Since 2007, the ecosystem of the Gulf of Lions has shifted to a different regime, characterised by a low anchovy (Engraulis encrasicolus) and sardine (Sardina pilchardus) biomass and a remarkably high sprat (Sprattus sprattus) biomass. Surprisingly, the abundance and recruitment of anchovy and sardine remained high. To understand which processes (bottom-up or top-down control, etc.) could have caused this shift, we studied the changes in body condition, growth and size and age of anchovy, sardine and sprat over 1984-1985 and 1992-2012, using data from scientific surveys. The annual age structure of anchovy and sardine was estimated using Bayesian mixture models based on size frequency data with priors on the age-length relationship derived from independent otolith readings. The results indicated periods during which anchovy and sardine were in an average (1992-2004), good (2005-2007) or poor (2008-2012) overall state of condition. For sardine, the shift towards smaller fish observed during these past 4 years was explained by a combination of slower growth and the disappearance of older individuals (ages 2+). Despite the increase in biomass of sprat since 2008, indications were found that sprat was also smaller than in the past. As growth and condition decreased and overexploitation has not been documented or suspected for those three species in this area, we propose that the current decline in sardine and anchovy biomass could be due to qualitative and/or quantitative modifications in the planktonic production (i.e. a bottom-up control) or mass mortalities of adults due to an epidemic disease.

Around 2008, an ecosystem shift occurred in the Gulf of Lions, highlighted by considerable changes in biomass and fish mean weight of its two main small pelagic fish stocks (European anchovy, Engraulis encrasicolus; European sardine, Sardina pilchardus). Surprisingly these changes did not appear to be mediated by a decrease in fish recruitment rates (which remained high) or by a high fishing pressure (exploitation rates being extremely low). Here, we review the current knowledge on the population's dynamics and its potential causes. We used an integrative ecosystem approach exploring alternative hypotheses, ranging from bottom-up to top-down control, not forgetting epizootic diseases. First, the study of multiple population characteristics highlighted a decrease in body condition for both species as well as an important decrease in size resulting both from a slower growth and a progressive disappearance of older sardines. Interestingly, older sardines were more affected by the decrease in condition than younger ones, another sign of an unbalanced population structure. While top-down control by bluefin tuna or dolphins, emigration and disease were mostly discarded as important drivers, bottom-up control mediated by potential changes in the plankton community appeared to play an important role via a decrease in fish energy income and hence growth, condition and size. Isotopic and stomach content analyses indicated a dietary shift pre-and post-2008 and modeled mesozooplankton abundance was directly linked to fish condition. Despite low energy reserves from 2008 onwards, sardines and anchovies maintained if not increased their reproductive investment, likely altering the lifehistory trade-off between reproduction and survival and resulting in higher natural mortality. The current worrying situation might thus have resulted from changes in plankton availability/diversity, which remains to be thoroughly investigated together with fish phenotypic plasticity.