No abstract

A numerical model (MARS-2D) was developed, with the aim of describing the hydrodynamics that prevail in Arcachon Bay. Direct model results as well as derived mixing and transport timescales (tidal prism, local and integrated flushing times, age of water masses), were used to understand the behaviour of water masses and exchanges between the Bay and its frontiers. Particular attention was paid to the processes that drive the hydrodynamics (tides, wind and rivers), in order to understand their respective influence. The Arcachon Bay hydrodynamic system appears primarily to be highly influenced by tides; secondarily, by winds. About two third of the lagoon total volume is flushed in and out at each tidal cycle, which represent a mean tidal prism of 384 millions of cubic meters. The percentage of seawater flushed out during the ebb, that returns into the lagoon during the following flood flow is very high (return flow factor=0.95). This pattern leads to calculated integrated flushing times (IFT) ranging from 12.8 to 15.9 days, respectively, for the winter 2001 and summer 2005 simulations (two contrasting climatological situations: in summer, light northwesterly winds and low discharges in the rivers and, in winter, stronger southwesterly winds and higher river flows). Moreover, it has been found that northerly and westerly winds tend to reduce the flushing time, whilst southerly and easterly winds tend to hinder the renewal of the water in the Bay. The behaviour of the waters originating from the two main rivers of the lagoon, was studied also by means of the mean age assessment, under varying conditions of river flow and wind regime.

The Soil and Water Assessment Tool (SWAT model, 2001 Version) has been applied to the Thau lagoon catchment area in order to simulate water discharges and nutrient inputs into the lagoon over a 10 years period (1989-1999), and to provide routing inflows to a previously developed lagoon ecosystem model. The watershed model has been calibrated and validated using measured data available for the two main rivers. The results indicate that the mean annual nitrogen inputs into the Thau lagoon is 117 ± 57 tons y−1, with the two main rivers, contributing for 80% of total annual nitrogen export. The variations of outputs to the lagoon are nonetheless important from 1 year to another. Due to the local agricultural practices and a reduced in-stream natural depuration, point sources seem to be the main factor affecting the fresh water quality. The coupling with the lagoon model allowed to estimate the impact of those terrestrial inputs on the lagoon nitrogen cycling and primary productivity. Influence of river discharges makes itself felt essentially near the river outlets. The northern bordure of the lagoon is then characterised by highly variable dissolved inorganic nitrogen concentrations, especially during flood events, while more stable and lower concentrations were simulated in the southern part of the lagoon. Simulated chlorophyll a ranged 1-15 μg l−1, with maximums in March. Mean annual phytoplankton production was 364 ± 142 gC m−2. The simulations showed that maximum annual productions are due to macrophytes (up to 1300 gC m−2 y−1), but at the whole lagoon scale, annual phytoplankton production resulted greater. From our results it also appeared that the greatest part of primary producers nitrogen requirements is satisfied by nutrient regeneration within the lagoon.

The primary production and the respiration of Zostera noltii beds in the Thau lagoon were studied by means of the benthic bell jar technique. Concurrently, environmental data (temperature, light and nutrients) as well as morphological data of seagrass meadows (leaf width and height, density of shoots, above/below-ground biomass ratio) were collected with the purpose of explaining most of the observed variations in metabolism. Seagrass plus epiphyte respiration rates were influenced mainly by the water temperature, showing a typical exponential response to an increase in temperature. Surprisingly, measurements of production rates were not related to incoming light intensities recorded at the seagrass canopy level. An equation frequently used for terrestrial standing crops, involving the leaf area index (LAI) and the characteristics of the canopy architecture (parameter K, depending on leaves optical and geometrical properties), was applied to the seagrass ecosystem in order to estimate the light energy actually available for the plants, i.e. the light intercepted by the seagrass canopy (Q(abs)). Linear relationships were then validated between gross production rates and calculated Q(abs) for Z. noltii beds, and the best fits were obtained with K values nearing 0.6, confirming the similarities between terrestrial graminaceae and seagrasses. A linear regression model for primary production is proposed, involving the calculated Q(abs), the water temperature and the leaf nutrient content.

The origin and composition of sediment organic matter (SOM) were investigated together with its spatial distribution in the Arcachon Bay-a macrotidal lagoon that shelters the largest Zostera noltii meadow in Europe-using elemental and isotopic ratios. Subtidal and intertidal sediments and primary producers were both sampled in April 2009. Their elemental and isotopic compositions were assessed. Relative contributions of each source to SOM were estimated using a mixing model. The SOM composition tended to be homogeneous over the whole ecosystem and reflected the high diversity of primary producers in this system. On average, SOM was composed of 25% of decayed phanerogams, 19% of microphytobenthos, 20% of phytoplankton, 19% of river SPOM and 17% of macroalgae. There was no evidence of anthropogenic N-sources and SOM was mainly of autochthonous origin. None of the tested environmental parameters-salinity, current speed, emersion, granulometry and chlorophyll a-nor a combination of them explained the low spatial variability of SOM composition and characteristics. Resuspension, mixing and redistribution of the different particulate organic matters by wind-induced and tidal currents in combination with shallow depth probably explain the observed homogeneity at the whole bay scale. Highlights ► SOM composition reflected the diversity of primary producers and POM sources of the system. ► SOM is mainly of autochthonous origin. ► There is no influence of anthropogenic N-sources in Arcachon Bay. ► SOM composition is homogeneous at the bay scale. ► None of the tested environmental forcings explained the spatial variability of SOM composition and characteristics.