1- Biogeochemical cycling of major nutrients is already changing in the Mediterranean Sea.

In the Algero-Provencal basin, the statistical analysis of historical data set of nitrate phosphate and silicic acid concentrations show contrasted trends for the different elements. In the deep water, the silicic acid concentration has remained nearly invariant since the early 1960s. In contrast phosphate and nitrate concentrations increased from the early 1960s and from 1972, respectively. The changes in N and P (0.5% yr-1 between 1960 and 1994) very likely result from anthropogenic activities which translate in a 0.3 % yr-1 increase of the atmospheric and terrestrial inputs of N and P. In contrast, Si input was mainly driven by natural processes. It has been suggested that the increase by a factor 3 of the input of N and P in the Mediterranean Sea has resulted in an increase of new production of 2.7. Consequently the amount of O2 consumed in the deep water to remineralise the exported material could also have increased. The dense water formation is a major process that supplies oxygen to the deep layers. If this process is kept constant, O2 in the deep water might decrease with time and critical values for the benthic fauna might be reached in the middle of the 21century.

How new production and carbon export will change during the next decades in response to the changing nutrient supply? What will be the consequences for the O2 concentration in the deep water?

In the next decade the weakening of the dense water formation is a possible consequence of the climate change in the Mediterranean Sea with two possible antagonist effects on the cycling of the major elements. In the surface waters, the increase in the stratification will reduce the supply of nutrients and will decrease the new production. This process should counteract the predicted increase in new production and in carbon export due to the raising of nutrient terrestrial discharge. In the deep water, the consumption of O2 for the remineralisation could be reduced due to the decrease in carbon export but concomitantly the supply of O2 to the deep water could also be reduced. The net effect on the O2 level in the deep water is difficult to predict.

The nutrient ratio (P:N) in the Mediterranean sea (1:24 in the eastern basin and 1: 22 in the western basin) deviates largely from the canonical values of Redfield. (1:15). Deficit of phosphate (for example due to scavenging by dust) or excess of nitrogen related to efficient dinitrogen fixation by diazotroph organisms have been proposed to explain the high P:N ratio.
The number of studies aiming to quantify the N2 fixation in the Mediterranean sea is growing rapidly, but the exact contribution of this process to the new production and to the N:P ratio has still to be determined. The organisms that are fixing N2 in Mediterranean Sea are not well known. But it is largely recognized that P and Fe are important nutrients for the diazotrophs. Atmospheric deposition is a major mode of supply for both elements (Bonnet and Guieu, 2006; Ridame et al., 2003; Guieu et al. 2002). The nature (anthropogenic versus natural dust) and the magnitude of the atmospheric deposit to the surface waters of the Mediterranean seas are also expected to change during this century (see section WP5). This will impact the nutrient cycling in a way that is actually very difficult to predict.

How will the steochiometry of the major element change?

A long term change of the N:Si:P ratio was also detected in the deep water of the Algero-Provencal basin where the ratio (N:Si:P) increased from 1:24:22 in 1960 to 1:19.5:22 in 2000. The decrease of the ratio Si:P is related to the difference in the time evolution of the supply ; Si was constant whereas P increased. This change may drive important changes in the structure of the phytoplankton community a different levels. A major shift from a diatom to a non diatom dominated system but also more subtle changes within the specific composition of the diatom community are possible. These shifts might favour episodic blooms of toxic phytoplankton species, promote shifts in the composition of the zooplankton community but also in return modify the cycling of the nutrients (Béthoux et al., 1990, 2002; Roether et al., 1998; Tsimplis and Baker, 2000)

2- Carbon cycling and acidification

The issues related to organic particulate matter are discussed in the previous paragraph but dissolved organic carbon is also an important pool in the carbon cycling. The available historical data set does not allow to draw conclusions on the impact of anthropogenic perturbation on the cycling of DOC. However, it is obvious that the magnitude of the sources, the nature and the mode of supply to the surface ocean, the mode of transfer from coastal areas to offshore or from surface to deep waters are not enough constrained.
Does the seawater Cant level increase with time in parallel with that in the atmosphere?

predicted that the wh

Increasing atmospheric concentration of CO2 results in increased seawater CO2 concentrations and consequently modified the carbonate equilibrium decreasing the pH. In a general context calcifying taxa are largely exposed to the acidification although the real effect of high level CO2 seawater is still debated. In the case of the Mediterranean Sea, in addition to the higher carbon input from the riverine waters, the Mediterranean Sea takes up a large part of the anthropogenic CO2 (Cant) injected into the atmosphere through air-sea interactions. Unfortunately, today, data of the carbonate system properties (total dissolved inorganic carbon, CT; total alkalinity, AT; pH; and CO2 partial pressure, pCO2) in the Mediterranean Sea are very scarce compared with other ocean areas.
As indicated by the levels of CFC-12 (Schlitzer et al., 1991), the ventilation of the deep Mediterranean waters is fast (less than 50 years). Consequently, it is not surprising that the recent estimates indicate that the deep layers are already contaminated by Cant (Aït-Ameur and Goyet, 2005; Touratier and Goyet, 2008).

Today, the scientific community is still unable to answer this question mainly because of the lack of data of the carbonate system properties (CT, AT, pH, pCO2). In particular, time-series measurements are necessary to determine and then forecast any temporal evolution of any ocean property. In the Mediterranean Sea there was the time-series station DYFAMED (DYnamique des Flux Atmosphériques en MEDiterranée; http://www.obs-vlfr.fr/sodyf/home.htm) located in the central part of the Ligurian Sea (43°25 N, 7°52 E). Although, there were relatively few and discontinued measurements of the carbonate system properties (CT, AT, pH, pCO2) at this station, it was possible via interpolations and use of the simple TrOCA approach (Touratier and Goyet, 2004a, 2004b, Touratier et al., 2007) to estimate the distribution of Cant over more than a decade (1993-2005) at the DYFAMED site (Touratier and Goyet, 2008). Surprisingly, the results indicated that in the intermediate and the deep waters, the concentrations of Cant decreased with time. The Cant and O2 signatures suggest that an invasion of one or several older water mass did occur during this decadal times-series. The older water masses may have originated either from the eastern basin as a consequence of the transients LT (Levantine Transient; Theocharis et al., 2002) and EMT (Eastern Mediterranean Transient event; Roether et al., 1998), or from the western basin via the transient WMT (Western Mediterranean Transient; Schroeder et al., 2008).

Nevertheless, the concentrations of Cant in the Mediterranean Sea are high, much higher than in the open oceans. Thus the Mediterranean Sea is a source of Cant for the Atlantic Ocean via the Gibraltar Detroit. A direct impact of this high accumulation of Cant in the Mediterranean Sea is a decrease of its pH by 0.15 since the pre-industrial era. However, in the case of the Mediterranean Sea, the actual high degree of supersaturation of carbonate indicates that the dissolution should not be a matter of concern in the coming years. But other effects of high CO2 waters are possible. For example at the phytoplankton level, the increase in CO2 could change the stoechiometry of C:N (Riebesell et al. 2007) or impact N2 fixation. It is also ole ecosystem could be impacted.

Questions :
How will the carbonate chemistry of Mediterranean Sea response to the increase in atmospheric CO2?
– Is the rate of Cant penetration in the Mediterranean Sea similar to its increase in the atmosphere?
– Is this Cant penetration rate variable with time?
– What is the impact of the observed pH change on air-sea exchange?
– What is the impact of the observed pH change on the buffer capacity of seawater?
– What will be the impact of the modified carbonate chemistry on the functioning of the ecosystem?