Rochet, M-J., Prigent, M., Bertrand, J. A., Carpentier, A., Coppin, F., Delpech, J-P., Fontenelle, G., Foucher, E., Mahé, K., Rostiaux, E., and Trenkel, V. M. 2008. Ecosystem trends: evidence for agreement between fishers' perceptions and scientific information. – ICES Journal of Marine Science, 65: 1057–1068. The results of a survey on fishers' perceptions of recent changes in the eastern English Channel ecosystem carried out in 2006 were compared with fishery and bottom-trawl survey data. A hypothesis-testing framework was used, testing the null hypothesis that fishers' statements were true, which permitted evaluation of both agreement and disagreement. Overall good agreement between fishers' statements and scientific data was found, and both sources suggested that the fish community in the Channel is undergoing large changes, among which are decreases in some commercially important species; in addition, a number of human pressures impact the ecosystem. Fishers had an accurate perception of changes and their time-frames, but not necessarily of their causes. They had a greater power than survey data to detect recent changes, showing that fishers' perceptions have great potential as early warning signals.

A glass eel fishery exists downstream of the Arzal estuarine dam in the Vilaine (Brittany, France). Catch statistics were collected between 1996 and 2000, and processed using a subsampling technique which allowed data from a reliable subsample to be extrapolated to the whole fishery. During the same period, glass eel migration into fresh water was monitored using a glass eel trap located on the dam -the upstream limit of the fishery. The analysis of glass eel biology and exploitation shows that the glass eel fishery is very intensive and that there was more or less no escapement during the fishing seasons studied. The proportion of the stock successfully migrating towards fresh water, as compared with the total catch is estimated to range between 0.3 and 3.9%. In such a fishery, the fishing effort affects the abundance. Consequently, the total catch has to be used instead of catchper-unit-effort to estimate abundance.

The construction of the Arzal dam eel ladder in 1996 allowed enhancing fluvial recruitment from a negligible level, limited to yellow eels crossing sluice and dam overflow, to a level ranging from 0.2 to 2.4 million glass eels per year. The effect of such recruitment on the eel population (Anguilla anguilla) of the Vilaine watershed was analysed at 19 electrofishing sampling stations. From 1998 to 2003, average densities of eel varied from 0.34 to 0.72 eels.m -2 , with a maximum reached in 1999 of 0.82 eels.m -2 . Fluvial recruitment failure, together with density dependent mortalities, explains the drop in yellow eel densities observed from 1998 in the downstream area (< 50 rkm) and from 2000 in the middle stream area (50-100 rkm). The largest densities consisted of a large proportion eels classified as age 1 and the increased density in middle stream sectors was interpreted as the consequence of density-dependent migration at the periphery of a saturated area. The construction of 13 eel ladders on the Vilaine waterway in 1999 and 2000 was followed by increased densities in the upstream area (> 100 rkm) in 2001. The comparison to an electrofishing survey performed at 17 stations in 1981, ten years after dam construction, confirmed that ladder installation increased densities by a factor 6 and modified the population structure, with densities of eels age 0 and 1 multiplied 29 fold. This change corresponded to reduced escapement rates to the glass eel fishery (1% to 5%). Considering this result, a preliminary escapement target of 240 glass eel per km 2 of watershed area or 1 500 glass eel per ha of water surface, is proposed and discussed for glass eel fisheries.

Compte rendu des synthèses et discussions de l'atelier de lecture d'écaillés (Rennes, 1 er au 4 avril 1985).

Glass eels (Anguilla anguilla) arriving after the fishing season in the Vilaine estuary (France) were caught by boat and at a trapping ladder located at a blocking dam upstream in the estuary. They were marked with Bismarck brown, rhodamine B or neutral red, and released into the estuary. The effect of marking on behaviour, mortality and recapture differed according to the dye and concentration used. The glass eels also differed in response to marking according to their origin. Estuarine glass eels were clearly desynchronized in their selective tidal stream transport and left the release area after a few tidal cycles. Trapped glass eels were less sensitive to marking. They remained in the release area and were recaptured in larger number both in the trapping ladder and in the estuary. These results supported a model of two behavioral fractions in the estuary: flow-carried migrants and active glass eels with stronger rheotactic behaviour. The proportion of active glass eels in the estuary shifted from none for the two weeks after the fishery closure, to 10% value of the total stock. At the trapping ladder, the migration ended in June each year, while a large stock was still present in the estuary. These observations favour the interpretation of a short duration of the active phase at this site followed by the transition towards a more benthic behavior. The daily efficiency of the trapping ladder was calculated as 4% of the active stock from both stock estimations and average recapture rate. This value was consistent with the low annual efficiency of the trapping ladder, calculated as 30% of the estuarine stock of glass eels that arrived after the fishing season. Key-words:Anguilla anguilla, vital dye, glass eel, marking recapture. ESTIMATION DE STOCK DE CIVELLES EN ESTUAIRE PAR MARQUAGE-RECAPTURES À L'AIDE DE COLORANTS VITAUX RÉSUMÉLes civelles (Anguilla anguilla) arrivant après la fin de la saison de pêche dans l'estuaire de la Vilaine (France) ont été capturées à l'aide d'un bateau de pêche et sur une passe piège située à l'amont de l'estuaire sur le barrage qui limite la progression du stock migrant. Elles ont ensuite été marquées à l'aide de Brun-Bismarck, rhodamine B et rouge neutre et relâchées en estuaire. L'effet du marquage sur le comportement, la mortalité et les recaptures, a différé en fonction du colorant et de la concentration utilisés. Les civelles ont également montré une réponse différente au marquage en fonction de leur origine. Celles qui étaient prélevées en estuaire étaient clairement désynchronisées dans leur transport tidal sélectif et ont quitté la zone de lâché après quelques cycles de marées. Celles qui provenaient de la passe ont été moins sensibles au marquage. Elles sont restées dans la zone de lâché, et ont été recapturées en plus grand nombre à la fois dans la passe et dans l'estuaire. Ces résultats permettent de proposer un modèle ou deux fractions comportementales sont distinguées dans l'estuaire, des civelles en phase de migration portée et des civelles en migration acti...