Tréguer¹, Paul, and Uli Bathmann²

¹IUEM, Brest-France (Chair of the Southern Ocean Synthesis Group) and ²AWI, Bremerhaven, Germany (Vice-Chair of the SOSG)

Southern Ocean-JGOFS: a step forward (I)

The JGOFS effort in the Southern Ocean (SO) was coordinated by the SO JGOFS Planning Group, led successively by Julian Priddle (BAS, Cambridge, UK), Uli Bathmann (AWI, Bremerhaven, Germany) and Paul Tréguer (IUEM, Brest, France). SO-JGOFS addressed 6 majors questions.

Question 1: What role does the Southern Ocean play in the contemporary global carbon cycle? Present estimates of the Austral annual sink of atmospheric CO₂ vary between -0.1 to –0.6 GTC yr^-1 for the part of the SO >50°S. Up to day this negative air-sea d pCO₂ fluxes are not reconcilable with outputs of atmospheric inversion models validated from the few CO₂ Southern Hemisphere land stations. Increasing the number of land stations is requested. New approaches for a better integration of ocean and atmospheric data of CO₂ and O₂ are also strongly recommended.

The modelled penetration of anthropogenic CO₂ is very active south of 50°S but its storage is low because anthropogenic carbon is rapidly transported northward isopycnally into the SubTropical Convergence. The modelled interannual variability of the net atmospheric CO₂ sink, due to the 7-year cycle Antarctic Circumpolar Wave (linked to ENSO), is estimated at ± 0.2 GT C yr^-1.

Question 2: What controls the magnitude and variability of primary production and export production? Substantial uncertainty remains about the processes/factors that regulate primary productivity, and particularly its variability. Numerous studies showed the co-limitation of the primary productivity by Fe and/or Si. Iron fertilization experiments (SOIREE: 1999 ; EISENEX: 2000 ; SOFeX: 2001-2002) show the biological pump of CO₂ reacts rapidly to dissolved Fe addition, although the export of biogenic material out of the photic layer seems to be variable. SOFeX showed that 1 tonne of iron spread at the ocean surface could force only 1,000 T of carbon > 100 m, much lower than expected, leading to the conclusion that the Southern Ocean might not be a good target for our quest for atmospheric CO₂ sinks. EISENEX shows much larger export, altough indirectly.

Access to satellite data has allowed more realistic estimates of the primary productivity, ranging between 60 and 100 gC m^-2 yr^-1, i.e. 3 to 5 times higher than those of the beginning of the 1990s, deduced from extrapolated ¹⁴C measurements. The total annual primary production >50°S is estimated at 3 - 4 GT C yr^-1.

There is a big gap between studies which consider export fluxes out of photic layer (especially using ²³⁴Th techniques) and those concerned by the measurements of biogenic matter in deep waters and at the water-sediment interface. To take into account the processes that control the fluxes of remineralisation and recycling in the « twilight » zone (100-1000m) should be a high priority for future programmes.

Dissolved nutrient distributions have been used to derive the rate constants of biogeochemical processes responsible for the observed fields using inverse modelling. The export of POC the world ocean being 10 GTC yr^-1, about 1 GT C yr^-1 is exported >50°S. This shows the export production of biogenic matter out of the surface layer is very efficient for the SO, which is not in agreement with satellite derived estimates. If we are to give more realistic estimates for the export and primary production of the SO these two approaches have to be reconciled. The conversion of algorithms is especially questioned, in areas like the SO where subsurface chlorophyll maxima are not detectable by satellites.

Question 3: What are the major features of spatial and temporal variability in the physical and chemical environments and key biotic factors? The classical view of latitudinal bands of contrasted marine environments around the Antarctic continent, although still alive has been strongly shaken as numerous SeaWIFs images reveal the importance of west-east gradients, which has to do with eastward aeolian transports of trace-metals. The importance of the physical-biological coupling at mesoscale has been demonstrated both from SeaWIFS images and from circulation models. The Polar Front is now being regarded as a high export production system.

Although numerous sophisticated biogeochemical models are now available they remain preliminary tools to account for the complexity of the merry-go-round Antarctic ecosystems, characterised by the sequestration of limiting nutrients by hard or soft-armoured species (e.g. large diatoms and colonies of Phaeocystis antarctica). Large scale distributions of krill and salps (the two major large grazers of phytoplankton) show they usually inhabit in different environnments and are sensible to global change. To improve the models outputs in terms of carbon retention and/or export, attention is to be put on the role of key species in the key ecosystems, on the community structure and on the dynamics of the higher trophic levels.

Several data sets are available regarding the seasonal variations of total organic carbon (TOC) vs. latitude and of the bacterial production in surface waters. High concentrations of semi-labile TOC have been evidenced in frontal areas. They represent high levels of CO₂ potentially produced through bacterial respiration.