Accueil > Objectifs scientifiques > Objective 3

Objective 3

lundi 12 mai 2014



The multidisciplinary field data gathered during the OUTPACE cruise will be available to the international community as soon as possible (CYBER data base) in order to help represent the interactions between climate, biogeochemical cycles and marine ecosystems. A multi-scale modelling approach is proposed from the beginning of the OUTPACE project. Its objective is to complement the observations with the analyzed data in order to understand the interactions between the biogeochemical cycles of the biogenic elements and the dynamics of the planktonic trophic network. Modelling will also be used in interaction with observation and measurements at several scales and at several moments of the project in order to optimize the potential of each of these tools.

Objective 3.1. Process scale studies

We first plan to focus on some physical and biogeochemical processes through (i) the zero-dimensional (0D) modelling of microcosm and through (ii) the simplified 1D vertical modelling of the physical and biogeochemical processes at the long duration stations. These stations will be positioned in regions where the horizontal variability of the hydrological and biogeochemical properties is supposed to be low over the duration of sampling making it possible to consider a 1DV modelling approach. Moreover, the majority of the model’s needs, in terms of state variables and parameters, will be acquired through in situ measurements or by controlled activities i.e. sampling and microcosms during the cruises.
Microcosm experiments and modelling (0D) are important for several reasons, the main one being that they provide a vision of the biogeochemical and trophic web functioning deconvoluted from hydrodynamic aspects. This is a crucial step towards the understanding of the ecosystem. Microcosms are also useful to acquire certain model parameters (characteristics of nutrient assimilation, photosynthetic parameters, respiration rates, mineralisation...) and to test or develop new mathematical formulations for processes where current formula are not validated or represented. It is also important that these microcosm experiments are undertaken during the cruises (or field studies) with in situ water samples in order to obtain parameters which are representative of the studied ecosystem.
In practice, a biogeochemical model relevant for SW Pacific waters and developed during the VAHINE project will be used to model the 0D experiments. It is based on a previous version of a biogeochemical model developed for the Mediterranean Sea which included 6 functional types, namely bacteria, two phytoplankton size-classes, HNF, ciliates and mesozooplankton and which was already implemented in the Eco3M modelling tool. In this model, organisms have flexible stoichiometry and are represented in terms of C, N, P biomasses as well as in abundances. During the VAHINE project, the diazotroph Trichodesmium spp. has been added and the introduction of other unicellular diazotrophs should be done in near future (Post Doc ANR JCJC VAHINE). This model should be further validated in the frame of the VAHINE project but it can already be used as an a priori model to help in designing the microcosm experiments that will be held on board. The measurements will in turn be useful to refine the a priori model.
In terms of processes, this model gives a mechanistic or semi-mechanistic representation of the key processes of plankton dynamics (photosynthesis, photoacclimation, nutrient absorption, DOM release, respiration, grazing, mineralization, excretion,...). The majority of these processes have been validated using microcosm and mesocosm experiments (Baklouti et al., 2006b, Baklouti et al., 2011, Alekseenko et al., subm.). Among the processes that specifically need to be worked on, the diazotrophy process is certainly the main one.
For Trichodesmium spp., the mechanistic model of Stephens et al. (2003) adapted to Trichodesmium spp. by Rabouille et al. (2006) has been used and adapted to our model. However, for unicellular diazotrophs, existing models for diazotrophy are scarce and rely on empirical laws deduced from macroscopic observations. In order to improve such models, the underlying mechanisms of diazotrophy need to be understood, especially the interaction between N2 fixation and nutrient uptake of other N sources, phosphate, and iron.

Objective 3.2. In situ experiments at long duration station (1DV)

The major objective of 1DV modelling is to represent fluxes of biogenic elements in the surface layer and provide an accurate budget on the fate of the organic C produced (Re-mineralisation, export). The use of a flexible stoichiometry model (non Redfieldian) will potentially make it possible to illustrate the decoupling of the biogenic element cycles in the oligotrophic ocean. The objective is to understand the production and fate of organic matter in the oligotrophic food web, and their role in the global C cycle. To achieve this, we will proceed to the extraction of those criteria which appear essential in providing a satisfactory representation of the marine pelagic ecosystem and biogenic element fluxes. The complexity of the trophic relationships will be simplified by focusing on the groups of species having similar biogeochemical functions (i.e. functional groups). The relevance of the 1DV modelling will depend on in situ measurements carried out at the long duration stations. Indeed, a certain number of parameters which will be measured are essential in achieving a realistic representation of the dynamics of the pelagic ecosystem and thus a precise estimate of the C budget. These parameters include all the parameters related to phytoplankton photosynthesis and to nutrient uptake by the plankton. In coupled physical-biogeochemical models, the parameterization of turbulence is frequently a source of error, because it is based on empirical laws, and because these laws are not then re-calibrated for each studied site (typical values found in the literature are used to parameterize these laws). Here, it will be possible to use realistic vertical profiles in the hydrodynamic 1DV model to accurately estimate the coefficients of turbulent vertical diffusion - Kz- (cf obj 2.1). The use of such realistic profiles is an essential step in representing the mixing processes in the surface layer and the nutrient fluxes at the base of the euphotic layer. However, a simplified 1D version could be achieved instead if the 1DV model is too difficult to parametrize. This simplified model would consider that the water column is divided in a number of boxes that would receive different light intensities according to the vertical profile of PAR, and which would exchange matter only through the sinking of detrital matter.
The biogeochemical model that will be used for the 1DV configuration is the same as the 0D one except that it will benefit from the refining that will have been done thanks to the microcosm experiments. This type of model enables us to obtain a representation of the biogeochemical cycles of the principal biogenic elements (C, N, P) as well as representing organisms ("functional" groups) that play a key role in the dynamics of organic C production and transfer. The whole data set collected from the long duration stations, and in particular the data obtained in a quasi continuous manner such as nitrate fluxes and stocks, will offer the possibility of validating the coupled model which up to this point has never been carried out due to the strong mismatch between experiments and modelling.

Specific questions: Is it possible to reproduce, with only one uniquely structured model, the principal physical and biogeochemical characteristics of the different oligotrophic areas studied? If so, the model should enable us to identify the origin of the processes and their different functions. In particular, this approach will enable us to answer an important question in the OUTPACE project: are the differences in functioning between the sites attributable to one or more production process of organic matter, remineralisation and export?