Overwintering dormancy behaviour was studied in female silver eels in Lake Mälaren in Sweden between 2008 and 2010. Depth choices and movements in relation to temperature were analysed from pressure andtemperature records for 13 eels with implanted data storage tags, covering 17 overwintering periods and threeintervening summer periods. Dormancy commenced in October–November as temperatures fell below 4–12 °C.Eels tended to remain motionless throughout the winter, with some short periods of activity signalled by smallchanges in depth distributions. During dormancy, the eel shows a clear avoidance of shallow areas <5 m in favourof the 10–25-m-depth interval. Activity tended to resume 4–6 months later in April–May as temperatures roseabove 3–7 °C and ice cover broke, and eels spent more time at shallower depths of <5–10 m. The majority of theeels were assessed as being in the silver eel stage at the time of tagging. During the autumn months, the divingbehaviour, with frequent and large vertical excursions and periods at the surface, was similar to that seen inmigrating eels in the Baltic and Atlantic Ocean. In spring and summer, the behaviour differed, being dominated bymore gradual depth variations, implying that the eels reverted from silver eel migration behaviour to yellow eelforaging behaviour. Body weight declined during dormancy, but other studies of starvation over comparable timeperiods showed significantly higher average specific weight losses, implying that the Mälaren silver eels must havefed between the end of dormancy and recapture.

Eels (Anguilla spp) are in decline worldwide and the signs of a reduced recruitment have been observed in continental Europe ever since the early 1970s. In order to protect and recover the European eel, EU (the European Union) decided in 2007 to establish a recovery plan, aiming at protection and restoration. Stocking is, together with reduced fisheries and higher survival when passing hydro-electrical power plants, a measure accepted by the EU, and is today used by many European countries, according to their management plans. In the early 20th century eel stocking programs started in Sweden and in other countries around the Baltic Sea, and in Sweden the responsible authorities encouraged stocking activities already at the end of 19th century.In this study, tagging experiments were conducted to follow eel migration from Lake Mälaren and from four sites along the Swedish east coast in the Baltic Sea. Recaptured tagged eels were retrieved from the fishermen, allowing for the opportunity to investigate origin by otolith microchemistry and to discover morphological differences after tagging. Several changes took place; e.g. eye index increased while weight and condition decreased with migrated distance and time until recapture.A majority of the tagged eels in Lake Mälaren did not migrate out of any of the outlets in the eastern part of the lake, irrespective of their origin. Most of them were caught in the opposite direction and continued to be caught in the lake 1-3 years after tagging, with significant weight losses. Overwintering is suggested to be a poorly chosen option, but it is uncertain whether this is a natural behaviour or a result of translocation and restocking.Concerning coastal eels, origin had no effect on the migration behaviour; a majority of the tagged eels migrated towards the outlet of the Baltic Sea. Interestingly, a minority of the recaptured eels originated from stocked fish. Instead, they were dominated of natural immigrants who had spent most of their lives in brackish waters.

One objection to the stocking of translocated eels as a management measure for the European eel Anguilla anguilla L. is that these eels may lack the ability to find their way back to the spawning area in the Sargasso Sea because the translocation will confuse their imprinted navigation. We undertook a series of tagging experiments using satellite tags, data storage tags and acoustic tags to test the hypothesis that eels translocated 1200 km from the UK to Sweden differed in their ability to migrate compared to naturally recruited eels. Eels to be tagged were caught in 2 locations, one with a record of eel stocking for more than 20 yr and with a series of barriers to upstream migration and another in a river with only natural immigration and without barriers to upstream migration. In the first year, the naturally recruited and stocked eels were released in a fjord where the initial escapement behaviour could be monitored by acoustic tagging in addition to using archival tags to track the subsequent marine migration. In the second year, eels were tagged with archival or satellite tags and released on the open coast, and only their marine migration was investigated. Eels were tracked more than 2000 km along a route that, after leaving the Skagerrak, followed the Norwegian Trench to the Norwegian Sea, turned south and west along the Faroe-Shetland channel before emerging into the Atlantic Ocean, and then continued west. There were no statistically significant differences in estuarine or oceanic behaviour regarding route, swimming speed and preferred swimming depth between stocked and naturally recruited eels. These results provide the first empirical evidence of a Nordic migration route and do not support the hypothesis that a sequential imprinting of the route during immigration is necessary for adequate orientation or behaviour during the adult spawning migration.

Stocking of eels is one of the management measures in the eel regulation (EC No 1100/2007) to recover the stock. The Swedish Eel Management Plan doubles the numbers stocked to 2.5 million eels per year. Whether stocked eels contribute to the spawning stock or not has been questioned: stocked eels might not migrate as successful as wild recruited eels. The EIFAAC/ICES Working Group on Eel (2011) recommended that all stocked eel should be marked and thereby separable from wild eel in subsequent sampling'. Since 2009, eels stocked in Sweden have been bathed in a strontium (Sr) solution, which gives a detectable mark in their otoliths. So far, 5.5 million eels have been marked in Sweden; Finland imports eels for stocking via Sweden, and these 0.6 million eels were marked by two Sr rings. We present results on marking success and recapture rates and also from marking with alizarin complexone and PIT tags in combination with Sr. If all eels stocked in the Baltic are marked, their contribution to the spawning run can be estimated. Using different combinations of marks in different regions, the relative contribution from separate stocking programmes can be evaluated. To increase the set of suitable marks, barium was tested as an additional tracer.

In a mark–recapture study in 2006, migrating European Anguilla anguilla silver eels were caught,tagged and released in the Baltic Sea and recaptures in commercial pound nets examined for possibleeffects on migration of infection with the swimbladder parasite Anguillicola crassus. The overallrecapture rate was 36%. The prevalence of infection was lowest at the northernmost sampling site.There were no significant differences between infected and uninfected A. anguilla in conditionindices, body fat content and estimated migration speeds. Parasite infection intensity levels weresignificantly negatively correlated with times and distances covered between release and recapture,but did not correlate with migration speed. It appears that more heavily infected A. anguilla wererelatively more vulnerable to recapture in pound nets. It is hypothesized that parasite-induced damageto the swimbladder inhibited vertical migrations and infected A. anguilla tended to migrate inshallower coastal waters, relatively close to the shore.

Restoration of the depleted stock of the European eel (Anguilla anguilla (L.)) requires anthropogenic impacts to bequantified, reduced, and controlled. In this article, we assess the impact of the silver eel fishery on the Baltic Coast in Sweden, applying survival analysis to 60 years of mark–recapture experiments, involving 8000 recaptures out of 18 000 releases. Thehazard of being recaptured (overall 46%) varies along the coast and declined substantially over the decades. But, most notably,the hazard for the individual diminishes strongly after the first kilometres en route. This individualized hazard disqualifies themore traditional mark–recapture methodology, which assumes random recaptures. We advocate the general use of survivalanalysis for conventional mark–recapture data. The result of our analysis indicates that the impact of the fishery just prior the2009 fishing restrictions was in the order of 10%—in itself well within sustainability limits, though only but one of the factors contributing to the mortality in the Baltic Sea.