WP1 NET ECOSYSTEM DYNAMICS DURING COCCOLITHOPHORID BLOOMS

Net ecosystem production (NEP) is the balance between gross primary production (GPP) and community respiration (R), where NEP=GPP-R. Conversely, net ecosystem calcification (NEC) is the balance between gross CaCO₃ production (GC) and CaCO₃ dissolution (D), where NEC=GC-D. Whether an ecosystem is source or sink for atmospheric CO₂ depends on the sign of NEP and NEC. In all marine systems, there is a strong decoupling in time and space (in particular vertically throughout the water column) of GPP and R, and of GC and D. This decoupling is much looser in marginal seas than in open oceanic waters due to the strong benthic-pelagic coupling. Importantly, R and D are intense at the water-sediment interface and affect the photic zone on much shorter time scales in the shelf seas than in the open ocean. Hence, it is essential to study organic and inorganic carbon metabolism at the appropriate temporal and spatial scales.

Organic and inorganic carbon fluxes during coccolithophorid blooms will be determined using incubation and mass balance approaches, based on considerations of various parameters of carbon and 234Th derived fluxes. Particular attention will be given to the characterisation of carbon metabolism in relation to phytoplanktonic biomass and composition, and bacterial community composition. These carbon fluxes will also be interpreted in relation to general environmental forcings (mixed layer depth, solar irradiance, nutrient availability) based on standard hydrographic sampling (salinity, temperature, nutrients, oxygen, etc.).

The air-sea CO₂ fluxes will be compared to the mass balance approaches of CO₂ derived from NEP and NEC, at a station-wise basis and at a cruise-wise basis. This will allow us to unravel the role of the different stages of the coccolithophorid bloom in air-sea CO₂ exchanges.

Task 1.1 Net ecosystem organic carbon fluxes (NEP)

Gross primary production of organic carbon during cocolithophorid blooms will be determined by ULB during short-term incubation (4h) using 14C incorporation experiments. Esterase activity will be determined by ULB to evaluate the status of the bloom development. Vertical distribution of nutrients (NOx, PO4, Si) and dissolved organic carbon (DOC) will be determined by ULB: nutrients by classical colorimetric methods and DOC by the widely accepted high-temperature catalytic oxidation (HTCO) technique. Discrete underway samples will also be taken for nutrient measurements. The four parameters of the seawater inorganic carbon chemistry (pCO₂, pH, TA, DIC) will be determined by ULg using standard methods (respectively: equilibration, combined pH electrode, Gran potentiometric titration and acidification coupled to a NDIR analyzer). Sampling will be carried out on vertical profiles and underway on surface waters. Community respiration will be assessed by ULg by determining O2 changes (Winkler potentiometric end-point analysis) during incubations of pelagic water samples and benthic sediment cores. Phytoplankton samples will be taken by UGent for their identification by microscope examination and for their pigment determination by High Performance Liquid Chromatography (HPLC). Particulate samples will also be taken by UGent for coccolithophorid morphology and intra-specific identification using scanning electron microscopy (SEM). Bacterial samples will be collected by UGent on 0.2 µm filters and analyzed using Denaturing Gradient Gel Electrophoresis (DGGE) based on the 16S rDNA gene. Selected (dominant) DGGE bands will be excised and sequenced for identification purposes. Bacterial abundance will also be determined by UGent by epifluorescence counts of DAPI-stained cells. Particle residence time will be assessed by ULB using 234Th in collaboration with the University of Bordeaux. Particulate matter will be collected by ULB by centrifugation and by in situ pumping. Samples will be measured for CaCO₃ content by ULB by dissolution with acetic acid of particulate matter. Other major inorganic elemental composition such as aluminium will also be determined by ULB by inductively coupled plasma-atomic emission spectrometry (ICP-AES) after complete dissolution of the suspended matter. Particulate organic carbon (POC) content will be measured by ULB by high temperature catalytic combustion using an elemental analyzer.

Task 1.2 Net ecosystem inorganic carbon fluxes (NEC)

Production of particulate inorganic carbon will be evaluated by ULB also using 14C incorporation experiment by making the difference between total particulate 14C production and particulate organic carbon (PO14C) production. Dissolution of CaCO₃ in the sediments will be assessed using incubation experiments of sediment cores by ULg. Concentrations of TA and Ca2+ will be followed as a function of time by ULg.

Task 1.3 Net impact of NEP and NEC on air-sea CO₂ fluxes

Air-sea CO₂ fluxes will be evaluated by ULg with a high temporal and spatial resolution from underway pCO₂ measurements in seawater and the atmosphere (1 min. sampling frequency) and gas transfer velocity parameterisations as a function of wind speed derived from ship born measurements and reanalysis products (ERA-40 ECMWF and NCEP). The main biogeochemical drivers of pCO₂ (and hence the air-sea gradient and flux of CO₂) will be examined using a multiparameter cross-analysis with ancillary underway data (temperature, fluorimeter derived chl-a and O2 - 1 min. sampling frequency) and high temporal resolution discrete data (TA, DIC and nutrients).

WP2 UNRAVELLING THE LINK BETWEEN THE BACTERIAL COMMUNITY, GRAZING, TEP DYNAMICS, DMS CYCLING AND CARBON EXPORT DURING COCCOLITHOPHORID CALCIFICATION

In the ocean, the dissolved polysaccharides (PCHO) originating from phytoplankton exudation contain acidic sugars that can facilitate the aggregation of single chains through divalent cation bridging and/or ester-sulphate bonding between the anionic ends. PCHO aggregation leads to the formation of TEP that can alternatively be formed by the release of particulate material (mucus) from phytoplankton cell surfaces or colonies. TEP are known to contribute to the export of suspended material through particle aggregation. During field studies, we will examine particulate suspended matter composition in the water column in relation to TEP dynamics (abundance, size spectrum) in order to elucidate the potential for retention/export of C. TEP form also an important microenvironment for bacteria and may serve as hot spots for nutrients and substrate concentrations, and hence bacterial activity, or simply as attachment surfaces. Recent work suggests that concentration of bacteria in TEP may alter their availability to different grazer groups and hence may lead to short-circuiting of food chains, on the one hand by serving as a refuge against microzooplankton grazing, but on the other hand by making bacteria more available for larger grazers feeding on TEP. Bacteria, and especially α-proteobacteria related to the genus Roseabacter, have been shown to play a major role in DMSP consumption in E. huxleyi blooms. However, the phytoplankton cells themselves are also capable of DMSP cleavage. It is as yet not clear to what degree DMSP and DMS cycling is mediated by bacteria and algae respectively, and how this is influenced by grazing.

Task 2.1 Temporal and spatial distribution of TEP and their polysaccharide precursors during coccolithophorid blooms

To determine TEP production during coccolithophorid blooms, AWI will conduct a temporally and spatially highly resolved sampling procedure. Samples will be collected at least two times a day (early morning, late afternoon) at different depths within the surface mixed layer and below. The carbon content of TEP will be measured. Dissolved mono/ polysaccharides (MCHO and PCHO) will be determined. Dissolved neutral and acidic sugars will be analysed by ion chromatography on a Dionex ICS-3000. Prior to analysis, the sample will be pre-processed including a concentration, neutralization and desalting step. ULB will contribute to the colorimetric determination of TEP concentration in samples filtered onto 0.4µm Nuclepore filters using the Alcian Blue dye method. An indication of the relative importance of aggregation processes for TEP production can be derived from the size spectrum of organic matter. Aggregation of dissolved organic matter (DOM) results in removal of high molecular weight DOM (HMW-DOM) and a subsequent increase in microscopically visible particles, e.g., TEP, while degradation shifts the size spectrum of DOM towards the low molecular weight fraction. AWI will therefore determine the size frequency distributions of TEP and its dissolved precursors. The size distribution of TEP will be examined by microscopy and Image analysis (NIH-Image). To characterise the size spectrum of dissolved precursors, size fractionation of DOM - using different membrane cut-offs, e.g. 1kDa, 10kDa, 100kDa - and the measurement of micro-sized aggregates will be necessary. Task 2.2 Bacteria community structure and TEP dynamics

We will use microcosms to study the succession in bacterial community biomass and composition associated with phytoplankton (E. huxleyi) cells and TEP in relation to the phytoplankton population dynamics, and to the presence or absence of grazers. UGent will determine the phytoplankton biomass (Chl-a), bacterial biomass (epifluorescence counts of DAPI-stained cells) and community composition (DGGE) during the development and decline of an experimental E. huxleyi population inoculated with natural bacterial communities. UGent will also explore the use of isolation techniques to assess differences in bacterial dynamics and composition associated with the algal cells and TEP respectively, and bacteria present in 'TEP-free' water. TEP abundance and production will be quantified during these experiments (AWI, ULB). Different algivorous and bacterivorous grazers will be added to study their impact on phytoplankton population dynamics, and bacterial dynamics and biomass.

Task 2.3 DMSP production and its transformation in DMS

DMSP lyase activity will be determined to establish its consumption rate (ULB). Concentrations of DMS and dissolved and particulate DMSP in samples collected during field investigations will be measured using gas chromatography equipped with a flame photoionization detector (ULB). The analysis of DMS and DMSP will be subcontracted by ULB. Air-sea DMS fluxes in the Gulf of Biscay will be evaluated by ULg. E. huxleyi culture experiments (with natural bacterial inocula) will be set up to study DMSP lyase activity. We will use axenic vs. bacterised cultures to differentiate between algal and bacterial DMSP cleavage. Bacterial community biomass (counts) and composition (DGGE) will be assessed (UGent).

Task 2.4 Influence of TEP and DMS production on rain ratio PIC/POC

The enzymatic cleavage of DMSP into DMS releases one acrylic acid. The potential acidification that could result from this transformation will be examined under experimental conditions. We will further test the hypothesis that this acidification could be buffered by CaCO₃ dissolution in microenvironments provided by TEP formation by following the evolution during incubations of enzymatic activity (ULB) and Ca2+ concentration, DIC and TA (ULg).

WP3 IMPACTS OF OCEAN ACIDIFICATION ON COCCOLITHOPHORID METABOLISM AND TEP PRODUCTION

The increase of atmospheric pCO₂ will affect ocean biogeochemistry through the invasion of CO₂ in surface waters and directly related decrease in pH. This ocean acidification can potentially stimulate the primary production of certain phytoplanktonic species and reduce the calcification rates of calcifying organisms. Both of these processes are negative feedbacks on the increase of atmospheric CO₂. Using phytoplankton from the central Baltic Sea incubated at different pH values, it has been shown that the production of TEP under nutrient depleted conditions was related to pH. In a more recent study with the coccolithophore E. huxleyi grown at three different pH values, it has been observed that the TEP concentration per cell increased significantly at lower pH than at present day or elevated pH. Although these studies suggest that pH affects carbon partitioning, our knowledge is still insufficient to evaluate the importance of these processes for the future marine carbon cycle. Rates of PCHO exudation and aggregation as a function of pH and in terms of carbon fluxes still need to be quantified. The relationship between PCHO exudation, nutrient concentration and pH also need to be parameterised, if we want to identify where or when these processes could significantly affect carbon partitioning in the ocean. In addition, we need to trace the fate of pH dependent POC production to assess whether these processes can significantly affect the biological pumping of carbon to the deeper ocean and eventually provide a sink for anthropogenic CO₂ on longer timescales. Finally, acidification stimulates in diatoms and coccolithophores the production of DMSP although the potential of DMS to enhance CaCO₃ dissolution has not yet been investigated.

Task 3.1 Influence of pCO₂/pH on calcification by selected coccolithophorid species using batch culture experiments

To assess the effect of increased temperature and acidity related to the future rise in atmospheric CO₂ on phytoplankton, primary production, calcification, and DMSP and DMS production, controlled (pCO₂/pH, temperature) laboratory (batch culture) experiments will be conducted on E. huxleyi and other selected species. ULB and ULg will perform and compare culture under crossed conditions of pCO₂ and temperature at an irradiance of 400 µmol photons m-2 s-1. Adaptation to oceanic acidification will be performed on batch cultures at pCO₂ of 180, 380 and 750 ppm (corresponding respectively to glacial, present, and year 2100 atmospheric CO₂ concentrations) by bubbling gases at fixed CO₂ concentration, and crossed increase of temperature of 3° corresponding to the predicted sea surface temperature increase by year 2100 in the Gulf of Biscay. We will also study the influence on calcification of saturation state with respect to calcite by modifying the CO32- or Ca2+ concentration in the batch cultures. Two batch culture experiments are planned per year at ULB for three years with each lasting one month. UGent will use the microscopic and molecular techniques described in Tasks 1.1 to monitor changes in community composition of prokaryotic and algal assemblages in relation to various temperature and pCO₂ regimes. SEM micrographs of the coccolithophores will also be taken (UGent). During the the exponential growth phase of the experiment, production of organic and inorganic carbon will be measured by incubating an aliquot of the culture using 14C (ULB). In addition, evolution in the batch cultures of nutrient, PIC and POC concentrations will be followed (ULB), and concentrations of TA and Ca2+ will be also determined (ULg). Samples will be taken by ULB for DMSP and DMS production measurements, especially during the exponential phase of the growth. DMSP lyase activity will also be determined.

Task 3.2 Influence of pCO₂/pH on TEP dynamics and DMS cycling by selected coccolithophorid species using chemostat experiments

We will describe the kinetics of polysaccharide exudation and the subsequent TEP production by coccolithophores in relation to CO₂ concentration in a series of chemostat studies. Chemostats are culturing systems that allow balanced growth of phytoplankton to be achieved and controlled by adjusting the flow of growth medium through the system. At equilibrium, the cell growth rate, supported by inflow of supply medium, matches the removal of cells in the outflow. Cell abundance in the chemostat is regulated by adjusting the concentration of nutrients in the supply medium. Because chemostats allow for separate control of growth rate and abundance, they are ideally suited to study the effects of pCO₂/pH on the production of algal exudates and on their aggregation. Experiments will be performed with coccolithophorid cultures where DOM production rates will be compared at various pCO₂ and pH values. Chemostat experiments will be set up by AWI, ULg and ULB, corresponding to five different pCO₂ levels: 1) ~180 ppm (glacial), 2) ~280 ppm (pre-industrial), 3) ~380 ppm (present), 4) ~750 ppm (year 2100 in the IPCCC business as usual scenario IS92a), and 5) ~ 1500 ppm (year 2300 in the IPCCC IS92a scenario). The experiments will be conducted using natural seawater filtered through a set of filter catridges to remove a potentially high background concentration of HMW-DOM. Each chemostat incubator, will be filled and continuously supplied with seawater from a reservoir containing the nutrients and equilibrated with the desired pCO₂. The medium inside the chemostat incubator is stirred using a magnetic stirrer bar mounted 1cm above the bottom of the incubator to prevent cell damage. CO₂ concentration in each reservoir, containing input solution to the chemostat incubator, will be regulated using a pH-stat system. The desired pCO₂ concentration and pH can be achieved at constant temperature, salinity and TA by control of the pH. For higher-than-present CO₂ concentrations, the reservoir will be aerated with pure CO₂ until the desired pH value is reached. For present CO₂ concentration experiments, the water will be bubbled with air pumped into the reservoir by a compressor. For less-than-present CO₂ treatments, compressed air will be passed through a CO₂-adsorber (Na2CO3 platelets) first. All experiments will be performed in a temperature controlled room, at an irradiance of ~400 µmol photons m-2 s-1 using an alternating light dark cycle. Two chemostat culture experiments are planned per year at AWI for three years with each lasting one month. Excess material collected in the receiving bottles of the chemostats will be used to examine the influence of TEP on the formation, composition and sinking velocity of coccolithophorid aggregates using roller tables. These experiments will help determine whether there are any effects of pCO₂/pH on chemical composition of particles that might influence their stickiness, and hence tendency to aggregate and eventually sink (AWI). Algal growth and characteristics including SEM examinations (UGent), PIC and POC concentrations (ULB), TA (ULg), DMSP lyase activity and DMS production (ULB), nutrient concentrations (ULB), bacterial community structure and diversity (UGent) will also be determined.

WP4 MATHEMATICAL MODELING OF COCCOLITHOPHORID DYNAMICS AND THEIR IMPACT ON THE CARBON CHEMISTRY AND CLIMATE REGULATION AND CHANGE

Despite the spatial extent and the biogeochemical significance of coccolithophorid blooms, few models have been developed to describe their dynamics. Based on existing mesocosm experiments reproducing a bloom of E. huxleyi and on experiments to be performed in the framework of this project, we will develop and parameterise a mechanistic model of coccolithophorid dynamics. This will allow us to test the wide range of hypotheses on the environmental conditions that regulate the onset, development and decline of coccolithophorid blooms. Once fitted, this mathematical model will be extended in order to represent the ecosystem of the PEACE study site and it will be coupled with a physical model of the region. The coupled model will be used to simulate the succession of phytoplankton blooms in spring and to determine environmental conditions that promote the growth of coccolithophores.

Task 4.1 Development and parameterisation of a mechanistic model to simulate the coccolithophorid bloom reproduced in mesocosm experiments

A mesocosm experiment was carried out between 31 May and 25 June 2001 at the Marine Biological Field Station of the University of Bergen. The aim was to follow the development and decline of a bloom of a natural plankton community dominated by E. huxleyi exposed to various pCO₂ conditions (i.e. simulating the "glacial", "present", and "year 2100") in order to investigate the pCO₂ related effects on calcification at the community level. Chemical (e.g. TA, pCO₂, DIC, nutrients, O2, pH, PIC), biological (e.g. algal cell density, bacterial abundance, 14C primary production, Chl-a, net primary production, DOC, particulate organic matter) and physical (e.g. temperature, salinity, light intensity) variables and fluxes were monitored. Prof. Riebesell, coordinator of the experiment, has agreed to make these data-sets available. This experiment offers an optimal way to develop and parameterise a mathematical model due to the large amount of data available for model validation and calibration. Existing models will be tested and adapted in order to reproduce the experiments. In particular, a carbon submodel will be elaborated to describe the DIC chemistry (pH, TA, DIC and pCO₂).

Task 4.2 Development and parameterisation of a coupled physical-ecosystem for the PEACE study site

We will use a 1D General Ocean Turbulence Closure model (GOTM model) to simulate the vertical structure of the water column (temperature and salinity fields) and its seasonal modifications in response to the variability of atmospheric forcings (ERA-40 ECMWF). The mathematical model of coccolithophorid and DIC dynamics developed in Task 4.1 will be extended in order to obtain an ecosystem model of the region. This model will then be coupled with the GOTM model and the resulting coupled model will be validated using data collected in the framework of the present project but also from available data-sets (OMEX, CCCC). The model will be used to investigate which environmental conditions that regulate the onset, development and decline of coccolithophorid blooms in the PEACE study site and to estimate the rates of organic and inorganic carbon production, degradation and export, and air-sea exchange of CO₂.

WP5 PROJECT MANAGEMENT AND COORDINATION

An efficient project management and coordination maximises its integration and outputs. The objectives of WP5 are 1) to organise the general project planning and the coordination among the various work packages, 2) to establish contacts with other relevant national and international programmes, 3) to ensure data and information exchange among the partners and other national and international scientific communities, and 4) to valorise and disseminate the project outputs.

Task 5.1 Internet site

An internet site will be set up by ULg at the beginning of the project and maintained on a monthly basis, to provide general and detailed information to both the public and scientific community. All partners will supply material to ULg concerning their contribution to the project.

Task 5.2 Database

Data generated by the project from field and laboratory studies will be made available and regularly updated in a database posted on the internet site (Task 5.1) in a restricted area available to the network members (including the follow-up committee members) and PPS Science Policy. This will allow the partners to exploit at maximum the project outputs. At the end of the project, the data and metadata will be posted in various public and international databases. The database in the PEACE web site will also be made public.

Task 5.3 Cruise preparation

ULB, in consultation with all partners, will be responsible for the cruise planning. Foreign scientists involved in similar studies in the same area will be invited to participate in the cruise. We will attempt to combine our cruises with those of other projects in the same area, in order to optimise the usage of shiptime. Three cruises, one per year starting project year 1, are envisaged in the spring to early summer period when coccolithophorid blooms are observed in the Gulf of Biscay. This will allow us to finalise the analyses of all field and laboratory samples during the early part of project year 4, so that partners can concentrate on the data interpretation and publication in peer-reviewed journals. In addition, the newly acquired data-sets will be rapidly available for validation of the biogeochemical model.

Task 5.4 Organisation of annual workshops and follow-up committee meetings and preparation of annual science reports

During the first three project years, annual workshops will be organised by ULB to discuss the results obtained by the partners, and members of the follow-up committee will be convened once a year. Whenever possible, the annual workshops will be combined with the follow-up committee meetings. Since the last year of the project will be devoted entirely to data processing and interpretation, two workshops will be held, in addition to the follow-up committee meetings. Relevant external participants, national and/or international, can be invited to stimulate the exchange of knowledge on the same subjects. Annual science reports integrating the results from individual partners will be assembled by ULB and submitted to PPS Science Policy.

Task 5.5 Publications

The results obtained by the project will be published in joint papers in international peer-reviewed journals. This valorises and disseminates the outputs of the project.