AO INSU 2011

Section « Océan-Atmosphère » – MISTRALS

Dossier scientifique

SPEciMed –

Structures of Planktonic Ecosystems in the North-western Mediterranean

Foreword

The following document is the program that was produced in late 2009 and proposed for the first call of the ‘Chantier Méditerranée’ Project. This program received a favorable opinion of the selection committee but has, nevertheless, not been funded by the MISTRALS Project in 2010. The project was again included in the overall MISTRALS Project in response to the 2011 call for a du-ration which was extended until 2014. Following the recommandations of the first selection commit-tee, SPEciMed keeps on the same objectives but wishes to gradually initiate the same type of ap-proach in other Mediterranean sectors (Nice-Calvi transect in collaboration with the Université de Liège, transfer of competences towards the University of Bizerte then the University of Algiers, with which co-operations have been in progress with the LOPB for a few years).

The following document has been amended to include the present situation of SPEciMed. In-deed, despite the absence of institutional funding, the program was launched in May 2010 and con-tinued until now, however in a reduced geometry. The program was initially restricted to the site of Marseille and a monthly visit of the two stations for CTD sampling and plankton and biogeochemi-cal parameters. At the end of 2010 we have implemented the physical oceanography aspects by add-ing a regular deployment of SCAMP measurements of turbulence as well as a deployment trial of the MiniBat undulating towed vehicle equipped with a CTD. Since the beginning of 2011 we have been able to add the regular deployment of the MiniBat in between our two stations. From March 2010 we have included the deployment of LISST and LOPC. An application for 4 cruises with NO Tethys II has also been sent recently.

Details of the implementation (as well as an updated detailed budget) are indicated in light blue at the end of the following document.

Bernard Quéguiner, PI of the SPEciMed Project

2

AO INSU 2010 Section « Océan-Atmosphère » – MISTRALS

Dossier scientifique Référence du projet : AO2010- 501593

Nom du porteur du projet : Bernard QUÉGUINER

SPEciMed –

Structures of Planktonic Ecosystems in the North-western Mediterranean

1 - Project Summary SPECiMed is designed as a part of MISTRALS to be included to MERMEX in medium-term strategy of Enhanced

Observing Periods (EOPs). In comparison with Long Observation Periods (LOPs) implemented in the framework of MOOSE (Mediterranean Ocean Observation multi-Sites on Environment), SPECiMed positions itself as an EOP designed to meet its specific scientific goals. MOOSE will benefit from SPECiMed results for its strategy implementation of long-term monitoring of biological parameters. Within MERMEX SPECiMed will be more focused on the objectives of WP1 “Stock and fluxes of biogenic elements: Impact of hydrodynamic changes on Mediterranean biogeochemical budgets” and WP2 “Ecological proc-esses: Biogeochemistry and food-web interactions”.

The rapid development of the Mediterranean basin had significant positive impacts on living standards of people but it was largely achieved at the expense of environmental balances essential to human well-being. With increasing anthropogenic pressure, the Mediterranean basin has now become an endangered environment both in terms of its ecological balance and exploitable resources and of its water systems that sustain its activities.

Regarding the marine environment, despite the intensive research efforts undertaken in the Mediterranean Sea for over a century, an integrated vision of how its ecosystems function is still lacking. Yet this knowledge is indispensable to meet the expectations of the Mediterranean basin development and sustainable management issues it raises.

In the north-western Mediterranean (NWM), studies on the impact of climate on plankton communities are limited by the small number of long time series data. Nevertheless, few studies have addressed the question of the long-term drift in com-position and dynamics of plankton. A long-term evolution of phytoplankton communities has been at least detected in several places of the French NWM coast, e.g. at DYFAMED station and especially in the Gulf of Lions (GoL), during the research projects EC2CO/GolPhyZ and the ongoing EU/SESAME partly devoted to plankton series data mining. The decadal variabil-ity of coastal phytoplankton in the Bay of Marseille from 1994 to 2006 suggests a close link to the North Atlantic Oscillation (via processes that still need to be assessed at the mechanistic level), a possible regime shift in the years around 1999, as well as signs of biodiversity loss.

The Mediterranean Sea is often compared to the World Ocean given its thermohaline anti-estuarine circulation. It is also characterized by an eastward gradient of oligotrophy associated with a succession of different plankton communities. There-fore it is difficult to observe the evolution of the Mediterranean as a whole. Even if trends can be predicted using numerical models, these must be validated continuously in view of ongoing climate change. Therefore, the regional level appears appro-priate. At first glance, the NWM basin is a mosaic of nested ecosystems offering similarities with the general situation of the World Ocean: An estuary at the mouth of a great river, the Rhône River, which brings locally large nutrient loads on a conti-nental shelf, the GoL, and a coastal current, the Northern Current (NC), which separates the land-to-ocean aquatic continuum from an oligotrophic gyre.

SPECiMed aims at establishing a three-year observation platform of plankton communities incl. bacteria, phyto-, micro-zoo- and mesozooplankton and associated biogeochemical cycles of major elements (C, N, P, and Si). As a step towards opera-tional management of marine ecosystems SPECiMed will develop a predictive understanding of how marine biogeochemical cycles and ecosystems respond to changing forcings, including how large-scale climatic variations impact regional ecosystems quantitative functioning through the changing physical dynamics and the alteration of biogeochemical cycles. SPECiMed relies on the joint expertise of oceanographers from the fields of physics, chemistry and biology to comprehend the whole reactional continuum ultimately controlling the response of ecosystems.

The 3-years project relies on the existing SOMLIT coastal stations of Marseille (SOFCOM) and Banyuls/mer (SOLA) and will implement the two shelf stations JULIO (SE entrance of the GoL) and MOLA (SW exit). Implementation also con-cerns the SOMLIT stations for new parameters of biodiversity observation (bacteria, phytoplankton, and zooplankton). Spe-cific objectives assigned to SPECiMed include the establishment of a multi-sites plan of joined observations relying on the OSUs of the GoL, the implementation of modern methods of investigation to describe the temporal evolution of the coast-shelf continuum (moving vessel profilers equipped with in situ particle analyzers, in situ chemical analyzers), and the development of a coupled physical-biogeochemical modelling approach as a predictive tool for the temporal evolution of disturbed ecosys-tems.

3

2 - General Scientific Background The IPCC conclusions are clear: climate change is underway and its effects in the half century

ahead are partly inevitable (IPCC, 2007). The issue of climate change and its ecological, therefore socie-tal, outcomes has become the focus of international concerns within the space of barely two decades.

This is particularly true in the Mediterranean basin. Its rapid development in recent decades had significant positive impacts on people living standards but it was largely achieved at the expense of envi-ronmental balances essential to human well-being. With increasing anthropogenic pressure, the Mediter-ranean basin has now become an endangered environment both in terms of its ecological balance as its exploitable resources and water systems that sustain its activities.

Regarding the marine environment, despite the intensive research efforts undertaken in the Mediter-ranean Sea for over a century, an integrated vision of how its ecosystems function is still lacking. Yet this knowledge is indispensable to meet the expectations of the Mediterranean basin development and sustain-able management issues it raises.

2.1 The plankton community: a complex community

Oceanic productivity, fishery yields and the net marine sequestration of atmospheric greenhouse gases are all controlled by the structure and function of planktonic communities (Karl et al., 2001).

Planktonic organisms generally are classified on the basis of size, nutritional, and physiological characteristics or phylogeny. Regardless of the criterion used, broad diversity is revealed, as illustrated in Figure 1Erreur ! Source du renvoi introuvable.. Thus, the planktonic organisms include several size-classes, from < 1 µm to several cm, so that sampling as well as quantitative observation of all requires the use of several complementary means of investigation, each appropriate to one part of the size spectrum.

The physical, chemical and biological interactions occurring in pelagic ecosystems operate on a range of time and space scales from seconds to decades and from microns to tens of kilometres (Erreur ! Source du renvoi introuvable.). Time-series programs provide the ideal opportunity to resolve seasonal, inter-annual and decadal variability of pelagic community structure in the context of other relevant bio-geochemical data.

Planktonic organisms generally are classified on the basis of size, nutritional, and physiological characteristics or phylogeny. Regardless of the criterion used, broad diversity is revealed, as illus-trated in Figure 1Erreur ! Source du renvoi in-trouvable.. Thus, the planktonic organisms include several size-classes, from < 1 µm to several cm, so that sampling as well as quantitative observation of all requires the use of several complementary means of investigation, each appropriate to one part of the size spectrum.

The physical, chemical and biological inter-actions occurring in pelagic ecosystems operate on a range of time and space scales from seconds to decades and from microns to tens of kilometres (Erreur ! Source du renvoi introuvable.). Time-series programs provide the ideal opportunity to resolve seasonal, inter-annual and decadal variabil-ity of pelagic community structure in the context of other relevant biogeochemical data.

Figure 1 : Representative classification of planktonic organisms by size showing the diversity of various autotrophic and heterotrophic groups (modified from Karl, 1999). The blue area corresponds

4

to the compartments for which historical data are available in the GoL, the pink area corre-sponds to the compartments for which detailed information is lacking.

2.2 Global change and changes of plankton communities

On relevant time scales to society (e.g. decades to centuries), oceanic biological processes sequester large quantities of atmospheric carbon, thereby modulating CO2 concentrations in the lower atmosphere (IPCC, 2001) together with physical processes regulating CO2 invasion in oceanic waters.

Figure 2 : Spatiotemporal representation of the interactions between physico-chemical and ecological proc-esses occurring in the mesope-lagic ‘twilight’ zone (Robinson et al., submitted).

While global change is accelerating (nutrient over-enrichment and eutrophication of downstream estuarine and coastal ecosystems–e.g. Yunev et al., 2007; Diaz & Rosenberg, 2008; Conley et al., 2009–, increase in the temperature of water masses–e.g. Levitus et al., 2000; Millot et al., 2006–, appearance of anomalous warming of surface seawater –Garrabou et al., 2009; Coma et al., 2009–,decrease of seawater pH–e.g. Caldeira & Wickett, 2003; Orr et al., 2005; Doney et al., 2009–, contaminant release at global oceanic scale), there is a paucity of information on the manner in which the structure and function of planktonic communities mediating production and cycling of major elements (i.e., C, N, O, P, Si) in pe-lagic ecosystems are altered. However, changes in physical and chemical forcings resulting from global change and rising emissions of greenhouse gas are likely to affect marine biodiversity, structure and func-tioning of marine ecosystems, the level of exploitable resources, biogeochemical cycles, and finally ele-mentary fluxes at the interface with other reservoirs (atmosphere, continent).

5

Figure 3 : Contour plots of mean monthly Phytoplankton Colour Index (a semi-quantitative representa-tion of the total phytoplankton biomass) from the Continuous Plankton Recorder during 1948–95 for the central North Sea, central northeast Atlantic and northern northeast Atlantic (adapted from Reid et al., 1998).

2.2.1 Regime shifts are underway…

The complex physical and biogeochemical interactions that regulate carbon fluxes between the at-mosphere and the surface ocean –physical events, fluctuations in community structure and function, natu-ral climate cycles, and long-term changes in anthropogenic forcing– are best studied within the frame-work of ocean time-series observations. Biogeochemical models and primary productivity algorithms based on historical or even contemporary data sets may not be accurate predictors of future trends (Karl et al., 2001) because they are representative of past ecological conditions. These biogeochemical models and algorithms of primary productivity should be improved through better observations of the complex relationships linking marine biogeochemistry and community structure. Although most long-term bio-logical time series are based on once-a-season or even less frequent observations, a growing number of studies further supports the idea that community- and/or functional-level analyses of plankton should help elucidate processes of ecosystem changes, which may be concealed otherwise (Karl et al., 2001; Mackas et al., 2001; Batchelder & Powell, 2002; Beaugrand et al., 2002; Hooff and Peterson, 2006; Chiba et al., 2008). Biogeochemists and ecologists are facing a difficulty due to the fact that many regime shifts are already underway in every part of the World Ocean (e.g., Reid, 1977; Reid et al., 1998 –see Figure 3; Beaugrand & Reid, 2003; Chavez et al., 2003; deYoung et al., 2008; Beaugrand 2009).

Therefore, we must adopt a strategy of long lasting observation to 1) understand how systems cur-rently in place are structured given the fact that they are already subject to disturbance, 2) monitor current evolutionary trends in order to provide relevant data to improve coupled biogeochemistry / ecology mod-els for predicting the evolution presently taking place in ecosystems and identify mitigation actions and management options.

Figure 4 : Interannual evolution of the vertical distribution (left) and depth-integrated stocks of diatom (fucoxanthine) and cyanobacteria (Divynil-chlorophyll a) diagnostic pigments at the DY-FAMED site during the 1991–1999 period (Marty et al., 2002).

2.2.2 The situation in the North-western Mediterranean Sea…

The Mediterranean Sea is very sensitive to human influence and climate forcing (Turley, 1999). In the north-western Mediterranean, studies on the impact of climate on phytoplankton communities are limited by the small number of long time series data. Nevertheless, few studies have addressed the ques-tion of the long-term drift in composition and dynamics of plankton (Gómez & Gorsky, 2003; Pereira Quiroga, 2005; Tunin Ley et al., 2007). Through various methods, the long-term evolution of phytoplank-ton communities has been at least detected. At station DYFAMED, using HPLC pigment data, Marty et

6

al. (2002) demonstrated a trend of increased biomass of phytoplankton they interpret as a response to the lengthening of the summer stratification period. This trend is mainly an increase of small phytoplankton, especially cyanobacteria and nanoflagellates (Figure 4). These observations are consistent with those made by Goffart et al. (2002) which already indicated a particular sensitivity of phytoplankton to envi-ronmental conditions, particularly temperature.

The objective of DYFAMED, under the international JGOFS program was primarily to get a long-term series of biogeochemical parameters but no long-term (i.e., more than a seasonal cycle) study of plankton community structure has actually been conducted there.

In a recent paper, Conversi et al. (2009) have review the interannual variations of the dominant co-pepod species in the Gulf of Trieste, Adriatic, and identified two periods (1970–1987 and 1988–2005) characterized by ecosystem-wide changes. Long-term evolution was evidenced by an approximate dou-bling in total copepod abundance, the arrival of a new species (Diaixis pygmoea), the rise (Paracalanus parvus, Oncaea spp., Oithona spp., and Euterpina acutifrons) or decline (Pseudocalanus elongatus, Clausocalanus spp.) of several taxa, and changes in the phenology of several species, with predominantly forward shifts in the timing of the maximum peak. Conversi et al. (2009) ascribed the long-term trend in zooplankton community structure to the general warming in the sea surface temperature and the associ-ated northerly displacement of the ecosystem, as well as to the basin-wide changes that altered the Medi-terranean surface and deep circulation since 1987, known as the Eastern Mediterranean Transient.

Figure 5 : Long-term fluctuations of selected species of copepods of the Gulf of Trieste from 1970 to 1985 (A) – Diaixis pygmoea is a sentinel species never observed before 1987–1988 in the study area; other species evolutions show a rise or a decline characterizing the regime shift. Changes in the seasonal peak of Pseudocalanus elongatus: Changes in the timing of the sea-sonal peak (B) and differences in the mean seasonal cycle for the 1970–1987 versus 1988–2005 periods (C) (redrawn from Conversi et al., 2009).

2.2.3 State of the art knowledge of the pelagic ecosystems of the Golfe du Lion

Under the European Commission's FP6 integrated program SESAME and the INSU/EC2CO GoL-PhyZ (Gulf of Lion: Characterization of Phytoplankton and zooplankton successions and implications for the modelling of biogeochemical cycles) project (PI: B. Quéguiner), the LOPB undertook a data mining

7

and study of biogeochemical parameters previously acquired simultaneously with the identification and enumeration of plankton populations on the French coastal area. This study, carried out especially during three Master training internships1, has allowed a first description of the interannual evolution of pelagic ecosystems. The studies were mainly based upon data acquired by the French CNRS-INSU monitoring program SOMLIT (Service d'Observation en Milieu LITtoral) and parallel programs at two stations (Banyuls/mer and Marseille) regarding phytoplankton and/or zooplankton data collections.

The nutrient environment

Major efforts were undertaken by the community SOMLIT allowing the current availability of high quality data of biogeochemical parameters. The three Mediterranean stations show different trends. When looking at nutrient limitation of the first trophic level (Figure 6), the series of Banyuls/mer and Marseille both show a gradual change that results in an increased period of limitation by phosphate which has gradually extended from January to the first 6 months of the year between 1996 and 2007. The situation appears more stable in the bay of Villefranche/mer. This contrast partly reflects changes in river flows during the last decade in the GoL. Recently, Ludwig et al. (2009) have documented a drastic increase of the NO3/PO4 ratio of Rhône River freshwater from ~20 to ~80 in between 1990 and 2000. This trend is a general feature of the Mediterranean Sea and appears related to mitigation strategies for urban pollution treatment targeting the reduction of phosphorus emissions. Ludwig et al. (2009) also enlighten the in-crease of potential limitation of marine primary production by both phosphate and silicic acid, the latter being more a problem for coastal phytoplankton community changes (shift from diatoms to flagellates).

Figure 6 : potentially limiting nutrients in order of priority (see codes inside the left graphs) for the three Mediterranean SOMLIT stations between 1996 and 2008 (adapted from M. Girault, not pub-lished, for details on the data analysis method see Leblanc et al., 2003).

1 Johann Rousselot, 2007 – Etude des évolutions des populations zooplanctoniques. Distributions spatiales et temporelles de

Centropages typicus et Temora stylifera, under supervision of F. Carlotti. Mathias Girault, 2008 – Caractérisation des communautés planctoniques en relation avec la biogéochimie en Méditerranée

Nord Occidentale, under supervision of B. Quéguiner. Boris Espinasse, 2008 – Etude à différentes échelles des variations spatio-temporelles des communautés de copépodes du Golfe du Lion, à partir de données historiques compilées de 1957 à 2003, under supervision of F. Carlotti & D. Nerini.

8

The phytoplankton community

Chlorophyll a concentrations show contrasting trends between the three network stations (Figure 7). Although located in the coastal area, the station of Villefranche/mer shows a marked oligotrophic charac-ter, a classic situation generally explained by the proximity of the NC. The station of Banyuls/mer, under the influence of the drift of river plumes, has high values rising up to 2–4 mg m–3 in spring. The temporal evolution of the Bay of Marseille is perhaps the most intriguing. Until 2003, its waters show an oligotro-phic character close enough to what is observed in Villefranche/mer. From 2004, the chlorophyll a con-centrations in the bay of Marseille increase gradually, reaching in 2007 levels comparable or superior to those of Banyuls/mer.

Figure 7 : Evolution of chlorophyll a concentrations at the different SOMLIT stations on the French Mediterranean coast between 1997 and 2007 (adapted from M. Girault, not published).

The data on the taxonomy of phytoplankton are more disparate as it is not a core parameter of the SOMLIT network. On the Mediterranean coast, the Centre d'Océanologie de Marseille is the only one to have a consistent set of data acquired through the hard work (“travail de bénédictin”) done by Mrs. Beatriz Beker. An examination of taxonomic data collected since 1994 on a fortnightly basis shows a high variability of phytoplankton developments in coastal waters. Seasonality of blooms is only slightly marked although they tend to occur during winter and spring when the environment is replenished in nu-trients after strong wind (mistral) events. If no trend in the long-term background is observed, the data set shows, however, a disruption marked by decreasing cell concentrations around the years 1996-1997 (Figure 8).

Figure 8 : Long-term evolution of total microphytoplankton concentrations at station SOMLIT in the Bay of Marseille from 1994 to 2006. The red curve represents the 12 months central moving average of daily interpolated data (adapted from M. Girault, not published).

Of the 228 taxa of diatoms and of dinoflagellates that were observed in the Bay of Marseille, only a few shows an interesting development that is manifested by their appearance or their disappearance in the ecosystem. Thus the dinoflagellate Prorocentrum minimum decreases sharply after 1999, while the genus

9

Gymnodinium seems to find more favourable conditions for its development. For diatoms, from the same period, Chaetoceros curvisetum becomes very common while Thalassionema frauenfeldii almost disap-pears in the recordings. These species can be considered as sentinel species whose presence or absence is a reflection of changes in environmental conditions. It is interesting to note that the period shift corre-sponds well to a persisting maximum in the monthly NAO index temporal evolution.

Figure 9 : Evolution of 4 sentinel species of the Bay of Marseille in parallel to that of the monthly NAO index (the green curve represents 12 months central moving average of the monthly NAO in-dex) (adapted from M. Girault, not published).

Figure 10 : Long-term evolution of the Shannon (1948) index (H) and of the Pielou (1966) evenness (E) of microphytoplankton in the Bay of Marseille. The red curves show the variations of the 6 months central moving average of daily interpolated data normalized against the average of each data set. The blue lines represent the trends of declining diversity for the 1994–2006 pe-riod.

Generally speaking, changes in biodiversity shows a surprising opposition with that of the NAO in-dex. Thus, the period 1999-2000, characterized by a positive index, corresponds to an impressive de-crease of the Shannon index and the Pielou evenness concomitant in with evolution of sentinel species mentioned above. The decrease of biodiversity also matches the seasonal extension of potentially limiting conditions for phosphate although available nutrient data do not offer a sufficient background to describe accurately the evolution of the nutritional environment of phytoplankton for the period prior to 1996. It is therefore possible that anthropogenic changes have directly influenced the microphytoplankton commu-nity structure, including the dissolved Si/P ratio known to control the seasonal succession. However, the system does not appear to have suffered from severe disturbance, as no proliferation of flagellates has so far been observed over the study period. A second sharp decrease of biodiversity was observed in 2004, particularly marked for the H index, and occurring in parallel with an increase in phytoplankton bloom intensity (in terms of biomass) in the Bay of Marseille.

10

The accumulation of phytoplankton biomass reflects the imbalance between the production proc-esses and the consumption processes including grazing by zooplankton and the shift in phytoplankton species and biodiversity could be linked to changes in the community of grazers (Katechakis et al., 2002). It is therefore possible that the trend observed at the end of the series is a reflection of profound changes in the pelagic ecosystem.

The mesozooplankton community

Mesozooplankton taxonomy is not a core parameter of the SOMLIT network but studies have been conducted in parallel at Marseille’ SOFCOM and Villefranche/mer’ Point B stations. Based on the analy-sis of a 1966-1993 weekly time-series, Molinero et al. (2005) were able to document the opposite tempo-ral evolution of copepods and jellyfishes, suggesting that climate forcing on both trophic groups might act through water temperature changes.

Figure 11 : Spatial distribution of Calanus helgolandicus and Centropages typicus in the GoL during pe-riods of low (right panels) and high (left panels) seasonal abundances. The seasonal abun-dance is normalized by the relative abundance index which is the ratio of species abundance at a given station to the total abundance of the same species at all stations of the same cruise (adapted from B. Espinasse, not published).

During the SESAME program, studies on the distribution of zooplankton across the GoL and its connection with the NC helped to highlight the different behaviours of mesozooplankton species that seem directly influenced by the temperature. Some species such as Calanus helgolandicus, Temora stylif-era and Centropages typicus have shifted and reduced their presence seasonal windows in the ecosystem in response to changing environmental parameters, including temperature. Maps of spatial distributions enable the separation between coastal, offshore, or ubiquitous species. They also indicate the importance of specific physical or topographic structures in the survival strategies of some species (Figure 11). The Rhone River plume seems particularly important for the survival of C. typicus in winter, while the deep areas of the canyons on the continental slope allow C. helgolandicus to remain in a state of diapause dur-ing the summer then allowing it to expand its area of distribution in winter after the cooling of shelf wa-ter. C. helgolandicus is found at its southern biogeographical limit in the North-western Mediterranean and is therefore a sentinel species which should be strongly influenced by water warming.

11

Comparing the time series conducted in the Bay of Marseille from 1960 to 1961 by Gaudy (1962) with the series initiated by F. Carlotti at station SOFCOM since 2002 provides valuable insights on the evolution of mesozooplankton diversity. Although the sampling conditions have varied it is clear that the diversity of mesozooplankton decreased between these two study periods. The average H index fell from 3.79 to 1.95, respectively between the 1960-1960 and 2002-2003 periods while evenness declined from 0.89 to 0.43. This decrease in diversity was accompanied by an increasing dominance of the genera Ca-lanus and Pseudocalanus which rose from 16.5 to 65.5% in the countings. Hence the changes in meso-zooplankton are similar to what has been described above for microphytoplankton. Changes in the dy-namics of mesozooplankton also resulted in radical changes of the seasonality characterized by develop-ments earlier in the season and could be related to altered climatic conditions.

The missing links

So far, studies that have been conducted in the GoL over periods long enough (i.e. > one year) are concerned with water circulation, trophic environment and microphyto- and mesozooplankton community structures. Regarding the structure of planktonic communities several compartments remain poorly stud-ied so far. This is the case of the bacterial compartment which, for reasons primarily methodological (techniques to address the bacteria "taxonomy" have developed only since few years), has been little in-vestigated. It is also the case, also mainly for methodological reasons, for the complex (auto-, hetero-, and mixotrophic) nanoflagellate compartment, as well as that of the microzooplankton. Yet these functional groups (as defined “plankton functional types” sensu Le Quéré et al., 2005) play an essential role in the dynamics of organic matter and probably also in the transfer of primary production to the classic food web.

Many challenging and up-to-date questions remain open worldwide for heterotrophic nanoflagel-lates (HNF): What is their abundance and what are their temporal dynamics? What is the HNF diversity? Is HNF diversity affected by the succession of their bacterial preys? Which are the dominant species? Is it possible to track less abundant species? In the Mediterranean there are no long or even medium term stud-ies on HNF. A recent review paper, based on all existing Mediterranean open waters studies and sup-ported by the Euroceans Network (Siokou-Frangou et al., submitted) mentions relationships between HNF and bacterial abundances which suggests that HNF are bottom-up (resource) controlled by bacteria (Figure 12).

Figure 12 : Relationship between heterotrophic nan-oflagellates (HNF) and bacterial prey abun-dances from the model of Gasol (1994). All HNF abundances fall below the Maximum Attainable Abundance (MAA) line while 70 and 75 % HNF fall above the Mean Realised Abundance for marine environment (MRA) line in the East and the West Mediterranean respectively, generally suggesting bottom-up control prevailing on HNF (from Siokou-Frangou et al., submitted).

The scientific rationale for examining bottom-up control of HNF abundance by bacteria can be re-versed to consider the top-down control of bacterial abundance by HNF. The direction in which this question is approached from normally depends on the specific scientific objective(s) of a given study. In the context of SPeciMed, it is two parts of the same question for understanding the ecological, trophic and biogeochemical structure of pelagic ecosystems. Our objective to quantify the abundance of nanoflagel-lates and bacteria will also enable us to estimate the rates of bacterivory by HNF as a function of physical, chemical and environmental parameters which may vary seasonally and interannually, as well as over shorter-timescales during which the ecosystem is perturbed. It is possible to estimate rates of HNF bac-terivory using empirical models which are parameterised only by the abundance of predator and prey

12

along with the temperature. In the NW Mediterranean these models are in excellent agreement with di-rect measures of HNF bacterivory (Figure 13, Unrein et al., 2007). The cycling of organic matter and transfer of energy between trophic levels in oligotrophic ecosystems, such as the NW Mediterranean, is highly dependant on this particular trophic link, which will be addressed systematically during SPECi-Med.

Figure 13 : Comparison of rates of HNF bac-terivory using the direct method of fluorescently labelled bacteria uptake and disappearance with the output of the empirical model of Unrein et al. (2007) describing the relationship be-tween HNF community grazing rate and abundances of HNF and their bac-terial preys.

Ciliate abundance in the Mediterranean Sea at different sites and in different seasons displays a re-markably high variability (Siokou-Frangou et al., submitted). In a recent study in the W Med coastal wa-ters it was shown that among ciliates, specific mixotrophic species, can be more efficient grazers of phytoplankton than heterotrophs (Christaki et al., 2009). In another study (2006-2009) undertaken at the SOMLIT station in Wimereux the community structure of microheterotrophs was clearly shown to be shaped by the species community structure of their autotrophic preys (Grattepanche et al., in prep.). Con-sidering that microzooplankton is the preferential prey for copepods the relationship between microzoo-plankton community structure and the variation of copepod dominant species is clearly to be further in-vestigated. These “species dependent relations” –which can not be accessed by simple biomass studies– may have further implications to higher trophic levels and the overall ecosystem response to global change. Surprisingly, as for HNF, long term studies on ciliates in Mediterranean coastal waters are quasi inexistent (Modigh, 2001).

2.2.4 The physical environment

The physical environment of the GoL as far as its offshore limit is under the control of three major processes: Intrusions of the NC at the East entrance, freshwater inputs from the Rhone River plume and the generation of mesoscale eddies at the western exit, processes themselves largely controlled by the Western European climate via the frequency of prevailing winds (mistral and tramontane) and rainfall regimes.

The NC is the northern branch of the cyclonic circulation in the western Mediterranean. It flows along the continental slope in an eastward direction from its area of formation, the Ligurian Sea, out to the Catalan Sea through offshore GoL. Gatti (2008) documented occasional intrusions of the NC in sev-eral parts of the continental shelf of the Gulf of Lions, especially on the eastern plateau entrance, and pro-posed potential mechanisms of their generation processes. The intrusion flux ranges between 0.04-0.37 Sv which represents 4 to 30% of the flux of the NC. A realistic simulation on the first six months of 2002 with the SYMPHONIE model (Estournel et al., 1997) revealed that intrusions occur about three to four times per month with a duration of a few days to two weeks. Both in situ measurements and numerical modelling show that intrusions develop either as a separated branch of the main vein of the NC or as a part of the NC itself encroaching on the shelf. These two types of intrusions can change from one type to the other, both in time and in space.

Three wind situations can generate intrusions: mistral cessation, channelled mistral and the East wind. The first two situations conducive to intrusion can be explained by a single process: the heterogene-ity of the wind field. In the case of East wind two processes are acting together, Ekman transport and dis-placement of the NC core towards the coast, are favouring intrusions. Other factors include the vertical and horizontal extents of the NC as well as its degree of mesoscale instability. In contrast, neither the sea-

13

sonal variability of the NC intensity or the variation of water budget of the GoL seems to impact on the occurrence of intrusions.

The monitoring of seasonal cycles of carbon, nitrogen and silicon, and phytoplankton communities at station SOFi (Station d’Observation Fixe) had already indicated the existence of a high variability of the pelagic ecosystem at the eastern entrance of the GoL, a feature that had been linked to rapid changes of the local water mass circulation (Diaz, 2000; Leblanc et al., 2003, 2005).

Figure 14 : Example of an intrusion of the Northern Current by encroachment on the continental shelf of the GoL. AVHRR image of SST on 29/11/2002 at 2:32 am on which are superimposed hull-mounted ADCP measurements of the currents at 24 m during the GOLTS cruise. On the right diagrams, top to bottom: vertical sections of the east-west component (U), north-south com-ponent (V), amplitude and direction of currents along a shelf–open sea transect.

The other main feature of the circulation in the GoL is the occasional presence in its west part of an anticyclonic circulation (Figure 15), highlighted by Millot (1982) and recently re-examined by Hu et al. (2009) in the INSU/LEFE LATEX (LAgrangian Transport EXperiment) project (PIs A. Petrenko and F. Diaz). Hu et al. (2009) indicate that the eddy-like structures in the western part of the GoL, potentially influenced by the distal plume extension of the Rhône river, could play an important role in the shelf-offshore transport of nutrients and phytoplankton because of the presence of the NC nearby. The western cyclonic circulation has been shown to be related to the wind stress curl of the channelled northwesterlies (tramontane) by Petrenko et al. (1997) as previously hypothesized by Millot (1982). The study of the role of this anticyclonic structure on biogeochemical dynamics and on the shelf-offshore transport processes constitutes the core of the LATEX main cruise which will take place in fall 2010 and will develop a com-bined strategy of inert tracer release (SF6), Lagrangian drifters, satellite imagery, Eulerian moorings and numerical modelling. Preliminary results, mainly obtained from pre-project cruises and modelling, al-ready evidenced the potential role of the studied structure when the anticyclonic eddy at its life end ap-proaches the NC and interacts with it.

Figure 15 : (a) Infrared thermography on the August 1, 1977 and comparison of the signature of the anti-cyclonic eddy on (b) chlorophyll a map (SeaWiFs image processed with OC4 – courtesy

14

E. Bosc) and (c) the eddy, as identified by the wavelet analysis (contour black) of simulated relative vorticity at 20 m on July 25, 2001. (a, from Millot, 1982, b and c, from Hu et al., 2009).

Figure 16 : Time table of the major operations in SPECiMed indicating implementation of the continuous measurements, ship requirements, scheduled analy-sis for water and sediment trap samples, modelling steps and reporting.

2.3 Seasonal monitoring of the North-western Mediterranean Sea in a detailed ecological/biogeochemical/physical perspective

From this summary of the major results hitherto obtained in the area of interest, it follows a nesting of spatio-temporal scales including : 1) the seasonal variability of the regional circulation, chemistry, and plankton community structures, 2) a (sub)mesoscale variability directly influenced by the dominant role of wind on circulation and 3) a long-term evolution dictated by the weather regimes (may be related to the NAO) as well as by the possible drift of ecosystems and biogeochemical cycles as a result of global change.

SPECiMed aims to document the variability at all these scales by developing and implementing an adapted strategy using synoptic observation tools, in situ instrumented moorings, and discrete water sam-pling during monthly monitoring operations over several years of the entire planktonic ecosystem and of the associated biogeochemical variables.

The Mediterranean Sea is often compared to the Global Ocean given its thermohaline anti-estuarine circulation. The Mediterranean Sea has been shown to be characterized by an eastward gradient of oligotrophic associated with a succession of plankton community structures. Therefore it is difficult to observe the evolution of the Mediterranean as a whole. Even if such a trend can be predicted using nu-merical models, they must be validated continuously in view of ongoing climate change. Therefore, the regional level appears appropriate. At first glance, the NW Mediterranean basin is a mosaic of nested eco-systems offering also similarities with the general situation of the World Ocean: An estuary at the mouth of a great river, the Rhône River, which brings locally large nutrient loads on a continental shelf, the GoL, and a coastal current, the NC, which separates the land-to-ocean aquatic continuum from an oligotrophic gyre.

Figure 17 : Main physical features of the NW Mediterranean. The Rhône River enters the GoL continen-tal shelf which is also influenced by temporary intrusions of the coastal Northern Current boarding the central oligotrophic North-western Mediterranean Gyre. The general clockwise circulation in the GoL is under control of dominant northern (mistral) and north-western (tramontane) strong winds events and (sub)mesoscale structures (3-D aerial composite photo-graph from IGN Géoportail at http://www.geoportail.fr/index.do). b – diagram showing the structure of nested ecosystems typical of the NW Mediterranean basin.

SPECiMed will focus on the seasonal and annual evolution of biological community structures of the continuum bacteria–phytoplankton–zooplankton, to give new insights into: 1) the characterization of community dynamics for 2) the definition of functional types to implement the biogeochemical–ecological–physical coupled model that will be used to identify 3) the temporal trends of ecosystem struc-ture changes. SPECiMed will investigate the sensitivity of marine biogeochemical cycles and ecosystems to global change by deciphering the mechanisms regulating the interrelationships between phytoplankton diversity, biogeochemical macronutrient cycles and the physical environment, between bacterial and phytoplankton diversity, and between zooplankton and phytoplankton diversity, under the different weather regimes that influence the ecosystems of GoL through its physical and chemical environment.

These objectives cannot all be achieved within the framework of the strategy being implemented in starting MOOSE. The description of community structures at the finest level possible and in conjunction

17

with the biogeochemical parameters, remains a priority if we want to be able to characterize the changes taking place in the pelagic ecosystem of the GoL. The partition of elemental stocks (C, N, P, Si) in the different compartments is now possible and these data will be essential to validate the complex (multi-elements, multi-species) numerical models coupling ecological dynamics and biogeochemical cycles.

We propose to characterize the physical and biogeochemical conditions at the eastern entrance and the western output of the study area together with the community structure encompassing the complete size spectrum from bacteria to mesozooplankton. Taking advantage of the previous studies conducted in the GoL the coupled physical-biogeochemical model (Symphonie-Eco3M) will account for the evolution of the entire study area. The comparison with remote sensing data will provide the validation of this ap-proach.

Figure 18 : Map of stations to be monitored under the SPECiMed project. Stations of the SOMLIT program (SOFCOM and SOLA) will be implemented to ensure a careful monitoring of parameters identical to those of the two continen-tal slope stations (JULIO and MOLA)

Positions des stations :

SOFCOM 05°17’30’’ E – 43°14’30’’ N JULIO 05°15’00’’ E – 43°06’00’’ N MOLA 03°32’28’’ E – 42°27’08’’ N SOLA 03°08’42’’ E – 42°29’18’’ N

3 - Project Objectives We have assigned five individual objectives declined into several specific questions to SPECiMed:

1. To confirm and clarify our knowledge of the frequency and variability of Northern Current intru-sions and its relations to the overall exchange between the shelf ecosystem and the open sea. Specific questions: What is the range of the annual variability in the frequency of intrusions? Is it possible to link these intrusions to typical weather regimes in order to consider their future evolution in the context of climate change? What are the consequences on dissolved and particulate matter transport between the continental shelf and the open sea?

2. To study the vertical structure of important hydrodynamic properties (turbulence) on the distribu-tion of planktonic communities, with special interest in discrete concentrated biomass layers. Specific questions: Is there a robust relationship between the vertical distribution of phytoplankton biomass and the mixing rates in the surface layer? Are our investigative means adapted to the actual vertical structure of the planktonic ecosystem?

3. To assess the current status of the NW Mediterranean pelagic ecosystems through analysis of exist-ing and newly collected data that will represent an important landmark for ongoing future evolution. Specific questions: What is the current diversity of the individual compartments (bacteria, flagellates, microphytoplan-ton, microzooplankton, mesozooplankton) of the pelagic assemblages in the GoL? What are the characteristic structures of food webs across the GoL and for different seasons? How are they spatially distributed during the different seasons between the coastal zone and the conti-nental slope? What is the range of the annual variability in these food web structures? Do the offshore stations exhibit the same trend of biodiversity loss as that observed in the more coastal environment?

18

What is the importance of the (sub)mesoscale in the spatial structuring of planktonic communities via their control by the physical and the chemical environments?

4. To upgrade the Med-Eco3M model using forthcoming new coupled observations of the physical, biogeochemical, and community structure environments, including validation of seasonal runs. Specific questions: Can we realistically simulate by a mechanistic approach the main seasonal features of the pelagic ecosystem of the GoL? What are the annual budgets of major elements (C, N, P and Si) and what are the most important exchange mechanisms? How does the physical environment variability translate into biogeochemical cycles? What scenarios (including shift in community structure and changes of elemental budgets) can be considered for the future in conditions such as changes in river loadings, a general increase in wa-ter temperatures, or an increased stratification?

5. To provide a robust basis for the long-term observation and future monitoring of planktonic ecosys-tem structures in the NW Mediterranean that will be useful for the long-term implementation of MOOSE. Specific questions: Do we have the capacity to conduct a long-term integrated monitoring of physical, biogeochemical, as well as community structural properties of such a coast-to-ocean continuum? What is the best strategy to answer to environmental questions related to the drift of pelagic ecosys-tems in response to global change? Can we envisage the development of predictive tools based on feedback interactions between model simulations and data acquisition?

4 - Strategy The overall strategy is based on monthly visits to the 4 stations and, while underway transit between

coastal and slope stations off Marseille, on continuous acquisition of phyto- and zooplankton size spectra with ad hoc sensors placed on a profiler, and of current parameters by hull-mounted ADCP. To take full advantage of the complementary expertises of the three laboratories involved, each compartment meas-ured will be analyzed by a single laboratory, following the recommendations made by Lebaron & Quéguiner (2008, see Annex 2). The LOPB will support the analysis of suspended particulate matter (C, N, P, Si) and the taxonomy of microphyto- and mesozooplankton (including elemental partition using the different image analysis systems).The nutrients, except for measurements already made within SOMLIT, will be analyzed by LOPB and LOBB and intercalibrations will be conducted on an annual basis. The LOBB will support the analysis of bacterial diversity and will work in close coordination with the LOG for the analysis of nanoplankton communities. The analysis of physical parameters (ADCP, SCAMP) and profiler data will be made by LOPB.

4.1 Physical environment (objectives 1 and 2)

JULIO and MOLA stations will feature 300 kHz RDI ADCPs that will enable identifying periods of exchange between the continental shelf and the NC. These stations will be maintained for the 3 years pe-riod of data acquisition. During visits to the four stations monitored vertical profiles of turbulence coeffi-cients will be made using a new type of low cost, versatile instrument: a SCAMP (Self Contained Autonomous MicroProfiler), equipped with a fast temperature sensor, a fast conductivity sensor, an accu-rate conductivity/ temperature sensor, a chlorophyll fluorometer, and a PAR sensor. The preliminary study of the basic principles (function and use) of the instrument and a thorough literature search on pre-vious applications made sure of the effectiveness of the instrument for this purpose.

4.2 Biogeochemical environment and planktonic community structures (objective 3)

During visits to the four stations monitored vertical profiles of biogeochemical parameters will be made. We have restricted the sampling to the evaluation of standing stocks of nutrients (NO3, NH4, PO4, H4SiO4) and particulate matter (organic C, N and P, biogenic and lithogenic silica, HPLC pigments). Ver-

19

tical profiles of biological parameters will also be performed, including picoplankton distributions (flow cytometry), bacterial communities (FISH), and microplankton analysis (FlowCam sorting and micro-scopical image analysis for biomass partitioning between species). HNF counts will be made by flow-cytometry and nanoheterotroph diversity will be assessed by cell sorting and card-FISH. Vertical nets will provide samples for mesozooplankton monitoring: microscopical identifications, ZooScan analysis and biomass partitioning between species, and a special attention will be given to the gelatinous zooplankton including jellyfishes whose developments are getting problematic in the study area.

In between station visits, nutrients will be analyzed at high frequency at both offshore sites. These instruments will enable capturing both high-frequency and long-term nutrient (nitrate, nitrite, phosphate, silicic acid and ammonium) dynamics. In situ analyzer deployments will be restricted to stations MOLA and JULIO to focus on nutrient variability coupling with short-term and episodic physical and/or biologi-cal events. The MOLA site has already acquired such an autonomous in situ analyzer.

During transits from the coastal to the offshore station of each location, the (sub)mesoscale spatial distribution will be assessed, based on the data collected by in situ captors which are in the course of vali-dation at LOPB : LOPC (Laser Optical Plankton Counter) and LISST (Laser In-Situ Scatterometer and Transmissometer), associated with a CTD sensor and a chlorophyll fluorometer. Preliminary tests re-cently conducted during the INSU–EC2CO/COSTEAU 2009 cruise (PIs: F. Carlotti/M. Zhou) have dem-onstrated the feasibility of this technique in the study area concerned.

4.3 Getting the regional picture through coupled physical/biogeochemical modelling (objective 4)

The Symphonie physical code and the biogeochemical modelling tool Eco3M (Ecological Modular Mechanistic Model) developed by the LOPB (Baklouti et al., 2006-a, b) will be used to integrate the ob-servations all over the GoL. The coupled model Eco3M MED / Symphonie, developed by F. Diaz is now operational (Herrman, 2007). It will be implemented in terms of specific compartments using data ac-quired by SPECiMed. Zooplankton processes have been incorporated into Eco3M with a focus on the functional responses of dominant populations in the GoL and the representation of predation preferences in a multi-prey environment. A size-structured zooplankton module has recently been coupled to the code Eco3M (Einsenhauer et al., 2009). Such size structured models are able to assimilate size spectrum data from captors such as LOPC or LISST and to derive the fluxes of matter between zooplankton and prey and predators and within zooplankton (Zhou et al., 2009). The coupled model including the trophic levels from nutrients to zooplankton will be used to simulate the observed seasonal changes. The validation will be done through the use of acquired data and satellite imagery. Following this validation, several scenar-ios of climate disruptions will be tested.

4.4 Towards an integrated biogeochemical/physical observation system (objective 5)

SPECiMed's ambition is also to serve as a demonstrator of the capacities of several OSUs (Observa-toire des Sciences de l’Univers) from the Mediterranean coast to develop an Integrated Regional Observ-ing Strategy. After three years of data acquisition, a study will be conducted to draw lessons from the operation in terms of feasibility by the community on a longer time scale, the adequacy of the initial strat-egy vis-à-vis the assigned objectives, and ability to develop a predictive tool susceptible to open into as-pects of ecosystem management sensu largo. Such an approach seemed reasonable to get a sufficient dis-tance before implementing a monitoring strategy for the long-term which will take its place within the MOOSE program.

4.5 Data management

The results obtained will be openly available through scientific publications, presentations at con-ferences, and press articles. All raw data will also be available on the LEFE/CYBER national database (http://www.obs-vlfr.fr/proof/index_vt.htm), in close coordination with the INSU ‘base de données’ unit. This approach will ensure that the data acquired will be visible during the life of the project and available at the end of it. The visibility of the project supervision is strengthened by planned intermediary report-ing, on an annual basis, which will review the quality of monitoring, the sample analyses and the data supply to the database. Finally, this project will support the work of students (Master 2, Thesis) and post-doctoral fellows hosted by the different laboratories involved.

5 - References Baklouti M., Diaz F., Pinazo C., Faure V., Quéguiner B.

(2006a) Investigation of mechanistic formulations de-picting phytoplankton dynamics for models of marine pelagic ecosystems and description of a new model. Progress in Oceanography, 71:1-33.

Baklouti M., Faure V., Pawlowski L., Sciandra A. (2006b) Investigation and sensitivity analysis of a mechanistic phytoplankton model implemented in a new modular numerical tool (Eco3M) dedicated to biogeochemical modelling. Progress in Oceanography, 71:34-58.

Batchelder, H.P., Powell, T.M., 2002. Physical and bio-logical conditions and processes in the northeast Pa-cific Ocean. Progress in Oceanography, 53, 105-114.

Beaugrand G. (2009) Decadal changes in climate and ecosystems in the North Atlantic Ocean and adjacent seas. Deep Sea Research Part II: Topical Studies in Oceanography, 56:656.

Beaugrand G., Reid P.C. (2003) Long-term changes in phytoplankton, zooplankton and salmon related to cli-mate. Global Change Biology, 9:801-817.

Beaugrand G., Reid P.C., Ibanez F., Lindley J.A., Edwards M. (2002) Reorganization of North Atlantic Marine copepod biodiversity and climate. Science, 296, 1692-1694.

Caldeira K., Wickett M.E. (2003) Anthropogenic carbon and ocean pH. Nature, 425:365.

Chavez F.P., Ryan J., Lluch-Cota S.E., Niquen C.M. (2003) From anchovies to sardines and back: multide-cadal change in the Pacific Ocean. Science, 299:217-221.

Chiba S., Aita M.N., Tadokoro K., Saino T., Sugisaki H., Nakata K. (2008) From climate regime shifts to lower-trophic level phenology: Synthesis of recent progress in retrospective studies of the western North Pacific. Pro-gress In Oceanography, 77:112-126.

Christaki U., Courties C., Joux F., Jeffrey W.H., Neveux J., Naudin J.J. (2009) Community structure and trophic role of ciliates and heterotrophic nanoflagellates in Rhone River diluted mesoscale structures (NW Medi-terranean Sea). Aquatic Microbial Ecology (in press).

Coma R., Ribes M., Serrano E., Jiménez E., Salat J., Pas-cual J. (2009) Global warming-enhanced stratification and mass mortality events in the Mediterranean. Pro-ceedings of the National Academy of Sciences of the United States of America, 106:6176-6181.

Conley D.J., Paerl H.W., Howarth R.W., Boesch D.F., Seitzinger S.P., Havens K.E., Lancelot C., Likens G.E. (2009) Controlling eutrophication: Nitrogen and phos-phorus. Science, 323:1014-1015.

deYoung B., Barange M., Beaugrand G., Harris R., Perry R.I., Scheffer M., Werner F. (2008) Regime shifts in marine ecosystems: detection, prediction and manage-ment. Trends in Ecology and Evolution, 23:402-409.

Diaz F. (2000) Evolution saisonnière de la production primaire et des processus d'assimilation - régénération de l'azote dans le golfe du Lion. Estimation d'un bilan de carbone. Approches in situ et modélisation. Thèse de Doctorat, Université de la Méditerranée, Aix-Marseille II, 341 pp.

Diaz R.J., Rosenberg R. (2008) Spreading dead zones and consequences for marine ecosystems. Science, 321:926-929.

Doney S.C., Fabry V.J., Feely R.A., Kleypas J.A. (2009)

Ocean acidification: The other CO2 problem. Annual Review of Marine Science, 1:169-192.

Eisenhauer L., Carlotti F. , Baklouti M. , Diaz F. (2009). Zooplankton population model coupled to a biogeo-chemical model of the North Western Mediterranean Sea ecosystem. Ecological Modelling (in press).

Estournel C., Kondrachoff V., Marsaleix P., Vehil R. (1997) The plume of the Rhône: numerical simulation and remote sensing. Continental Shelf Research, 17:899–924.

Garrabou J., Coma R., Bensoussan N., Bally M., Chevaldonné P., Cigliano M., Diaz D., Harmelin J.G., Gambi M.C., Kersting D.K., Ledoux J.B., Lejeusne C., Linares C., Marschal C., Pérez T., Ribes M., Romano J.-C., Serrano E., Teixido N., Torrents O., Zabala M., Zuberer F., Cerrano C. (2009) Mass mortality in North-western Mediterranean rocky benthic communities: ef-fects of the 2003 heat wave. Global Change Biology, 15:1090-1103.

Gasol J.M. (1994) A framework for the assessment of top-down vs. bottom-up control of heterotrophic nanoflag-ellate abundance. Marine Ecology Progress Series, 113:291-300.

Gatti J. (2008) Intrusions du Courant Nord Méditerranéen sur la partie est du plateau continental du Golfe du Lion. Thèse de Doctorat, Université de la Méditer-ranée, Marseille, 156 pp.

Goffart, A. Hecq, J.H. Legendre, L. (2002) Changes in the development of the winter-spring phytoplankton bloom in the Bay of Calvi (NW Mediterranean) over the last two decades: a response to changing climate? Marine Ecology Progress Series,. 236:45-60.

Gómez F., Gorsky, G. (2003) Annual microplankton cy-cles in Villefranche Bay, Ligurian Sea, NW Mediterra-nean, Journal of Plankton Research, 25:323-339.

Herrmann, M., 2007. Formation et devenir des masses d'eau en Méditerranée nord-occidentale – Impacts sur l'écosystème planctonique pélagique – Variabilité inte-rannuelle et changement climatique. Thèse de Docto-rat, Université Paul Sabatier, Toulouse, 250 pp.

IPCC (2001). Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assess-ment Report of the Intergovernmental Panel on Climate Change [Houghton, J.T.,Y. Ding, D.J. Griggs, M. No-guer, P.J. van der Linden, X. Dai, K. Maskell, and C.A. Johnson (eds.)]. Cambridge University Press, Cam-bridge, United Kingdom and New York, NY, USA, 881pp.

IPCC (2007). Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 996 pp.

Karl D.M., Bidigare R.R., Letelier R.M. (2001) Long-term changes in plankton community structure and produc-tivity in the North Pacific Subtropical Gyre: The do-main shift hypothesis. Deep Sea Research Part II: Topical Studies in Oceanography, 48:1449-1470.

Katechakis A., Stibor H., Sommer U., Hansen T. (2002) Changes in the phytoplankton community and micro-bial food web of Blanes Bay (Catalan Sea, NW Medi-terranean) under prolonged grazing pressure by doliolids (Tunicata), cladocerans or copepods (Crusta-cea). Marine Ecology Progress Series, 234:55-69.

21Leblanc K., Quéguiner B., Garcia N., Rimmelin P., Raim-

bault P. (2003) Silicon cycle in the NW Mediterranean Sea: seasonal study of a coastal oligotrophic site. Oceanologica Acta, 26:339-355.

Leblanc K., Quéguiner B., Raimbault P., Garcia N. (2005) Efficiency of the silicate pump at a coastal oligotrophic site in the Mediterranean Sea. Biogeosciences, 2:219-229.

Le Quéré , C., S.P. Harrison, I.C. Prentice, E.T. Buiten-huis, O. Aumont, L. Bopp, H. Claustre, L. Cotrim da Cunha, R. Geider, X. Giraud, C. Klaas, K. Kohfeld, L. Legendre, M. Manizza, T. Platt, R.B. Rivkin, S. Sathyendranath, J. Uitz, A.J. Watson, D. Wolf-Gladrow (2005). Ecosystem dynamics based on plank-ton functional types for global ocean biogeochemistry models. Global Change Biology Vol. 11 (11), p. 2016

Levitus S., Antonov J.I., Boyer T.P., Stephens C. (2000) Warming of the World Ocean. Science, 287:2225-2229.

Lomas M.W., Bates N.R., Knap A.H., Karl D.M., Lukas R., Landry M.R., Bidigare R., Steinberg D., Carlson C.A. (2002) Refining our understanding of oceanic biogeochemistry and ecosystem functioning. EOS, Transactions, American Geophysical Union, 83:559-567.

Ludwig W., Dumont E., Meybeck M., Heussner S. (2009) River discharges of water and nutrients to the Mediter-ranean and Black Sea: Major drivers for ecosystem changes during past and future decades? Progress In Oceanography, 80:199-217.

Mackas L., Thomson R.E., Galbraith M. (2001) Changes in the zooplankton community of the British Columbia continental margin, 1985-1999, and their covariation with oceanographic conditions. Canadian Journal of Fisheries and Aquatic Sciences, 58:685-702.

Marty J.C., Chiavérini J., Pizay M.-D., Avril B. (2002) Seasonal and interannual dynamics of nutrients and phytoplankton pigments in the western Mediterranean Sea at the DYFAMED time-series station (1991-1999). Deep Sea Research Part II: Topical Studies in Ocean-ography, 49:1965-1985.

Millot C. (1982) Analysis of upwelling in the Gulf of Li-ons. In: Nihoul J.C.J. (Ed.), Hydrodynamics of Semi-Enclosed Seas: Proceedings of the 13th International Liège Colloquium on Ocean Hydrodynamics. Elsevier Oceanographical Series, vol. 34, Elsevier Scientific Publications, Amsterdam, The Netherlands, pp. 143-153.

Millot C., Candela J., Fuda J.-L., Tber Y. (2006) Large warming and salinification of the Mediterranean out-flow due to changes in its composition. Deep Sea Re-search Part I: Oceanographic Research Papers, 53:656-666.

Orr J.C., Fabry V.J., Aumont O., Bopp L., Doney S.C., Feely R.A., Gnanadesikan A., Gruber N., Ishida A., Joos F., Key R.M., Lindsay K., Maier-Reimer E., Matear R., Monfray P., Mouchet A., Najjar R.G., Platt-ner G.-K., Rodgers K.B., Sabine C.L., Sarmiento J.L., Schlitzer R., Slater R.D., Totterdell I.J., Weirig M.-F., Yamanaka Y., Yool A. (2005) Anthropogenic ocean

acidification over the twenty-first century and its im-pact on calcifying organisms. Nature, 437:681.

Petrenko A., Dufau C., Estournel C. (2008) Barotropic eastward currents in the western Gulf of Lion, north-western Mediterranean Sea, during stratified condi-tions. Journal of Marine Systems, 74:406-428.

Pereira Quiroga A.I. (2005) Evolution des populations phytoplanctoniques de la Baie de Banyuls-sur-Mer, avec une attention particulière pour les genres possé-dant des espèces potentiellement toxiques. Thèse de doctorat en Océanologie biologique, Université Pierre et Marie Curie, Paris, 186 pp.

Pielou E.C. (1966) The measurement of diversity in differ-ent types of biological collections. Journal of Theoreti-cal Biology, 13:131-144.

Reid P.C. (1977) Continuous plankton records: changes in the composition and abundance of the phytoplankton of the northeastern Atlantic Ocean and North Sea, 1958-1974. Marine Biology, 40:337-339.

Reid P.C., Edwards M., Hunt H.G., Warner A.J. (1998) Phytoplankton change in the North Atlantic. Nature, 391:546.

Robinson C., Steinberg D.K., Anderson T.R., Arístegui J., Carlson C.A., Frost J.R., Ghiglione J.-F., Hernández-León S., Jackson G.A., Koppelmann R., Quéguiner B., Ragueneau O., Rassoulzadegan F., Robison B.H., Tamburini C., Tanaka T., Wishner K.F., Zhang J. (submitted) Mesopelagic ecology and biogeochemistry – a synthesis. Deep Sea Research Part II: Topical Studies in Oceanography.

Shannon C.E. (1948) A mathematical theory of communi-cation. Bell System Technical Journal, 27:379-423 & 623-656.

Siokou-Frangou I., Christaki U., Mazzocchi M.G., Montresor M., Rassoulzadegan F., Ribera d’Alcalá M., Vaqué D., Zingone A. (submitted) Plankton in the open Mediterranean Sea. A review. Biogeosciences.

Tunin-Ley A., Ibañez F., Labat J., Zingone A., Lemée R. (2009) Phytoplankton biodiversity and NW Mediterra-nean Sea warming: changes in the dinoflagellate genus Ceratium in the 20th century. Marine Ecology Pro-gress Series, 375:85-99.

Turley C.M. (1999) The changing Mediterranean Sea - a sensitive ecosystem? Progress in Oceanography, 44:387-400.

Unrein F., Massana R., Alonso-Sáez L., Gasol J.M. (2007) Significant year-round effect of small mixotrophic flagellates on bacterioplankton in an oligotrophic coastal system. Limnology and Oceanography, 52:456-469.

Yunev O.A., Carstensen J., Moncheva S., Khaliulin A.,

Aertebjerg G., Nixon S. (2007) Nutrient and phyto-plankton trends on the western Black Sea shelf in re-sponse to cultural eutrophication and climate changes. Estuarine, Coastal and Shelf Science, 74:63-76.

Zhou, M. , Carlotti, F., Zhu, Y. (2009). A size-spectrum zooplankton closure for ecosystem models. Journal of plankton Research (in revision).