(click to go to ...) General context   General Objectives     The Southern Ocean and oceanic CO2 pump   A natural laboratory in the Southern Ocean: the Kerguelen plateau       Objective 1   Objective 2   Objective 3 KEOPS in the national context International co-operations / KEOPS and the international research programs     References cited

                                   

 

 

 

 

 

 

 

 

General background and interest for new carbon dioxide studies in the Southern Ocean has been already summarised in the previous sections and is not repeated here.

 

 

Estimates of annual global DMS emissions vary widely, but are expected to be in the range of  10  to 50 TgS yr^-1 (Intergovernmental Panel on Climate Change [IPCC], 1996).  The wide range results from uncertainties attached both to the global distribution of sea-surface DMS concentrations and to computing DMS air-sea exchange rates.  Besides reducing uncertainties on present-day emission estimates, it is also important to investigate the climate sensitivity of the marine DMS source. Although considerable progress has been made in understanding the marine and atmospheric biogeochemical cycle of DMS, the impact of global warming on marine DMS emissions remains to be established. Bopp et al. (2002) presented a modelled estimation of the response of DMS seawater concentrations and DMS fluxes to climate change, following previous work on marine DMS modelling (Aumont et al., 2002) and on the global warming impact on marine biology (Bopp et al., 2001). At 2xCO₂, the model estimates a small increase of global DMS flux to the atmosphere (+2%) but with large spatial heterogeneity (from ­15% to +30% for the zonal mean). In the 30°S-50°S band, the model predicts a +19% increase in the annual DMS flux (from 2.7 to 3.2 TgS yr^-1). Changes in DMS concentrations in that region result mainly from modifications of the community structure of the ecosystem (shift from diatoms to non-siliceous species). In situ experiments have shown that the availability of iron and silicate in the Southern Ocean plays an important role in controlling the structure of the phytoplankton community (Maldonado et al. 2001) and this can modify the budget of DMS (S. Turner pers. com.). The global 3-D ocean carbon cycle model used by Bopp et al. (2002) did not include these nutrient limitations. Consequently, these results strongly depend on the skill of the model to simulate marine productivity and on the relationships used between DMS and other biological variables. Improvements in both aspects are necessary to develop greater confidence in such future predictions. KEOPS will allow to better quantify the fluxes of DMS to the atmosphere and to better understand and parameterise the production processes in the ocean (Objective 3).

 

 In contrast to many previous studies that have investigated the sea to air transfer of lighter elements such as C and S, the importance of heavier elements in these processes is less recognised. Heavier elements so-called metalloids (Selenium (Se), Tellure (Te), Polonium(Po)) can form volatile alkyl species with reflux to the atmosphere.  The process of alkylation can occur either directly as a consequence of biological activity, or indirectly via abiotic reactions.  The pathways leading to the volatilisation and atmospheric transfer of selenium from oceanic environments are poorly understood. They may however affect the global distribution of selenium and its impact on marine and terrestrial ecosystems. Gaseous selenium compounds have been determined in the North Atlantic ocean during a spring bloom of identified phytoplankton species, known to be a large source of atmospheric sulfur. The results demonstrate that significant concentrations of gaseous selenium species occur in surface ocean waters, and their production is closely linked to the turnover of gaseous sulfur species.

 

 

The ocean is a significant source of the trace gases nitrous oxide (N₂O), methane (CH₄) and carbon monoxide (CO), which influence the radiative and oxidative capacity of the atmosphere. On both regional and global scales the marine source of these climate reactive gases is related to nutrient availability and phytoplankton production, and consequently any shift in nutrient supply, whether natural or deliberately induced, may have profound implications for atmospheric chemistry and climate . One such possibility is iron fertilisation of the oceans, which may occur naturally via climate-induced variations in supply, or by deliberate iron addition aimed at increasing carbon sequestration (Markels and Barber , 2001). Law and Ling (2002) reported that the decrease in radiative forcing resulting from carbon dioxide fixation and CO₂ uptake during SOIREE may be subsequently offset by 6-12% by N₂O production. It is therefore important that the potential impact upon biogenic gas cycling and oceanic emissions is adequately constrained (Fuhrman and Capone, 1991) for prediction of climate-induced effects and informed consideration of any artificial fertilisation.

 

The KEOPS experiment represents an ideal opportunity to examine the impact of natural fertilisation upon trace gas cycling, offering a direct comparison between “background” conditions and the fertilised waters in the wake of Kerguelen Island. This data will be invaluable in assessing whether the stimulation in N₂O and CO production observed in deliberate iron fertilisations (SOIREE - Southern Ocean, Law and Ling, 2001; SERIES - N.E. Pacific, Law, unpublished data) is also a feature of natural fertilisation events.

 

 

 

The oxidising capacity of the troposphere reflects the ability of the atmosphere to cleanse itself of man-made and natural compounds. It is primarily determined by the concentration of hydroxyl radicals (OH) which are formed mainly from the photodissociation of ozone by UV radiation. The emission of trace gases containing nitrogen and halogen (Cl, Br, I) atoms from the biosphere into the atmosphere affects the oxidising capacity, both as a source of reactive radicals such as NO₃, Cl and BrO, and as a result of their influence on the concentration of ozone. The alkyl nitrates are a reservoir species for NO_x (=NO₂ + NO). Photolysis of NO₂ is the only known way of producing ozone in the troposphere, therefore the photochemical processes occurring in the lower atmosphere are critically dependent on the level of nitrogen oxides. As the alkyl nitrates are relatively long lived in the troposphere, they can act as a source of NO_x in remote environments away from continental sources and so influence ozone concentrations on regional levels.

 

In general, the alkyl nitrates have a predominantly anthropogenic source, but during the 1990s, several authors invoked an oceanic source of the light alkyl nitrates (C₁-C₃) to explain the distributions seen over remote oceanic regions (Blake, 1999).  Recent studies in the Atlantic Ocean have confirmed a seawater source of methyl and ethyl nitrate (Chuck et al 2002), and their production mechanisms are, as yet, poorly understood.

 

The results of our participation in EisenEx 2000, an in situ iron addition experiment in the Southern Ocean, suggest that iron enrichment may increase methyl nitrate concentrations and has a variable effect on the concentrations of biogenic halocarbons (Chuck, unpublished).

 

 

 

 

Non methane hydrocarbons play a major role in the tropospheric chemistry as precursor of tropospheric ozone and their impact on the oxidizing capacity is relatively well known and quantified, mainly in the polluted atmospheres and in the northern hemisphere where anthropogenic emissions dominate .

 

The role of NMHC is very complex, its influence on ozone is non linear and dependent also on the NO_x level. NMHC may act as ozone precursors or ozone consumers depending on the considered atmosphere composition. Particularly in remote atmospheres a great uncertainty exists regarding the amplitude of the natural sources and the marine source. (T.J. Song, 2001). Our first estimations (Bonsang et al. 1988), have shown  the importance of the marine source of NMHC (particularly of alkenes) which has been estimated at roughly 35 x10⁶ Tons of carbon/year. A recent work by T.J. Song have shown that at levels of several hundreds pptv, this role may be significant in the atmospheric chemistry, and the ozone or oxidants budgets. Such atmospheric levels are compatible with the upper range of the oceanic source. Our time series of NMHC in the southern hemisphere at Amsterdam island are coherent with an oceanic source of the order of the upper level of the estimations, and more generally with the relatively high mixing ratios of NMHC observed in the southern Indian Ocean (Yokouchi et al., 1999;  Saito et al, 2000).

 

Our recent work undertaken at Kerguelen Island  in January February 2002, confirms the existence of a significant oceanic source either on the Kerguelen plateau or in this area of the Indian Ocean. The levels observed are very similar to the recent results of Saito et al. (2000). In a general way it appears clearly that subantarctic areas, particularly the southern Indian Ocean act as a significant sources of reactive trace gases, which can not be extrapolated to the global ocean, but should be taken into account  in their geographical variability.

 

 

 

Despite the fact that “iron from above” is probably not the main source of iron for the surface water of the Kerguelen plateau, an estimation of atmospheric deposition of iron is of major importance. On the global scale, there are still large uncertainties concerning the magnitude of total deposition rate of iron in the ocean. The lowest value, 6.6 10⁹ kg Fe yr^-1 is based on observations (Duce and Tindale, 1991) and the highest value, 32 10⁹ kg Fe yr^-1  is based on dust sources and a transport model (Fung et al., 2000). The uncertainty can be larger in the Southern Ocean where the observations are scarce. In addition the solubility of dust in seawater is also poorly constrained. This leads to a wide range of the estimation for the input of bioavailable iron in the Southern Ocean, from  0.2 10⁶ mol Fe yr^-1 (Lefevre and Watson 1999) to  2 10⁶ mol Fe yr^-1  (Measure and Vink, 2000).

 

Recently, Johnson (2001) investigated the iron supply and demand in the upper ocean and questioned : Is extraterrestrial dust a significant source of bioavailable iron ? Based  on input of interplanetary dust particles in the atmosphere and on the assumption that small particles, with a radius of the order of 1 nm, have a solubility of 100%, Johnson estimated that bioavailable iron input in the Southern ocean from IDPs is  0.35 10⁶ mol Fe yr^-1  this could account for 30 % of the terrestrial input of bioavailable iron and 20 % of the iron input due to upwelling flux in the Southern Ocean (Fung et al. 2000).

 

Interestingly, Johnson concludes « In fact it is likely that we know the flux of bioavailable Fe from IDPs with better accuracy than we know the flux of bioavailable iron from terrestrial dust ! ». This reinforces the need for a better quantification of the atmospheric iron deposition rate in the Southern Ocean, and for a better knowledge of the iron dissolution processes.

 

 

Specific questions :

 

 

 

 

                                   

 

During the first iron in situ fertilisation SOIREE, the phytoplankton  biomass peaked up to 3 µg l^-1 11 days after the beginning of the experiment, however  little export was observed. About 66 % of the photosynthetically fixed carbon remained in the mixed layer on day 13.  Lateral advection rather than grazing or export was the largest loss term. Boyd (2002) suggests that the SOIREE bloom –which was not significantly controled by grazing activity was not nutrient (including iron) limited due to the exchange with surrounding HNLC waters (chemostat effect) and hence phytoplankton remained healthy. As healthy diatoms sink relatively slowly, loss was also slow during the SOIREE experiment. The SOIREE bloom persisted for at least 50 days, longer than the delay observed by Buesseler et al. (2001) (in Boyd 2002) between the development of the bloom and the subsequent export along 170°W in the Southern Ocean. This could be due to a physical artefact during the SOIREE experiment. The lateral loss of phytoplankton could have prevented self-shading and phytoplankton aggregation.

 

 

The recent experiment SOFEX (Jan.-Feb. 2002) seems to lead to the same conclusions “….it looks as if natural variability will be as large as the iron-induced effects on particle export during this 5 week occupation of the South patch. While the details of this story will change, we did not witness the large-scale demise of the bloom and sinking out of carbon taken up by phytoplankton in response to iron…” K. Buesseler in SOFeX cruise logbook Februrary 2002.

 

 

The question of the carbon export following iron fertilisation remained an unsolved question that we will address during KEOPS

 

 

Related to objective 2.2, the major specific questions to be addressed within KEOPS are