Along the French Mediterranean coast, a complex of eight lagoons underwent intensive eutrophication over four decades, mainly related to nutrient over-enrichment from continuous sewage discharges. The lagoon complex displayed a wide trophic gradient from mesotrophy to hypertrophy and primary production was dominated by phytoplankton communities. In 2005, the implementation of an 11 km offshore outfall system diverted the treated sewage effluents leading to a drastic reduction of anthropogenic inputs of nitrogen and phosphorus into the lagoons. Time series data have been examined from 2000 to 2013 for physical, chemical and biological (phytoplankton) variables of the water column during the summer period. Since 2006, total nitrogen and phosphorus concentrations as well as chlorophyll biomass strongly decreased revealing an improvement in lagoon water quality. In summertime, the decline in phytoplankton biomass was accompanied by shifts in community structure and composition that could be explained by adopting a functional approach by considering the common functional traits of the main algal groups. These phytoplankton communities were dominated by functional groups of small-sized and fast-growing algae (diatoms, cryptophytes and green algae). The trajectories of summer phytoplankton communities displayed a complex response to changing nutrient loads over time. While diatoms were the major group in 2006 in all the lagoons, the summer phytoplankton composition in hypertrophic lagoons has shifted towards green algae, which are particularly well adapted to summertime conditions. All lagoons showed increasing proportion and occurrence of peridinin-rich dinophytes over time, probably related to their capacity for mixotrophy. The diversity patterns were marked by a strong variability in eutrophic and hypertrophic lagoons whereas phytoplankton community structure reached the highest diversity and stability in mesotrophic lagoons. We observe that during the re-oligotrophication process in coastal lagoons, phytoplankton shows complex trajectories with similarities with those observed in freshwater lake systems.

Carbon stable isotopes and fatty acids were measured in the suspended particulate organic matter (POM) of the Thau lagoon to study its qualitative temporal changes in relation to environmental factors and to identify the food sources of bivalves over a one-yr-cycle in relation to their growth. Reciprocally, the impact of shellfish farming on POM was also studied. Oysters and mussels were sampled and measured for biometry, stable isotopes and fatty acid composition. Water samples were collected at two sites, both inside and outside of the shellfish farming area, to determine concentrations in POM, chlorophyll a (Chl a) and stable isotopes. Carbon isotopes and fatty acids in bivalves reflected seasonal changes in food sources, which varied consistently with the environment. Seasonal changes in δ 13 Highlights ► Carbon isotopes and fatty acids in bivalves reflected seasonal dietary changes. ► Terrestrial organic matter and bacteria can contribute to the diet of bivalves. ► Mussels and oysters showed significant differences in their fatty acids. ► Periods of high growth rate in bivalves were mainly fuelled by diatoms. ► Stable isotopes allow investigating the effects of bivalve aquaculture.

The decrease in water nutrients confirms oligotrophication in the period 1998-2016. ► The decrease in phytoplankton biomass was associated with a decrease in diatom abundance. ► The dominant taxa shifted from Skeletonema-Chaetoceros to Chaetoceros-Pseudo-nitzschia. ► The median proportion of dinoflagellates relative to diatoms increased significantly. ► The total dinoflagellate abundance did not change significantly.

The consumption and assimilation rates of the woodlouse Armadillidium vulgare were measured on leaf litters from five herb species grown and naturally senesced at 350 and 700 µl l CO. Each type of litter was tested separately after 12, 30 and 45 days of decomposition at 18°C. The effects of elevated CO differed depending on the plant species. In Medicago minima (Fabaceae), the CO treatment had no significant effect on consumption and assimilation. In Tyrimnus leucographus (Asteraceae), the CO treatment had no significant effect on consumption, but the elevated CO litter was assimilated at a lower rate than the ambient CO litter after 30 days of decomposition. In the three other species, Galactites tomentosa (Asteraceae), Trifolium angustifolium (Fabaceae) and Lolium rigidum (Poaceae), the elevated CO litter was consumed and/or assimilated at a higher rate than the ambient CO litter. Examination of the nitrogen contents in these three species of litter did not support the hypothesis of compensatory feeding, i.e. an increase in woodlouse consumption to compensate for low nitrogen content of the food. Rather, the results suggest that in herbs that were unpalatable at the start of the experiment (Galactites, Trifolium and Lolium), more of the the litter produced at 700 µl l CO was consumed than of that produced at 350 µl l because inhibitory factors were eliminated faster during decomposition.

Highlights ► Significant reductions in water nutrient concentrations due to decreasing urban nutrient loading. ► Evidence for rapid recovery (1-3 y) of hypertrophic & eutrophic lagoons found in the water column. ► Almost complete recovery observed for the eutrophic lagoon, highlighting no hysteresis. ► The recovery patterns depend on the eutrophication status before remediation. ► Feedback responses suggest ecosystem regimes can become unstable.

The aim of this study was to distinguish between sources of the complex variety of Marennes-Oléron Bay suspended particulate organic matter (SPOM) contributing to the tropho-dynamics of the Marennes-Oléron oyster farming bay. Basic biomarkers (Chl a, C/N and POC/Chl a ratios), carbon and nitrogen stable isotopes from SPOM were analyzed and the microalgae community was characterized. The sampling strategy was bimonthly from March 2002 to December 2003; samples were taken from an intertidal mudflat. Four main sources contributed to the SPOM pool: terrigenous input from rivers, neritic phytoplankton, resuspended microphytobenthos and periodic inputs from intertidal Zostera noltii meadows. Seasonal fluctuations were observed in both years of the study period: (1) SPOM collected in the spring of 2002 (δ 13 C = −25‰ to −23‰) was mainly composed of fresh estuarine inputs; (2) SPOM from the summer and fall of 2002 and 2003 was predominantly neritic phytoplankton (δ 13 C = − 22‰ to −19‰); (3) SPOM from the winter of 2002, spring of 2003 and winter of 2003 (δ 13 C = −21 to −23‰) was composed of a mixture of decayed terrigenous river inputs and pelagic phytoplankton, which was predominantly resuspended microphytobenthos. In the summer of 2003-the warmest summer on record in southern France and Europe-SPOM was particularly enriched for 13 C, with δ 13 C values ranging from −14‰ to −12‰. Pulses in δ 13 C values, indicative of 13 C-enriched decaying materials, extended into the fall. These were attributed to benthic intertidal inputs, including both resuspended microphytobenthos and Z. noltii detritus. Changes in SPOM sources in Marennes-Oléron Bay may lead to differences in the quality of the trophic environment available for reared oysters.

Increases in the intensity of disturbances in coastal lagoons can lead to shifts in vegetation from aquatic angiosperms to macroalgal or phytoplankton communities. Such abrupt and discontinuous responses are facilitated by instability in the equilibrium controlling the trajectory of the community response. We hypothesized that the shift in macrophyte populations is reversible, and that this reversibility is dependent on changes in the pressures exerted on the watershed and lagoon functioning. Biguglia lagoon (Mediterranean Sea, Corsica) is an interesting case study for the evaluation of long-term coastal lagoon ecosystem functioning and the trajectory of submerged macrophyte responses to disturbances, to facilitate the appropriate restoration of ecosystems. We used historical data for a two hundred-year period to assess changes in human activities on the watershed of the Biguglia lagoon. Macrophyte mapping (from 1970) and monitoring data for dynamics (from 1999) were used to investigate the trajectory of the community response. The changes observed in this watershed included a large number of hydrological developments affecting salinity and resulting in changes in macrophyte distribution. Nutrient inputs over the last 40 years have led to a shift in the aquatic vegetation from predominantly aquatic angiosperm community to macroalgae and phytoplankton in 2007 (dystrophic crisis). Changes in hydrological management and improvements in sewage treatment after 2007 led to a significant increase of aquatic angiosperms over a relatively short period of time (4-5 years), particularly for Ruppia cirrhosa and Stuckenia pectinata. There has been a significant resurgence of Najas marina, due to changes in salinity. The observed community shift suggests that Biguglia lagoon is resilient and that the transition may be reversible. The restored communities closely resemble those present before Please note that this is an author-produced PDF of an article accepted for publication following peer review. The definitive publisher-authenticated version is available on the publisher Web site.disturbance. These findings demonstrate the need to understand watershed exploitation and ecosystem variability in lagoon restoration.

Since the mid-20th century, Mediterranean lagoons have been affected by eutrophication, leading to significant changes in primary producers. In the early 2000s, management actions have been implemented to reduce nutrient inputs with the aim to achieve a good ecological status as requested by the EU water framework directive. As a result of these actions, a sharp decline in nutrient loads has been recorded in several lagoons leading to an oligotrophication of the water column. The analyses of a long-term data set (1998−2015) of 21 polyhaline and euhaline lagoons with contrasting trophic status allowed us to infer a general scheme for the changes in macrophyte assemblages during the oligotrophication process. Placing hypertrophic and oligotrophic conditions end to end, we inferred that the general pattern for the re-oligotrophication trajectory in Mediterranean coastal lagoons is described by the following sequence, with regime shifts between each state: (1) bare non-vegetated sediments, phytoplankton-dominated state; (2) opportunistic macroalgae; (3) seagrass and perennial macroalgae dominated state. However, we did not observe the latter regime shift for the most eutrophicated lagoons, which, so far, remained stuck in the opportunistic macroalgae state. So far, the shift from dominance of opportunistic macroalgae to a system dominated by seagrasses was only observed in a single lagoon where seagrasses had never completely disappeared, which possibly relates to resilience. More generally, the conditions favoring regime shifts from opportunistic macroalgae to seagrasses are still poorly understood. In conclusion, we describe a generic pattern for re-oligotrophication of Mediterranean coastal lagoons, although a full recovery from highly eutrophied to oligotrophic conditions may require more than a decade and may include conditions that remain so far poorly recognized.