L'objectif de CHACCRA est d'étudier les flux de carbone et de simuler leur changement lors des modifications du climat et de l'hydrologie au débouché du Rhône, fleuve majeur de la Méditerranée occidentale, et dans le Golfe du Lion, la zone côtière adjacente. CHACCRA est une composante de l'initiative internationale RiOMar (River-dominated Ocean Margins).

Task 1- Fluxes from the Rhône to the Mediterranean Sea: characterization of organic matter, nutrients and particles and transformation in the downstream region
Partners : CEREGE, COM, CEFREM, IRSN

In order to gain understanding of the Rhone-Mediterranean system, a precise determination of dissolved and particulate organic matter, nutrients and particle river inputs is now required especially during floods which are critical events for particle inputs. Furthermore, the dynamics of transformation (degradation, aggregation, settling, residence time) of these inputs between their freshwater end-member and the river outlet must be known to constrain properly the elemental inputs which drive biogeochemical processes acting in the plume and the sediment of the coastal zone (Tasks 2, 3). This implies both correct estimation of fluxes and a fine characterization of the particles. These are the two main objectives of this task, based on the following subtasks:

SubTask 1.1 quantification, characterization and temporal variations of particles, organic matter and nutrients transported by the Rhône:
The various fractions of N, P (dissolved and particulate, organic and inorganic) and organic carbon, silicate and biogenic silica concentrations will be measured once a day at Arles (40 km upstream of the river mouth) with high frequency sampling using an automatic station run in cooperation between Agence de l'Eau, IRSN and COM. Collection strategy is based on automatic sampler offering daily samples in routine. During sudden rises of river flux (higher than 3000m3.s-1), sampling frequency is increased to 6 samples per day to estimate precisely variations during rising and falling of the water flux. In the river water, most nitrogen is present as nitrate which can be directly used by biological productivity, but almost half of the phosphate is under particulate form, the largest part of which is calcium bound phosphate. Considering the importance of PO4 limitation, it is essential to specify the various phases of P in particulate matter, in order to determine the quantity of bio-available P transported by the Rhône.
All samples will be analyzed for: suspended matter concentrations; nitrate, nitrite, phosphate, silicate (automated colorimetry); total and particulate nitrogen, phosphorus and organic carbon (wet-oxidation and successive automated colorimetry cross-calibrated with HTCO) and various forms of particulate phosphorus (iron-bound P; Ca-bound P). Dissolved organic diacids and sugars will be monitored as the reactive fraction of carbon. A selection of samples will be also analyzed for grain-size (laser granulometer).
On a biweekly base (with additional sampling during and after flood event resulting in a number of 30-40 samples per year), biogenic silica concentrations and quantification plus determination of phytolith species by microscopy will be done. This will allow to determine the relative fraction of phytolith in biogenic silica as recent work has shown that phytoliths could constitute 90% of BSi of the Congo river (Conley et al, in press). On a monthly basis (12-15 samples per year), carbon and nitrogen stable isotopes (Micromass GVI spectrometer) and total and hydrosoluble sugars (TPTZ method) will be determined. With these analyses, temporal variations in "normal" and "flood" conditions will be defined and available for plume and benthic studies and as boundary conditions for modelling purposes.

Parameters measured. Normal regime: 1 measurement per day
dissolved : NO3-, NO2-,H3PO4, H4SiO4 by automated colorimetry
DOC by wet-oxidation and automated colorimetry cross-calibrated with HTCO
Dissolved organic diacids and sugars by HPLC
particulate : Particulate P (iron-bound P; Ca-bound P) by digestion and colorimetry
POC and PON by elemental analyser
BSi by digestion and colorimetry
Flood regime : if flux is >3000m3.s-1 : 6 measurements per day
dissolved : NO3-, NO2-,H3PO4, H4SiO4 by automated colorimetry
DOC by wet-oxidation and automated colorimetry
particulate : Particulate P (iron-bound P; Ca-bound P) by digestion and colorimetry
POC and PON by elemental analyser

SubTask 1.2: Transformations along the "salt-water wedge"
The objective of this task is to better constrain our knowledge of processes occuring in the river from the fresh end-member zone (Arles) to the sea. Indeed, river monitoring stations are generally situated far upstream from the river outlet which causes a difficulty for estiamting the river inputs to the costal one as downstream processes may modify both the quality and quantity of riverine material delivered to the coastal zone. In the case of the river Rhône, 40km of relatively flat flood plain (Camargue) lead the water from Arles to the Mediterranean, therefore creating a potential transformation of particulates and dissolved substances delivered to the coastal region. Various processes can occur downstream of Arles (degradation, disagregation or aggregation) which are able to change the nature of particles entering the coastal zone. The Rhone is a stratified " estuary " with a salt wedge introducing about 20km inland at low water discharge, and about 8 km at mean water discharge (1700 m3/s). Aggregation, flocculation and desorption processes associated with large changes in ionic strength occur in this zone and lead to rapid exchanges between particulate and dissolved phase. Thill et al. (2001) showed that the Rhone particles have a poor reactivity regarding salt induced flocculation, but they also demonstrated that small particles (2-5 µm) strangely increased seaward. Colloidal aggregation or primary production could explain such a trend but data are lacking to confirm these hypotheses. Furthermore, a significant fraction of particulate material may be transported in the river in a bed load (a thick turbid layer). The fraction of material transported through that pathway ir rarely assessed but is thought to be important for quantity (generally estimated to 10%) and especially for the quality of organic carbon delivered to the coastal zone.
Sampling will be performed seasonally during expeditions at various water discharges from Arles to the mouth using an inflatable boat (spring normal flux and summer low flux). Grain-size by portable particle counter (Pacific scientific) and Hiac-Royco granulometer, cation exchange capacity (acoustic zeta potential measurement, DT 300 Horiba), river current (ADCP) and some of the parameters listed above (SubTask 1) will be analysed on these samples. Some of these parameters will be measured at sea on selected samples during plume field campaigns in order to further the analysis of aggregation/disagregation at the end of the salinity gradient.
Bed-load fraction will be sampled for the first-time (Van Delft bottle) during these river cruises in order to determine the percent of solid discharge transported by such process as well as the nature of these particles. C/N and d13C will be performed on a selection of bedload samples in order to assess the origin of organic matter in this fraction.

SubTask 1.3 Long term variations of liquid discharge and nutrient concentrations and model scenarii.
The last objective of this Task is to produce a database of temporal variations of liquid discharge over the last two centuries including, when available, values of nutrients concentrations. With these past fluxes, impact of human influences on the coastal zone will be evaluated using the model developed on Task4. Particular attention will be paid to floods (origin and discharge) in this retrospective analysis.

Retour au menu

Task 2 - Role of the Rhone plume ecosystem in transforming and transferring particulate and dissolved organic and inorganic riverine input to the benthic system. Field and laboratory experiments

Partners : LOBB Banyuls, IFREMER
The main objectives of this Task is 1/ the quantification of the fixation of nutrient delivered by the river to the coastal sea and of the flux exported from this plume production to the benthic ecosystem and 2/ the parameterization of processes to be included in the biogeochemical module of the coupled circulation-biogeochemical model. Benefiting from our experience of former projects in the Rhône plume (RhOfi, BiopRoPhi) during the Chantier PNEC-Méditerranée, we propose to perform specific sampling strategies accounting for the interactions between hydrodynamics and biogeochemistry in order to set up a consistent data base to fulfil models requirements (Task 4) and to get new insights on the processes involved in downward production transfer. On the Rhone system, few exhaustive and accurate data sets are concerned with the downward transfer of particulate organic matter to the sediments. In order to estimate the transformations in dissolved and particulate compounds transferred to the benthic system (Task 3), a particular attention will be paid to the microbial loop as its functioning is in the timescale of those of the hydrodynamics of the plume (Naudin et al., 2005; Joux et al., 2005; Pujo-Pay et al., 2006).
Field campaigns are not well designed to sample extreme events (availability of ship, work at sea). Consequently another part of this proposal introduces laboratory experiments under controlled conditions in order to estimate the influence of microbial diversity and function, as well as POM and DOM quantity and quality (terrigenous or biogenic origin) in organic matter transformation and transfer from the plume to the benthic system. Experiments will mimic either storms or floods conditions or modifications induced by climate changes.

SubTask 2.1 : Nutrient fixation, downward flux of organic matter and the role of the microbial loop (field campaigns).
The main objective of this subtask is the quantification of the fixation of nutrient in the river plume and of the flux exported from this plume production to the benthic ecosystem during field campaigns. To this purpose, we will perform particulate and dissolved C, N and P field measurement on a grid of 24 stations presented in the upper graph and a Lagrangian experiment to follow the plume and its biogeochemical transformations. During each cruise, estimation of the downward flux will be measured with specific devices (drifting traps, SWASS; Naudin et al., 2001) according to a lagrangian strategy. Both sampling inside (0.2 and 1.5m depth) and under the plume will be considered. Traps will drift at a depth out of the dilution influence and above the bottom boundary layer. Lagrangian sampling allows a single synoptic 2D view of the vertical transfer as the river plume spread over the shelf. In order to set up budgets all over the region of freshwater influence, quasi synoptic surveys (grid of 24 CTDs, 5 nautical miles spaced) of the hydrological (T, S, Fluorescence, turbidity), and dynamical structure (wind, ADCP) will be performed before each lagrangian sampling. The area surveyed will extend from 4°25 to 5°00 E and from. 43°18 to 43°00 N. Nutrient fixation in the plume will be investigated using N and P measurements and 14C and 15N production measurements which will provide rates of carbon and nutrient fixation during organic matter production. Concerning measurement frequencies, the proximal plume, characterised by a high velocity and a steep salinity gradient, will be differentiated from the distal plume where production processes dominate. Along each trajectory, the proximal plume will be sample hourly whereas the distal plume will be sample every 2 hours due to the decreasing velocity spreading. During each cruise, an exhaustive sampling will be performed at a day interval in the proximal and the distal plume and repeated twice for a better estimate.
Another focus will be given on bacterial activity as the timescales associated with bacteria is in the order of hours, similar to the transfer time of waters in the dilution plume. The colonization of organic matter particles seems to be an efficient strategy for bacterioplankton to enhance their activity (Simon et al., 2002). The significance of particle-attached microbial processes as key processes in the decomposition and downward flux of organic matter is largely influenced by the quantity and quality of particulate matter (Simon et al., 2002). In order to study the competition between phytoplankton and bacteria for key nutrients utilisation, bacterial production on both free-living and attached bacteria using 3H leucine will be compared to phytoplankton production. Furthermore, bacterial biomass will be estimated using flow cytometry and epifluorescence microscopy. Community structure will be followed by fingerprinting techniques based on 16S rDNA and rRNA (DGGE or CE-SSCP). The phylogenetic diversity of both fractions will be done by cloning and sequencing approach.
In order to estimate the settling flux of particles supplied by the Rhone and to characterise the modifications involved during the spreading of the plume (aggregation, colloids, particulate carbon), size spectra of particles and flocculates will be measured in situ using an in situ granulometer and video microscopy will be recorded in collaboration with IFREMER. Developments on quasi in-situ measurements of the settling velocity of particulate matter will be considered too as well as the changes in light limitation all along the dilution gradient. These measurements will directly fulfil models requirements.

Parameters measured on the array of 24 stations and lagrangian experiment for surface water
at 0.2, 1.5m and 10m depth covering the river plume and underlying seawater
during spring and fall campaigns:
- CTD, fluorescence, turbidity on water column profiles (Seabird probe)
- NO3-, NO2-,H3PO4, H4SiO4 by automated colorimetry
- POC, PON and POP by CHN analyser and wet oxidation procedures
- DOC and DON, DOP by high temperature catalytic oxidation (HTCO) and by wet oxidation respectively
- Primary production: 14C incorporation, 15N isotopic method, delayed fluorescence, photosynthetic parameters
- bacterial counts and production by flow cytometry, and [3H] leucine incorporation
- Particle-attached and free-living bacterial community structure : fingerprinting techniques based on 16S rDNA and rRNA (DGGE or CE-SSCP)and cloning and sequencing approach for phylogenetic approach.
- AOB and NOB activities will be measured by the use of specific inhibitors (Bianchi et al., 1999).
- Particle size: measured in situ using an in situ granulometer and videomicroscope (IFREMER participation), Spectroradiometer Trios.
Drifting sediment traps material will be analysed for POC, PON and POP, plus bacterial parameters.

SubTask 2.2: Effects of simulated changes in nutrients and terrestrial organic matter discharge upon microbial communities (Microcosm experiments).
The objective of this subtask is to investigate in vitro the possible alterations of the plume ecosystem in case of changes in river delivery. Several modification of the coastal environment may arise during the coming century due to climate or anthropogenic changes which might affect the pelagic ecosystem : increased particle delivery due to floods, change in nutrient input or their elemental ratio due to changing human practices on land, increased water stratification due to summer warming, … These changes could promote alterations of the prokaryotic community structure which could shift microbial diversity from free-living to particle-attached bacteria, if particle load changes. These changes in river delivery could also promote modifications of phytoplankton assemblages. Indeed, phytoplankton in the Rhone plume is characterized by large gradients in composition and activity, from highly productive communities dominated by large cells (mainly diatom) at mid salinities to low productive communities dominated by small cells (Synechococcus, picoeukaryotes) at high salinities (Joux et al. 2005). It can be therefore hypothesized that a modification in the nutrients concentrations or ratio delivered by the Rhone River will impact differently these phytoplankton communities in terms of activity and composition changes.
The microcosm experiments will start after the spring coupled (tasks 2 and 3) campaign in May 2008 from seawater collected at locations representative of the functioning of the plume and of the downward transfer. These experiments will last for one year (May 2009) to test the following protocols. The colonization of various amount of suspended particulate matter from terrigenous or biogenic origin by bacterial communities from the Rhone river plume will be assessed by batch culture incubations. The experiments will be designed to evaluate the relation between changes in particle-attached and free-living bacterial diversity and mineralization activity with changes in particulate and dissolved organic matter quantity and quality.
A set of nutrient enrichment experiments (PO4, NO3+NH4, and N+P) will be performed using seawater collected at different locations in the River plume and by following activity (primary production, exudation) and the structure of the communities during incubation. Primary production will be measured by different technique (14C incorporation, delayed fluorescence, photosynthetic parameters). Exudation of DOM and regeneration fluxes will be measured by 14C and 15N isotopic method. Flow cytometry, microscopy and pigment analysis (spectrofluorimetry, HPLC) will be used to study structural changes. As for the field campaigns, bacterial biomass will be estimated using flow cytometry and epifluorescence microscopy. Community structure will be followed by fingerprinting techniques based on 16S rDNA and rRNA (DGGE or CE-SSCP). The phylogenetic diversity of both fractions will be done by cloning and sequencing approach.

Retour au menu

Task 3- Benthic remineralisation of terrestrial and marine organic matter and its temporal variation
Partners : LSCE, BIAF, LOBB, LGE, CEFREM

The objective of this task is to understand deposition, regeneration, resuspension and burial processes of terrestrial and marine organic matter in the river prodelta and the adjacent shelf in order to parameterize the benthic module of the coupled biogeochemical-circulation model. Flood input and its impact on the sediment recycling/burial is also an important component of the relation between rivers and coastal seas and will be specifically investigated using a moored benthic station. The ecosystem characteristics of the prodelta sediments (macrofauna, meiofauna) and their relationship with river inputs and sediment recycling will also be investigated as this ecosystem provide the basis of the whole food chain up to fish resources. Part of this ecosystem (benthic foraminifera, meiofauna) may fossilize and constitute archives of past environmental conditions. In order to understand spatial and temporal varibility of benthic regeneration/burial processes at the Rhône river outlet, we will use two different strategies:
- field campaigns and related measurements for spatial variability coupled to plume investigations on 16 benthic stations splitted in two sets : one in the river prodelta (close to the river mouth) and one in the distal zone related to plume production (see above map for locations)
- a moored benthic station in the river prodelta for event-related temporal variation.
These two approaches will be synthesized using early diagenesis models in relation with Task 4 dealing with biogeochemical modelling.
Task 3- Benthic remineralisation of terrestrial and marine organic matter and its temporal variation These two approaches will be synthesized using early diagenesis models in relation with Task 4 dealing with biogeochemical modelling.

SubTask 3.1: Understanding spatial variability of deposition, recycling and burial in relation with river inputs
A large body of parameters is needed in order to assess the budget for recycling/burial and the processes which determine the biogeochemical dynamics in marine sediments. This is due to the complex origin and reactivity of organic matter which requires special assessment in mixed terrestrial/marine environments and to the number of biogeochemical and transport processes that interplay in the sediment and lead to the mineralization of organic matter. Therefore, combined approaches using different type of parameters are generally used to constrain sediment recycling (solid phase and porewater distribution, flux measurements). The relation between organic matter mineralization and benthic fauna characteristics will be investigated on combined cores for assessing the functional relationship between sediment chemistry and living foraminifera, as they may constitute archive of past margin conditions; sediment imagery will be used to investigate spatial distribution of macrofauna.
Three main groups of parameters will be studied during these cruises:

SubTask 3.1a parameters related to the origin or reactivity of organic matter
Organic matter (OM) in shelf sediments is a mixture of different origins and reactivity. Several approaches need to be combined to decipher the mineralization/burial capacity of this OM mixture. Together with measurements of organic carbon and nitrogen in sediment particulate matter, the measurement of d13C and d15N by EA-IR/MS allows roughly to discriminate between the two possible biological sources of OM terrestrial and marine (Kerhervé et al., 2001). Indeed the d13C-OM of terrestrial component is -27‰ in this region whereas the marine d13C-OM is around -22‰. However, both sources vary with time and hydrological flux and need to be quantified precisely in relation with WP1 and WP2 before mixing models are performed to obtain fractions of each component in the organic matter. Nitrogen stable isotopes (d15N) of the OM also provides information on the organic matter source, as the riverine OM contains generally more 15N (high d15N) than the marine OM. The combined use of C and N stable isotopes should strengthen the discrimination between the two main sources of OM. ?14C of the organic matter is linked to its long-term reactivity (the period of 14C is 5730 y) with a clear signature of the recent material linked to the 14C released during the atmospheric bomb tests. Thus, ?14C-OM provides a proxy of its residence time since production and thus the overall reactivity of the component (Raymond and Bauer, 2001; Rabouille et al., 2002).
These isotopic measurements will be complemented by characterization of the sedimentary OM by main biochemical classes (lipids, proteins and carbohydrates) by colorimetric methods and estimation of the labile fraction (Chl-a pigments and amino-acids measured using HPLC). These analyses performed in the prodelta and on the shelf will allow the estimation of the fraction of labile organic matter in the sediment. Furthermore, enzymatic digestion will be used to qualify the lability of organic compounds (Kerhervé et al., 2001, 2002; Accornero et al., 2003; Grémare et al., 2005; Medernach et al., 2001).

SubTask 3.1b parameters allowing the quantification of organic matter deposition, recycling and burial using sediment profiles and core incubations
Organic matter recycling will be approached using vertical distributions of chemical compounds related to bacterial regeneration: oxygen, nutrients, metals (Fe and Mn), sulfide. Oxygen, pH and resistivity microprofiles will be measured using an in situ profiler (UNISENSE) operated by LSCE (Rabouille et al., 2003; Lansard et al., 2003; Dedieu et al., accepted), which has been previously deployed on the Rhone prodelta providing preliminary data for this region. In relation with these in situ profiles, microprofiles of Mn, Fe and sulfide will be acquired on board on collected cores using a combination of techniques: High resolution porewater samples will be acquired using Diffusion Equilibration in Thin films (DET) implanted in additional cores supplemented by polarographic laboratory profiler if threshold concentration are reached (Luther et al., 1998; Deflandre et al., 2002). Other cores will be subsampled in a nitrogen atmosphere in order to extract sediment porewaters by centrifugation. These porewater samples will be analyzed for nutrients, sulphate and metal using small samples measurement techniques (Metzger et al, in press). In parallel, core incubations will be performed to provide total exchange fluxes between sediment and water and thus integrated recycling rates. Replicate core incubations will be performed in the laboratory at bottom temperature and oxygen and nutrient fluxes will be monitored throughout time (Denis and Grenz, 2003; Grenz et al., 2003).
Recent sedimentation of particles will be measured using short term natural radioisotopes (7Be-53 days, 234Th-24 days). These short live radioisotopes will be very useful for tracing the heterogeneity of new particles deposition after floods. Furthermore, 210Pbxs and 137Cs distribution in key sediment cores will be measured in order to assess century scale sedimentation rate. These will be combined with organic carbon measurements to calculate organic matter burial (Charmasson et al., 1998; Radakovitch et al., 2003a,b; Miralles et al., 2005).

SubTask 3.1c relation of organic matter inputs with benthic infaunal ecosystems
Investigation of ecosystem structure will be assessed using two different techniques. Sediment Profiling Imagery (SPI) will be used to observe the vertical distribution and areal density of macrofauna in the sediment, which will be calibrated by macrofauna sorting from sediment cores (Rosenberg et al 2003a&b). A particular insight will be given to the flood particulate discharge in shaping the macrofaunal community. Live foraminifera (Rose Bengal stained) will be investigated from undisturbed sediment cores collected with a classical multicorer. The density, specific composition and the vertical distribution of the foraminiferal faunas will be determined in relation with geochemical parameters such as organic matter deposition, bottom water oxygenation and redox conditions within the sediment (Fontanier et al. 2002). Benthic foraminiferal faunas should strongly differ in relation to the gradients of terrestrial or marine organic inputs and redox conditions in the sediment from the Rhone prodelta. Preliminary results based on material collected in Minercot 2 (2005) suggest such faunistic gradients in the vicinity of the Rhone mouth. As most benthic foraminifera possess calcite/aragonite shells, they will provide excellent bio-indicators of environmental conditions from the past that are used in several instances to reconstruct environmental parameters (see reviews Gooday, 2003). Using the ecological foraminiferal data gathered on the Rhone prodelta and near shelf, attempts will be made to calibrate the complex signals related to river mouth regime : terrestrial organic matter discharge, marine primary production exports, seasonal benthic anoxia (Van der Zwaan and Jorissen, 1991).

Parameters measured on the array of 16 stations during the three cruises
(2 in the spring and 1 in the fall)
Particulate matter (on 2-10 levels depending on parameter)
- POC, PON by elemental analyser
-d13C and d15N of organic matter by EA-IR/MS
- ?14C of organic matter by AMS
- Chl-a pigments and amino-acids by HPLC
-7Be, 234Th, 210Pbxs and 137Cs by Gamma spectrometry
Porewater components (10-15 levels per cores)
- microprofiles of O2, pH, resistivity by autonomous in situ microprofiler
- mm-resolution H2S, Fe2+ and Mn2+ profiles using DET or polarographic onboard profiler
- cm-resolution for NO3-, H3PO4, H4SiO4 by automated colorimetry
Sulphate by HPLC
Fe2+ and Mn2+ by ICP-AES
Core incubations (2 cores per station)
- oxygen by micro-winkler titration
- NO3-, H3PO4, H4SiO4 by automated colorimetry
Faunal analysis
- SPI pictures (2 to 3 per station)
- macrofauna sorting (2 cores per station, one cruise except post flood samples)
- Live foraminifera counting (1 core per station, 10 levels, 1 cruise except post flood samples)

SubTask 3.2 : Temporal variation of recycling performed by a moored benthic station
The use of a benthic station allowing continuous measurements is essential for the observation of temporal variations of the sediment component under various forcings including floods or large storms which promote primary particle input or secondary transport. Floods are supposed to thought to deliver a large proportion of annual particulate inputs to the sea (primary input) whereas storms will rework, resuspend and transport deposited particles. These two processes will be studied using the benthic station. The originality of this project will be the development/adaptation and validation of a benthic station measuring parameters related to organic matter recycling in the sediment continuously over a long period of time. More specifically, this benthic station will host an in situ microprofiler similar to the one already deployed by LSCE, which will measure series of oxygen profiles over a period of 1-2 months. The total deployment time should be 1 year in order to capture key periods of benthic recycling (phytoplankton blooms, floods, storms-resuspension). The ship Antédon II or Tethys II will visit the mooring every two months and service the benthic station (Change batteries, change electrode, memory download, fooling protection reactivated). This benthic station will be equipped with additional sensors which will register environmental parameters (oxygen, turbidity, salinity, fluorescence). In a second stage, these sensors will be used to perform conditional sampling in order to save power and memory space. Conditional sampling will consist in triggering oxygen profiling when specific conditions are met: strong resuspension or particle input from the river, phytoplankton bloom deposition, decrease of oxygen in the water column, freshwater input. The development of this benthic station is already underway in a FP6 European programme named Coastal Ocean Benthic Observatory (COBO) in which LSCE is participant. The developments made in this European programme will be used during our project. Yet, the new design of the benthic station will require substantial improvements and alteration to fit the river mouth context. The station should be deployed around an existing mooring (Bouée de Roustan) once the Affaires Maritimes will have given their agreement. This location is chosen because of its location at 25-30 meters at the river outlet and to avoid trawling of the structure by fisherman who are quite numerous in this productive area.
In relation with the project CARMA (S. Charmasson and C. Estournel) an Acoustic Doppler Current Profiler (ADCP) will be deployed near the benthic station which will be used for acquiring current velocities, and particle load in the whole water column. These data will document the hydrological movement and sediment resuspension and will be coupled afterwards to the observation of recycling acquired by the moored microprofiler. They will also be used for validation of the coupled circulation-biogeochemical model (WP4).

SubTask 3.3 : Diagenetic Modelling:
Models of early diagenesis (recycling of chemical elements in the sediment column) will be used to quantify the processes in the sediment from the multiple set of observations on both the liquid and the solid phase. These models were developed 20 years ago and are now used by several groups (Rabouille and Gaillard, 1991; Rabouille et al., 2001 Soetaert et al., 1996). The model used within this project is the 1D-model from the NIOO (FEMME-Omexdia) which describes early diagenesis of organic carbon by bacterial use of oxygen and nitrate and lumps iron, manganese and sulphur cycles (Soetaert et al., 1996). The model will be calibrated using field data (concentration profiles in porewater/solids and flux at the sediment-water interface). This will allow calculation of mass balances of carbon recycling in the sediments. Time varying models will be implemented in order to estimate the dynamics following sudden inputs (from floods or bloom sedimentation) in relation with the time series gathered on the benthic station. The validation of this model will be performed in close collaboration with Task 4.

Retour au menu

SubTask 4.1. Development of a coupled physical-biogeochemical model
To address our objectives, we will implement a three-dimensional model coupling hydrodynamics, pelagic ecosystem model, dynamics of suspended particulate matter and sediment resuspension and diagenesis (figure 1). The Symphonie hydrodynamics model that has already been validated in the Gulf of Lions (Estournel et al. 2003) and on the characteristics of the Rhone plume (Estournel et al., 2001; Ulses et al., 2005) will be used during this project. It displays the following strengths: dynamics of the surface plume, current-wave turbulence within the bottom boundary layer for the water-sediment exchanges are fully represented. The resolution of the coupled model will be of the order of the kilometre on the whole Gulf of Lions and of several hundreds of meters for the nested model near the Rhone River mouth. The oceanic and meteorological forcings will be realistic and directly given by the MFS (Mediterranean Forecasting System) European research program. The meteorological model (ALADIN from Météo-France) will provide forcings at very high frequency (hourly tempo). The ECO3M-MED model currently developed at the LOB-Marseille will be used to represent the functioning of the pelagic ecosystem. It is a rather complex model involving description of several functional key planktonic groups, non-Redfield dynamics, biogeochemistry of chlorophyll, carbon, nitrogen, phosphate, silicic acid, dissolved and particulate organic forms. The choice of this model will be efficient when performing simulations in future situations in which ecosystem might shift towards new equilibrium, which can be described by complex process biogeochemical models but not simpler budget-oriented models. Specific features linked to the dynamics of suspended matter in these turbid environments (river plumes) will be implemented from some existing parameterizations: dynamics of the suspended matter supplied by the Rhone River, processes such as aggregation, flocculation, desegregation which result from the changes in salinity and turbulence fields, light limitation associated to the content in particulate matter within the water column. Early diagenetic processes will be based on the benthic model of Soetaert et al. (1996, 2001) tested in Task 3 on experimental results. The process of resuspension of the superficial sediments with waves and currents developed by Ulses (2005) will be coupled to the diagenesis model. This sediment transport model has been already validated using data of the EUROSTRATAFORM European research program and has provided a budget of non biogenic suspended matter over the whole Gulf of Lions.


Figure 1. Processes studied by task 4

SubTask 4.2. Calibration and validation of the coupled model
A special effort will be devoted to the validation of each module of the coupled model in relation with field based workpackages (Tasks 2&3). The effort will particularly deal with new biogeochemical processes which will arise from the field experiment dedicated to the study of the Rhone River plume and superficial sediments. This present step of validation will be facilitated because the cruise strategy has been closely elaborated between field experimenters and modellers. The parameters of biogeochemical and ecological functions used in the ecosystem model will be calibrated from data resulting from some batch cultures of functional planktonic groups performed at the laboratory (see Task 2). Moreover it will be checked whether the coupled model will correctly reproduce (1) the main mechanisms occurring from the River mouth and all along the dilution of freshwater and (2), the organic matter decomposition in the water columns and superficial sediments. The time-series performed in the River prodelta as well as the remote sensing tool through satellite imagery will be used to validate the coupled model under conditions that will not be observed during the field cruises. This will be especially the case during flood events that transport large amounts of particulate and dissolved matter to the marine domain and during strong storm events that release nutrients and organic matter to the water column that were sequestered in superficial sediments.

SubTask 4.3. Budgets of biogenic elements
Following validation steps, the coupled model will be used as a tool to provide budgets of biogenic elements supplied by the Rhone River at the scale of the Gulf of Lions. In this type of budgets, the use of the coupled model will enable to separately quantify and characterise spatially and temporally some key processes such as primary production, recycling and sequestration of organic matter. In the latter case for example, we will try to discriminate the different timescales for sequestration of organic matter in the sediments of the shelf: from the short-term in sediment which can be resuspended and transported (i.e. in the prodelta zone) to long-term sequestration. Indeed Ulses (2005) recently demonstrated that strong storms involve resuspension over the whole shelf and even in the canyons heads. This type of events may thus periodically supply large amounts of nutrients and organic matter to water column and transport sediment out of the shelf to the deeper Mediterranean basin. One of the major objectives of the present work will be thus to see whether this type of resuspension episodes has an actual impact on the annual budget of biological production. Numerical simulations will be carried out on the most recent years with the dataset available for the validation over this period. The interannual variability of the budgets that may depend on the occurrence of extreme events such as floods and storms will be thus assessed.

SubTask 4.4. Scenarios of the evolution of the pelago-benthic ecosystem associated to climate change and to the alteration of Rhone River inputs (anthropic impact)
We propose to study the sensitivity of carbon budgets and biogenic element cycling to the foreseeable climatic and anthropogenic changes. Several types of alterations will be tested such as the variations in the composition of the Rhone River supplies in inorganic (nitrate, silicic acid, phosphate) and organic compounds and in the freshwater flow in the context of climate change. A first step would be to test the responses of the plankton community structure to the historical decrease in phosphate observed during the last two decades and the concomitant evolution of the N to Si ratio in the Rhone River supplies. Some other scenarios of evolution of Rhone River flows for the 21st century given by regional climatic models (Météo-France) and of long-term changes in the nutrient composition (W. Ludwig, CEFREM) will also force the coupled model. The sensitivity of the carbon budget to climate change will be explored through simulations forced by the results of oceanic and atmospheric models of the Mediterranean Sea carried out by the climate group of Météo-France. We are already involved with Météo-France in a similar study at the scale of the northwestern Mediterranean which will give to our project an intermediate grid necessary for the downscaling. The scenarios performed by regional climate models for the next decades indicate drier summers but possible increase of cyclogenesis and thus precipitation during autumns generally responsible of the major floods. As a consequence of the increase of summer temperatures and associated reduction of precipitation, an increase of stratification and surface water temperatures as already observed during summers 2003 and 2006 is expected. It is also likely that destratification will shift later in the fall season and possibly early winter while dense water formation responsible of the exportation of matter produced on the continental shelf toward the deep ocean (especially the late fall bloom) will also be affected by climate warming and by changes in freshwater flow. The Symphonie circulation model is prepared to perform climate change scenarios but so far no calculation has been achieved on the changes of circulation imposed by a change in climate and freshwater discharge. An important part of the modelling effort will thus concentrate on the fair representation of these physical phenomena before the biogeochemical model is used to study the impact of climate change on marine ecosystems and biogeochemical cycles. The biogeochemical model ECO3M already represents several planktonic functional groups competing together which may provide a flexible model to address the issue of climate change impact on marine ecosystems and carbon cycling. The complexity introduced in this model with functional groups is necessary to be able to take into account the changes in the annual sequence of planktonic populations as a response to physical (temperature, stratification) and biogeochemical (rivers composition) constraints.

Retour au menu