 , G. Carnat 2 , A.V. Borges 1 , S.F. Ackley 3 and B. Delille 1 .
Sea ice, a barrier to air-sea exchange of CO 2 ?
15 min. Conservation equations and parameters are detailed in Lancelot et al. (2004).
Model simulations were run with 1996–1999 climatolog- ical forcings for hydro-meteorological conditions and river inputs. These functions were computed from recorded daily global solar radiation (Oostende Station, Institut Royal de M´et´eorologie, Belgium), seawater temperature and monthly nutrient loads for the rivers Seine (Cellule Antipollution de Rouen du Service de la Navigation de la Seine, France) and Scheldt [Institute for Inland Water Management and Waste Water Treatment, The Netherlands) and Department of En- vironment and Infrastructure (Ministry of Flemish Commu- nity, Belgium). Organic carbon loads by the Scheldt were re- trieved from the Dutch water base (http://www.waterbase.nl). For the river Seine, we used data described in Servais et al. (2003). Land-based fluxes of DIC and TA were esti- mated based on a compilation of DIC and TA concentra- tions in the Seine and the Scheldt rivers (Frankignoulle et al., 1996, 1998; Frankignoulle and Borges, 2001b; Abril, per- sonal communication) and river discharges, making use of the “apparent zero end-member” method (Kaul and Froelich, 1984). Atmospheric pCO2 was extracted from the Mace Head (53 ◦ 33 0 N 9 ◦ 00 0 W, Southern Ireland) and the National Oceanic and Atmospheric Administration/climate Monitor- ing and Diagnostics Laboratory/Carbon Cycle Greenhouse Gases Group (NOAA/CMDL/CCGG) air sampling network (available at http://www.cmdl.noaa.gov/). Wind speed at 50.0 ◦ N 6.0 ◦ W was provided by the Pacific Fisheries En- vironmental Laboratory (PFEL) and is based on Fleet Nu- merical Meteorology and Oceanography Center (FNMOC) synoptic pressure fields.
Fig. 1. Conceptual model of seasonal carbon fluxes in ice-covered seas. The figure covers the upper 100 m of the water column. The processes
driving air–sea gas exchange are active throughout the cycle of sea ice formation and decay. During winter, ice growth results in the rejection of CO 2
along with salts dissolved in seawater from the ice crystal matrix, which gives rise to dense brine that sinks and is incorporated into intermediate and deep water layers. Subsequent sea ice melt during the summer thaw results in the formation of a strong halocline, with surface waters well below atmospheric CO 2 saturation, thus enhancing the uptake of atmospheric CO 2 from the ocean. In addition, primary productivity in both sea ice and
convection develops bottom-up in the sea ice cover and by fuelling microalgae primary production in nutrients, triggers strong undersaturation of CO 2 (pCO 2 down to 30 ppmV). Sea ice therefore
turns into a CO 2 sink with CO 2 fluxes ranging from 0 to -6 mmol.m -2 .d -1 depending, among other
parameters, on the ice texture. On the whole, spring and summer Antarctic pack ice appears to act as a CO 2 sink which magnitude could be of significant importance in the budgets of air-sea CO 2
Glaciology Unit, Université Libre de Bruxelles, Belgium
There are growing observations that sea ice exchange CO 2 directly with the atmosphere. To
explore the relationships between sea ice-specific biogeochemical processes and fluxes of
CO2 at the air-ice interface, we carried out three surveys which addressed spring and summer CO 2 dynamics in Antarctic land fast sea ice, and first year and multiyear pack ice.