Ecologie larvaire de l'huître perlière (Pinctada margaritifera, L.), croissance et dispersion dans un lagon d'atoll polynésien

Understanding key factors determining the bentho-pelagic species larval growth and dispersal remains one of the challenges of marine ecology, notably to understand the adult population structure and dynamics. In the present study, we explore this determinism on the development and dispersal of the black-lip pearl oyster (Pinctada margaritifera var. cumingii, Linné 1758) larvae in a French Polynesian atoll lagoon (Ahe, Tuamotu archipelago). Rearing of the pearl oyster for the production of black pearls takes on a major social and economic place in French Polynesia. However, the supply of pearl oyster seed to the farms depends entirely on the natural collection of wild spat of the only one specie P. margaritifera, and spatial and temporal variability of spat yield on collectors is very high. Needs cannot always be locally fulfilled. In this context, the objective of the present thesis is to analyze the relationship existing between the environment and the growth and dispersal of the pearl oyster larvae, in order to bring understanding elements and tools allowing the improvement of the collecting strategies. This work is based on an in situ study, which describes environmental heterogeneity and brings a vision of the larvae dispersal dynamics in the Ahe lagoon. In parallel, we developed a more explanatory approach of the links between the environment and the larvae, by an experimental approach and the application of deterministic models of growth and dispersal. Ahe is a relatively rich atoll lagoon, with higher picophytoplankton concentrations than the average of Polynesian atolls, and very high bivalve larvae abundances throughout the year. The study of environmental heterogeneity points out the supremacy of small spatial (vertical) and temporal (weekly) scales variations, compared with large scales (lagoonal and seasonal). The short-term variations of the weather conditions, mainly of the wind, explain most of this variability. In addition, we could highlight the combined influence of physical factors such as the hydrodynamics, and biological such as the access to the trophic resource or the larval swimming behavior, on the determinism and the maintenance of observed structures, as much for the phytoplanktonic communities than for the bivalve larvae. The development and application of a bioenergetic growth model, based on the dynamic energy budget theory (DEB), enable us to identify the sensitivity of the pearl oyster larvae to the space-time variations of the trophic resource, measured in situ. The Ahe lagoon environment appears thus potentially limiting for the P. margaritifera larvae growth. An average of 3 % of the larvae would reach the pedi-veligerian stage after 30 days of development. The validation of a dispersal model, integrating an empirical model of the larval vertical migration, allowed outlining the larval dispersion, under the combined effect of hydrodynamics and larval behavior. In spite of the important dispersal capability of the larvae at the lagoon scale, the wind orientation seems to be a major element determining the concentration sectors, and in fine the preferential area of settlement. The seasonal variability of the larval dispersal capability would be thus prevalent at the Ahe lagoon scale. These results allow identifying key sectors for the collecting practices, and the opportunity to develop a sanctuary area, providing a great quantity of larvae in the farmers collecting sector.