Fish migration studies in the Baltic Sea - a historical review

Transcription

1 ICES Marine Science Symposia, 215: Fish migration studies in the Baltic Sea - a historical review Eero Aro Aro, E Fish migration studies in the Baltic Sea: a historical review. - ICES Marine Science Symposia, 215: The distribution of Baltic Sea fish species throughout their life history may be random or patterned, and these changes have been estimated by tagging programmes since the early years of ICES. The adaptive value of fish migration is strongly coupled with the optimization of the surrounding environment. The hydrographic conditions in the Baltic form an uneven continuity from southwest to northeast, and the physiological selection of fish species, their distribution, and their capability to migrate has taken place in varying environmental temperatures and salinities. In the life history of fish species in the Baltic Sea, the eurythermie and euryhaline species have had more adaptive value than those species which prefer constant salinities and temperatures. The euryhaline and eurythermie species are usually able to migrate more intensively than stenothermic and stenohaline species. Marine species have both rather local and more migratory stocks. Freshwater species are, in general, more stationary than marine species. The advantage of migrations in a patchy environment is that the impact of environmental variability on reproduction, survival, foraging, and growth decreases. ICES, as a coordinating organization, has played an important role in migration studies during the last 100 years and has been significant as a forum for developing new ideas and hypotheses on fish migration in the Baltic Sea. Keywords: Baltic herring, Baltic salmon, cod. flounder, freshwater species, ICES, migration, plaice, sea trout, tagging. Eero Aro: Finnish Game and Fisheries Research Institute, Pukinmäenaukio 4, PO Box 6, FIN Helsinki, Finland; tel: ; fa x: ; eero.aro@ rktl.fi. Introduction If the environmental needs of fish species were constant through time, there would be no reason for them to move from place to place. The distribution of fish throughout their life history may be random or patterned, and the environmental needs of individuals may vary from day to night, summer to winter, and from the time of hatching to their adulthood. The spatial and temporal distributions of many fish species change in more or less regular and cyclical patterns. The adaptive value of fish migration is strongly coupled with the optimization of the surrounding environment. The Baltic Sea is the product of the last glacial period and has many unusual characteristics. It hosts an aquatic biota where both marine and freshwater organisms live side by side with a number of living relict species. The hydrographic conditions in the Baltic Sea form an uneven continuity from southwest to northeast, so the physiological selection of fish species, their distribution, and their capability to migrate has taken place in varying environmental temperatures and salinities. Historical background of ICES and fish migration research in the Baltic Sea Early years at the beginning of the 20th century Since the early years of ICES, fish migration, distribution, and interaction studies have formed an important part of ICES research activities (Hoek and Hjort, 1903; Garstang, 1903; Hoek and Garstang, 1903). These activities led to the development of Petersen's famous principle (Petersen button), attaching a tag button by a silver wire near the middle of the dorsal edge of the body. Finnish Fisheries Inspector Oscar Nordqvist, who was Convener of ICES Committee C (Committee for the Baltic) at the very beginning was the first to propose migration studies and large-scale tagging research in the Baltic Sea. He drafted a research proposal in 1902 for the Baltic (Nordqvist, 1903) which included basic work on salmon, sea trout, eel, sprat, cod, and flounder and the effect of seals on fish stocks. His first draft also included migration studies, but his work had to be com

2 362 E. Aro FIN NOR Salmon Eel SWE EST RUS LAT GER POL RUSI LIT Figure 1. The first recaptures of Baltic salmon and eel in early 1900 as reported to ICES in pleted by Filip Trybom (Sweden) and C. G. J. Petersen (Denmark) after Nordqvist s unfortunate departure caused by the political situation in Finland. The Czar of Russia issued an edict on 26 September 1902 to the Finnish Senate to discharge Oscar Nordqvist for political reasons, effective 1 October However, the outcome of Nordqvist s initiative was a draft report, produced in 1907 by P. P. C. Hoek and F. Trybom and assisted by Nordqvist s successor J. A. Sandman, to the governments of Denmark, Finland, Russia, and Sweden regarding the Baltic salmon fisheries (Hoek and Trybom, 1907). The report included a discussion about the reasons for the declining stock, the nature of the fishery and its distribution between coastal and open-sea areas, minimum landing size, and minimum mesh size (Nordqvist, 1907), all in relation to the timing and pattern of salmon migrations. Trybom (1907a) also drafted a special "proposal regarding an international Treaty between the Baltic countries" which contained a number of actions to be taken to "prevent any further diminution of the stock of salmon, maintain the salmon stock in the Baltic and possible increase of catch by the erection of hatcheries where the greatest possible quantities of salmon might be bred". This proposal, considered a predecessor of the present International Baltic Sea Fishery Commission s (IBSFC) Salmon Action Plan (Anon., 1995), was not supported by the Baltic countries, and the ICES Bureau was not yet prepared to bring forward a definite proposal on the matter in June The special report by Trybom (1907b) gave detailed results of Baltic salmon and sea trout tagging and migration studies conducted by various Member Countries in the Baltic area during Among the recaptures was one salmon originally tagged off Bornholm in April 1906 and recaptured four months later in Bothnian Bay. This is believed to be the first documented Baltic salmon homing record in the scientific literature. Another salmon tagged in September 1905 in the eastern part of the Gulf of Finland outside the mouth of the Kymijoki River was recaptured in May 1907 in the neighbourhood of Memel (Klaipeda) showing, as a first record, the possible extent of the feeding migration of Baltic salmon ( Figure 1). These early taggings of Baltic

3 Fish migration studies in the Baltic Sea - a historical review 363 A. Coastal zone B. Small restricted "shells" C. Rocky areas D. Steep slopes E. Open/deep sea area F. Larger reefs in open sea G. Special areas (wrecks etc). Figure 2. Typical fishing ground categories in the Baltic Sea. salmon and sea trout were aimed mainly at estimating the distribution area and proportion removed by opensea fisheries and coastal fisheries (Trybom, 1907b), and although the number of tagged fish each year was rather low, tagging provided new information on migration patterns of Baltic salmon and sea trout. The state of the eel stock in the Baltic also received attention (Trybom and Schneider, 1907) in Committee C at the beginning of O. Nordqvist and J. A. Palmén started tagging eels in 1903 at the Tvärminne Zoological Station founded by Palmén in 1902 on Hanko Peninsula, Finland, the westernmost part of the Gulf of Finland (Nordqvist, 1903). Eel taggings were expanded further in These first silver eel taggings and the subsequent recaptures in the Atlantic Ocean (Figure 1) contributed to the development of Johannes Schmidt s famous hypothesis on eel life history (Schmidt, 1906). Priorities during the 1920s-1940s Fish migration studies in the Baltic in the 1920s-1940s concentrated mainly on commercially important marine, anadromous, and catadromous species such as Baltic herring, sprat, cod, plaice, flounder, eel, and salmon. The freshwater species in the Baltic did not receive any special attention from ICES, although many laboratories tagged freshwater fish in coastal areas (Ekman, 1916; Henking, 1923). The main aims of all tagging studies during this period were to estimate the spatial and temporal distribution patterns and their annual changes. In the 1920s and 1930s, scientists began the practice of categorizing the recapture information according to distance travelled and to the typical fishing grounds in the Baltic Sea (Figure 2). In the 1920s, special attention was paid to the most abundant species in the pelagic system, Baltic herring and sprat, and to their predator, Baltic cod (Heidrich, 1925; Hessle, 1923, 1925, 1927). Hessle (1925) showed that the slow-growing, long-lived, open-sea herring spawn mainly along the Swedish east coast from Hanö Bay to the Åland Sea in deeper water than the coastal spring-spawning herring in other locations. Hessle (1925) also showed that Baltic herring along the Swedish coast undertook long migrations extending from the Åland Sea to the southern part of the Baltic. In the 1920s, sprat were shown to inhabit the Baltic from the Belt Seas and Western Baltic (ICES Subdivisions 22 and 24), to the Quark area in the north (Subdivision 30), and to the northeastern part of the Gulf of Finland (Subdivision 32) (Hessle, 1927). Three different sprat stock components were considered at that

4 364 E. Aro time, with a low rate of mixing with the Kattegat and Skagerrak stocks due to the salinity gradient and differences in many abiotic factors between the Western Baltic and the Kattegat (Hessle, 1927). Hessle ( 1923) and his colleagues reported on the distribution, spawning, growth, and periodicity of Baltic cod year classes along the Swedish coast. Although these studies included only areas in the vicinity of the Swedish coast, they indicated great fluctuation caused by changes in the distribution patterns of cod year classes. Kändler ( 1944), in his classical work in which small numbers of cod were tagged, showed differences between cod stocks in the Baltic in the Transition Area around Bornholm (14 E). In the case of Baltic salmon, Järvi (1938) reported that they nearly all fed in the Baltic Main Basin (Subdivisions 24-29) and that the most important feeding grounds were the Gotland Deep, the waters around Bornholm, and the Gdansk Basin. This important work summarized 15 years of information ( ) on the basic biology, the fisheries, and obvious reasons for fluctuations in the Baltic salmon. Decades of tagging and migration studies during Fish migration studies using tagging were intensified considerably in ICES circles in the 1950s. An initial stimulus came when a Danish cutter accidentally discovered dense concentrations of young herring on the Bløden Grounds in July 1950 about miles west of Esbjerg, Denmark, in the North Sea. This discovery and the fishery which subsequently developed stimulated discussion about the role and effect of the Bløden fishery on the North Sea herring fishery and led to an extensive tagging programme for herring in the North Sea and the Baltic Sea. The establishment of the International Baltic Sea Fishery Commission (IBSFC), and two ICES stock assessment working groups necessary for IBSFC activities served to elevate the need for tagging studies in the early 1970s. Thus, the present knowledge of fish migration in the Baltic is based mainly on taggings made from the early 1950s to the 1980s, except in the case of Baltic salmon and sea trout, large numbers of which have been tagged each year up to the present. Between 1950 and 1980, several hundred thousand fish were tagged throughout the entire Baltic area (Aro, 1989) including Baltic herring, sprat, cod, plaice, flounder, Baltic salmon, sea trout, and freshwater species such as whitefish, pike, pikeperch, perch, bream, and burbot. A common feature of all the tagging work has been to estimate annual migration patterns and their changes, the home range of various species, site fidelity for spawning, distribution of stock components and their mixing, and the effects of fishing at stock level, and also to evaluate the timing of annual migration patterns. Main results from tagging and migration studies during Herring Tagging and migration studies have shown that in some areas Baltic herring can be divided into two or three components by their different migration patterns, such as spring-spawning coastal herring, spring-spawning open-sea herring, and autumn-spawning herring. Baltic herring in the southwestern Baltic, Kattegat, and Skagerrak have both spring-spawning and autumn-spawning components (Anwand, 1963a, 1963b), with springspawning stocks having a very clear migration pattern. The majority of the fish migrate from the feeding and wintering areas in the Skagerrak and Kattegat through the Öresund and the Belts into the Baltic in autumn and early winter, with a minority coming from other feeding areas around Bornholm, off the Polish coast, and Hanö Bay to the spawning grounds (Biester, 1979; Otterlind, 1985). The migration pattern of adult herring from the Baltic to the Kattegat and Skagerrak has also been confirmed by the occurrence of Anisakis nematode larvae in the adult herring from the Southwestern Baltic (Strzyzewska and Popiel, 1974; Friess, 1977; Kiihlmorgen-Hille, 1983). Herring in the Southwestern Baltic seem to have a clear homing ability (Jönsson and Biester, 1979) despite some controversy. In the Central Baltic, there are assumed to be three different stocks: spring-spawning coastal herring, spring-spawning opensea herring, and autumn-spawning herring (Popiel, 1984). These stock components have different migration patterns in the Southeastern and Central Baltic. In the Bothnian Sea, there are two spring-spawning coastal herring components (Hannerz, 1955, 1956; Otterlind, 1957; Parmanne and Sjöblom, 1982, 1986), one along the west coast and one on the east coast, both of which show high spawning-site fidelity. Sprat Sprat inhabit the Baltic Sea from the Belt Seas and western Baltic (Subdivisions 22 and 24), to the Quark area in the north (Subdivision 30), and to the northeastern part of the Gulf of Finland (Subdivision 32) (Hessle, 1927; Veldre, 1986; Rechlin, 1986). There are three different sprat stocks in the Baltic, and mixing with the Kattegat and Skagerrak stocks is considered to be very low, although there is no significant difference in morphometric characters and vertebrae counts (Lindquist, 1968). Mixing is probably prevented by the salinity gradient and differences in many abiotic factors between the Western Baltic and the Kattegat (Hessle, 1927; Aps et al 1987). Boundaries between neighbouring stocks are unclear, and stock mixing during feeding and wintering is apparent (Rechlin, 1986). During the feeding period in July-November in the south and central parts

5 Fish migration studies in the Baltic Sea - a historical review 365 of the Baltic (van Khan et al., 1972) and in August- December in the north, sprat migrate from the more coastal areas to the offshore parts of the Baltic basins forming feeding shoals which contain mainly older age groups (Lindquist, 1971). The range and amplitude of the migration patterns of sprat stocks and their mixing in the Baltic is not yet very clear. Some observations indicate more locality (Aps and Lotman, 1984; Aps et at., 1987), but some, on the contrary, indicate more migratory behaviour and stock mixing (Grauman, 1976; Khoziosky et al., 1983). and Auvinen, 1984a, 1985). On these feeding grounds, the stocks mix thoroughly, and there is no clear evidence on the preferred feeding areas of individual stocks, although there are some aggregations on the main fishing grounds (Carlin, 1959; Halme, 1964). Feeding salmon are more or less clustered, and changes in density on the feeding grounds reflect their active movement in search of food (Thurow, 1973). All Baltic salmon stocks exhibit homing behaviour, but in some releases, homing has not developed because of environmental factors (Christensen, 1982). Salmon Salmon in the Baltic are distributed from the Belt Seas and Western Baltic to the northern parts of the Gulf of Bothnia and the Gulf of Finland and very seldom migrate outside the Baltic (ICES, 1980; Lindroth et al., 1982). There are naturally spawning stocks and artificially reared stocks of various origin. During the marine phase, the Baltic salmon have very clear migration patterns in the coastal and pelagic area. In general, both the wild and hatchery-reared stocks have the same migration patterns, with some exceptions. The migrations are divided into post-smolt, feeding, and spawning phases. Salmon smolts enter the Baltic usually in April-June, with those from the southern Baltic running a few weeks earlier than those from the north. The wild and hatchery-reared smolts start their post-smolt migration at about the same time. During the first few weeks, postsmolts are relatively stationary, adapting to the new environment, and they remain close to the river mouth or releasing place (Bartel, 1976; Ikonen and Auvinen, 1985). From releases on the Finnish side of Bothnian Bay (Subdivision 31), post-smolts migrate southwards along the Finnish coast to the Quark area where they shift to the Swedish coast. From the Swedish releases, the post-smolts migrate south along the Swedish coast of Bothnian Bay to the Quark where some shift to the Finnish side following the main current northwards along the Finnish coast, and do not leave Bothnian Bay (Larsson and Atheskar, 1979). In the Bothnian Sea (Subdivision 30) stock, the post-smolts behave like the northern stocks, except for the River Neva stock (from Russian origin in the eastern Gulf of Finland, Subdivision 32) releases in the Bothnian Sea (Ikonen and Auvinen, 1982), which seem to have more localized tendencies. Nearly all Baltic salmon stocks feed in the Baltic Main Basin (Subdivisions 24-29), except those of River Neva origin. From the releases in the Gulf of Finland, about 31% of the Neva-origin fish feed in the Main Basin, while the percentage from the Bothnian Sea releases is less. The most important feeding grounds are the Gotland Deep, the waters around Bornholm, and the Gdansk Basin (Järvi, 1938; Halme, 1964; Carlin, 1959; Thurow, 1973; Christensen and Larsson, 1979; Ikonen Sea trout Sea trout in the Baltic are distributed from the southwestern part to the northern parts of the Gulf of Bothnia and eastern parts of the Gulf of Finland. Their migration may be divided into two patterns: the long-distance migrating stocks and the more stationary stocks (Zarnecki and Duszynski, 1961; Ikonen and Auvinen, 1984b). In the Baltic, sea trout occupy an intermediate position between the pelagic and coastal fish communities, and their migration pattern is dependent on stock origin and the dimensions of the archipelago. During their feeding migration, the stocks in the Bothnian Sea and Bothnian Bay intermix when migrating from the east coast to the west coast and vice versa. The migration southwards from Bothnian Bay to the Bothnian Sea and to the Main Basin has also been observed even if the northern stocks are considered to be more local and less migratory in nature (Carlin, 1965; Ikonen and Auvinen, 1984b). In the Gulf of Finland, the feeding migration encompasses the whole area from west to east, and there appears to be some migration to the Archipelago Sea, the Bothnian Sea, and the Main Basin (Ikonen and Auvinen, 1984b). In the northeastern Baltic near the island of Saaremaa, the stock seems to be quite local, but in some cases it extends into the Bothnian Sea, the Åland Sea, near Gotland Island, and southwards to the Bornholm region (Rannak et al., 1983). In the Southern Baltic, the feeding migrations are more intensive and extend to quite large areas. From the Pomeranian coast and the Vistula region, sea trout migrate mainly eastwards, but parts of the stock also move westward along the Polish coast. The feeding migration of these stocks may reach the Gulf of Riga, the Gulf of Finland, and the Bothnian Sea (Backiel and Bartel, 1963; Chrzan, 1963). The stocks along the Swedish south and east coasts migrate northwards and to the Bothnian Sea, and the Gulf of Finland, the Gulf of Riga, as well as southeasterly to the Gulf of Gdansk (Svärdson and Fagerström, 1982).

6 366 E. Aro Cod There are two cod stocks in the Baltic which have been shown to differ in many ways. The Western Baltic stock (Gadus morhua morhua L.; the Atlantic cod or Transition Area cod) is distributed west of Bornholm, in the Western Baltic, the Belt Seas, and the Sound (Subdivisions 22-24) and has regular connections to the Kattegat (Division Ilia) and to the southeastern Baltic (Bagge et a i, 1994). The Eastern Baltic stock (Gadus morhua callarias L.; the Baltic Sea cod) is distributed east of Bornholm to the northern parts of the Bothnian Sea and to the eastern parts of the Gulf of Finland (Bagge et al., 1994). The border between these two stocks is diffuse, and mixing is evident in the Arkona and Bornholm Basins. Western Baltic cod migrate in all directions after spawning (Berner, 1981) from the deep waters to more shallow coastal areas to feed. The young, immature age groups usually remain in the coastal areas before joining the mature stock. From the Arkona region, feeding migrations of adults may reach the Belt Seas in the west and extend eastwards to the Slupsk Furrow, the Bornholm region, the Gdansk Deep, and even to the southern Gotland Deep (Lamp andtiews, 1974; Berner, 1981 ). Cod in the Arkona Basin are always a mixture of the two main stocks, and the spawning migration occurs in two directions. According to transplantation experiments, the homing of cod is not well developed (Bagge, 1983). Eastern cod migrate to the Bornholm Basin spawning grounds in December-February from the feeding grounds in Slupsk Furrow, Gdansk Bay, Hanö Bay, and the Gotland Deep (Netzel, 1968, 1974). The spawners in the Gdansk Deep originate from areas south of the Bornholm Basin and from southern parts of the Gotland Deep (Netzel, 1974; Otterlind, 1976). There is also a spawning migration from the Swedish east coast, the Åland Sea, the southern Bothnian Sea, and the Gulf of Finland to the Bornholm Basin, the Gdansk Basin, and the southern parts of the Gotland Basin in December- March (Otterlind, 1976; Sjöblom et al., 1980; Aro and Sjöblom, 1983a). The homing behaviour of cod in the Southeastern and Northern Baltic is also unclear, and cod may use different spawning grounds in successive years (Bagge, 1983; Otterlind, 1984, 1985). During the feeding period, cod are distributed over large areas and may undertake extensive migrations (Otterlind, 1985; Aro 1989) that exhibit no pattern. Flounder There are several rather distinct flounder stocks or populations in the Baltic (ICES, 1978). Flounder are regularly distributed in all parts of the Baltic, except Bothnian Bay, the easternmost part of the Gulf of Finland, and the deepest areas of the Gotland Deep (Anon., 1978). There are at least three stocks in the Southwestern and Southeastern Baltic (Subdivisions 22-26), three in the Central and Northeastern Baltic (Subdivisions 27-28), one each in the Åland Sea, Archipelago Sea, and southern Bothnian Sea, and two in the Gulf of Finland (Anon., 1978). Migrations between the mature flounder stocks in the southern and central Baltic are quite rare, extending from the southern part of Öland Island to the Rosewie on the Polish coast (Otterlind, 1965, 1967; Vitinsh, 1976, 1977). Migrations are effectively blocked in the Southeastern and Central Baltic by the Gdansk and Gotland Deeps and in the Northern Baltic by the Gotland Deep and the deep southwest of the Åland Islands (Otterltnd, 1965; Vitinsh 1972, 1976, 1977; Aro and Sjöblom, 1983b). The annual migration patterns of flounder stocks are quite well known and, because of their general locality, the homing behaviour of flounder is obvious. Plaice Plaice are distributed in the Baltic Sea from the Belt eastwards to the Gdansk Bay area and northwards to the southern Gotland Deep (Bagge, 1981). Plaice are very rare in the Northern Baltic. Feeding migrations from the deeper spawning grounds to shallower waters are primarily west and to a lesser extent east from the Arkona Basin. The westward feeding migration may reach the Belt Sea, the Sound, the southern Kattegat, and even the Skagerrak (Otterlind, 1967). Those emigrating out of the Baltic Sea are assumed not to return (Otterlind, 1967). The eastward migration is most intensive from the Arkona Region to the east and southeast of Bornholm during November-February. Plaice feeding in the Gdansk Deep have been shown to migrate to the Bornholm Basin to spawn (Cieglewicz, 1961). Freshwater species In the Southern Baltic where the freshwater species are almost totally absent, shallow coastal bottoms and pelagic areas serve as habitat for the young stages of several marine pelagic and demersal species of economic importance. These areas also serve as habitat for a number of other littoral and coastal species of substantial ecological but minimal economic importance. In the Central and Northern Baltic Sea, there is a clear dominance of freshwater species in the littoral and coastal areas (Neuman, 1982), but their abundance decreases towards the seaward limit of their distribution (Lehtonen and Toivonen, 1981). The number of freshwater species is greatest in the archipelagoes, bays, inlets, and river mouths. The migrations of freshwater species are generally local, although some species long distances. Some of the freshwater species lack a migration pattern or their migrations are too short to identify. Two whitefish

7 Fish migration studies in the Baltic Sea - a historical review 367 species, river-spawning whitefish (Coregonus lavaretus L.) and sea-spawning whitefish (Coregonus widegreni Malmgren), have different migration patterns. They are distributed along the coastal areas in the Baltic and are rare in the Belt Seas and along the south coast of Sweden (Svärdson, 1979), but are more abundant in the Gotland Island area, along the west coast of Estonia, and in the Gulf of Finland, and the Gulf of Bothnia (Lehtonen, 1981). The feeding migration patterns vary between species and populations. Sea-spawning whitefish do not migrate very much and feed near the spawning grounds (Lehtonen, 1981). Pike exhibit a local distribution pattern and territorial behaviour (Ekman, 1916; Segerstråle, 1951; Halme, 1957; Halme and Korhonen, 1960; Strandman, 1964; Lehtonen, 1973). Local migrations or movements during feeding are very limited since the pike is an ambush hunter or stalker (Neill and Cullen, 1974). The home range of an individual pike is usually several square kilometres and when displaced it attempts to return to its original home range (Halme and Korhonen, 1960). Pike-perch spawn in inlets and shallow bays in April-May (Lehtonen and Toivonen, 1981). After spawning, they remain on the spawning grounds and feed (Lehtonen and Toivonen, 1987) or migrate inside the archipelagoes, to the open sea, or along the coasts (Henking, 1923). Annual migrations take place between spawning inlets or bays, feeding areas in the archipelagoes, and wintering areas in deeper waters (Toivonen, 1968). The average dispersal area is smaller when the coast is open and larger when the archipelago is rich (Toivonen, 1968; Lehtonen and Toivonen, 1987). The mixing of local neighbouring stocks occurs in the wintering areas, but pike-perch exhibit a clear homing behaviour to their former spawning grounds (Lehtonen, 1985). In the Southern Baltic, the feeding migration is directed to the open sea, and pike-perch overwinter in shallow inlets (Henking, 1923; Filuk, 1962). The perch is one of the most common freshwater species in the Baltic archipelagoes. Perch form more or less separate local but non-discrete stocks along the Baltic coasts (Böhling and Lehtonen, 1985). Spawning occurs all along the coast, and the mixing of neighbouring stocks during spawning is unlikely (Ekman, 1916). After spawning, feeding migrations, usually short (Johnson, 1978), but sometimes long (Henking 1923), take place to deeper areas along the coasts and inside the archipelagoes. The range of feeding migrations is affected by the distribution of food resources, optimum temperature, and the abundance of neighbouring stocks, as well as the morphology of the archipelago (Böhling and Lehtonen, 1985). Conclusions The fish fauna in the Baltic Sea may be classified into three different and intermediate communities: a pelagic community, a benthic community, and a littoral and coastal community. The borders between them are not sharp, and individuals from neighbouring communities cross them frequently, particularly the littoral and coastal regions which serve the pelagic community as spawning and nursery areas. In the Gulf of Bothnia, the littoral and coastal community is dominated by freshwater species, which very seldom migrate outside this environment, and the Baltic herring is actually the only native pelagic species using that environment as a spawning and nursery area (Andreasson and Petersson, 1982). The migration and movements of the Baltic fish species occur at micro- and macroscales inside and between these communities in annual, diurnal, horizontal, and vertical patterns. The migration, spatial, and temporal distribution pattern of Baltic fish exhibits an annual cycle between spawning, feeding, and nursery habitats. Homing in spawning migrations and site fidelity has been observed, but results suggest that this link is weak in the case of cod, and the ratio of emigration to other spawning grounds seems to be dependent on the conditions of the spawning grounds. This also applies to springspawning Baltic herring and freshwater species. Baltic fish may use different spawning grounds in successive years, and evidence indicates that spawning migrations are more strongly linked to prevailing hydrographical conditions than to homing. The feeding migration of Baltic fish seems to have different patterns for Baltic herring, sprat, salmon, and sea trout, but not in the case of cod. The movements of Baltic fish from one place to another during the feeding period seem to be once-a-year events. However, they may be just local seasonal movements, dispersals, or "true" migrations. The advantage of movements and migration in a patchy environment is, in general, that the impact of environmental variability on reproduction, survival, foraging, and growth decreases. The migratory behaviour promotes more flexibility in the face of uncertainty. With migration, the risk from predation and cannibalism is balanced against the advantage of remaining in one place to exploit resources. The freshwater species in the Baltic have unfortunately been more or less neglected in ICES activities even though they are an important component of the ecosystem. The number of original freshwater species in the Baltic is 23 (Lehtonen and Toivonen, 1981). The freshwater species are practically absent in the southwestern part of the Baltic except in some estuaries, fjords, and bays (Henking, 1923; Hempel and Nellen, 1974). There are 18 species in the Central Baltic, and freshwater species dominate in the north and northeastern parts of the Baltic (Lehtonen and Toivonen, 1981). The general problem in quantifying movements and migrations (feeding or spawning) on a robust and usable spatial and temporal scale is that the tagging data rely totally on commercial fishermen s catch and effort and their willingness to return tags. Tagging of Baltic fish

8 368 E. Aro was very popular from the 1950s to the early 1980s. Most of these tagging studies and experiments were planned to estimate movements and fishing mortality. The results of these studies have only been partly published in the scientific literature. The analytical tools for mark-release experiments were developed decades ago and are well described (Seber, 1973; Burnham et a l., 1987). However, data analysis is hampered by a lack of essential information on fishing effort, spatial and temporal distribution of fishing activities, systematic determination of tag return rates from the commercial fishery, and so on. Many tagging experiments have neglected these issues by allocating resources mainly to catching and tagging of fish instead of to the recovery and reporting rates of tags. Before the establishment of the IBSFC in the early 1970s, there were national fishing zones and a huge international fishing zone in the Baltic Sea where fishing activities were largely unregulated. Consequently, detailed information on fishing and fisheries was very scarce, and the systematic collection of fisheries data did not exist on a large scale. The situation has changed considerably since the beginning of the 1970s, and the parameters necessary for the design and analysis of tagging experiments are now accessible. However, Baltic fish tagging experiments are not very appealing because conducting an experiment takes at least 6-8 years before final results are available. Despite these shortcomings, new tagging experiments on Baltic cod and herring should be encouraged. References Andreasson, S., and Petersson, B The fish fauna of the Gulf of Bothnia. In Coastal Research in the Gulf of Bothnia, pp Ed. by K. Müller. Dr W. Junk Publishers, The Hague. 462 pp. Anonymous Proceedings of the twenty first session. International Baltic Sea Fishery Commission. Warsaw, 4-8 September IBSFC, Warsaw. Anwand, K. 1963a. Markierungen am Riigenschen Frühjahrshering im Jahrel961 (Tagging of spring-spawning Rügen herring in 1961). Internationale Revue der Gesamte Hydrobiologie (International Review of General Hydrobiology), 48: 2-8. (In German). Anwand, K. 1963b. Vergleichende Untersuchungen an Frühjahrs- und Herbst-heringe aus den Gewässern um Rügen (Comparative studies on spring- and autumn-spawning herring off Rügen). Zeitschrift für Fischerei (Journal of Fisheries), 11: (In German). Aps. R.. and Lotman, K Natural tags of sprat in the Gulf of Finland. ICES CM 1984/J: pp. Aps, R., Benenson, I., Kaleis, M., Korytin, N., and Kryazhimsky, F Spatial model of the Baltic sprat population dynamics. ICES CM 1987/J: pp. Aro, E A review of fish migration patterns in the Baltic. Rapports et Procès-Verbaux des Réunions du Conseil International pour l Exploration de la Mer. 190: Aro, E., and Sjöblom, V 1983a. Cod off the coast of Finland in ICES CM 1983/J: pp. Aro, E., and Sjöblom, V 1983b. The migration of flounder in the northern Baltic Sea. ICES CM 1983/J: pp. Backiel. T., and Bartel, R Preliminary results of sea- and lake-trout tagging. ICES CM 1963/ pp. Bagge, O Demersal fishes. In The Baltic Sea. Elsevier Oceanography Series, 30: Ed. by A. Voipio. Elsevier Scientific Publishing Company, Amsterdam. 418 pp. Bagge, O Migration of transplanted cod. ICES CM 1983/J : pp. Bagge, O., Thurow, F., Steffensen, E., and Bay, J. 1994: The Baltic cod. Dana, 10: Bartel, R The Drava river salmon in the light of some recent tagging experiments. ICES CM 1976/M:6. 8 pp. Berner, M Dislocation parameters of tagging experiments on cod in the Baltic (Subdivision 22-25) from ICES CM 1981 /J: pp. Biester, E The distribution of the Rügen spring herring. ICES CM 1979/J:31. 6 pp. Burnham, K. P., Anderson, D. R., White, G. C., Brownie, C., and Pollock, K. H Design and analysis methods for fish survival experiments based on release-recapture. American Fisheries Society Monograph, 5., Bethesda, Maryland, USA. 437 pp. Böhling, P., and Lehtonen, H Effect of environmental factors on migrations of perch (Perea fluviatilis L.) tagged in the coastal waters of Finland. Finnish Fisheries Research, 5: Carlin. B Results of salmon smolt tagging in the Baltic area. Rapports et Procès-Verbaux des Réunions du Conseil International pour l Exploration de la Mer, 147: Carlin. B Märkning av odlade havsöringsungar i utvandringsfardig storlek (Tagging of reared sea trout at the size ready to migrate). Svensk Fiskeri Tidskrift (Journal of Swedish Fisheries), 74(7/8): (In Swedish). Christensen, O Danish experiments with salmon smolt released into the Baltic Sea from the Island of Bornholm. ICES CM 1982/M:9. 7 pp. Christensen, O., and Larsson, P. O Review of Baltic salmon research. Cooperative Research Report, pp. Chrzan, E Preliminary report on tagging experiment of sea trout in the region of Vistula Firth. ICES CM 1963/95. 7 pp. Cieglewicz, W Tagging experiments with flatfish in the southern Baltic. ICES CM 1961/95. 2 pp. Ekman, T Meddelanden rörande utförda märkningar af gädda m. fl. fiskar i Södermanslands och Östergötlands skärgärd (Communications on tagging of pike in the archipelago of Södermansland and Östergötland). Svensk Fiskeri Tidskrift (Journal of Swedish Fisheries), : (In Swedish). Filuk, J Nachkriegsstudium über Biologie und Fang des Zanders des Frischen Haffs (Studies after the war on the biology and catches of pike-perch in Frischen Haff). Zeitschrift für Fischerei (Journal of Fisheries), 10: (In German). Friess, C Investigations on the spring spawning Rügen herring infested by Anisakis in ICES CM 1977/ P: pp. Garstang, W Report of the Convener of Committee B. Rapports et Procès-Verbaux des Réunions du Conseil International pour l Exploration de la Mer, 1, Annex C: Grauman, G On the distribution and passive migrations of Baltic sprat eggs and larvae in Bornholm-Slupsk spawning ground. ICES CM 1976/P:8. 11 pp. Halme. E Gäddmärkningar utförda av Timmer Spinnfiskeklubb r.f. (Tagging of pike made by Timmer s Spinning Lure Club). Fiskeritidskrift for Finland, Ny Serie (Journal of Finnish Fisheries, New Series), 1: (In Swedish).

9 Fish migration studies in the Baltic Sea - a historical review 369 Halme, E Report on salmon tagged in Finland in ICES CM 1964/Salmon and Trout Committee, pp. Halme, E., and Korhonen, M Haukien vaelluksista rannikoillamme (On migration of pike off our coast). Kalamies (Fisherman), 4: (In Finnish). Hannerz, L Strömmingsmärkningen i Bottenhavet 1954 (Tagging of herring in the Bothnian Sea in 1954). Ostkusten (East Coast), 1: (In Swedish). Hannerz, L Preliminary results of the herring investigations in the Bothnian Sea Annales Biologiques du Conseil International pour l Exploration de la Mer, 11: Heidrich, H Über die Fortpflanzung von Clupea sprattus in der Kieler Bucht ( Reproduction of sprat, Clupea sprattus, in the Kiel Bay). Wissenschaftliche Meeresunter-suchungen (Scientific Marine Research), 20: (In German). Hempel, G., and Nellen, W Fische der Ostsee (Fishes in the Baltic Sea), ln Meereskunde der Ostsee (Baltic Sea Marine Research), pp Ed. by L. Magaard and G. Rheinheimer. Springer-Verlag, Berlin. 269 pp. (In German). Henking, H Die Fischwanderungen zwischen Stettiner Haff und Ostsee (The migration of fish between Stettiner Haff and the Baltic Sea). Zeitschrift für Fischerei (Journal of Fisheries), 22: (In German). Hessle, C Undersökningar rörande torsken (Gadus callarias L.) i mellersta Östersjön och Bottenhavet (Research on cod (Gadus callarias L.) in the Baltic Main Basin and in the Bothnian Sea). Meddelande från Kungliga Lantbruksstyrelsen (Communications from the Royal Agricultural Administration), 243: (In Swedish). Hessle, C The herrings along the Baltic coast of Sweden. Publications de Circonstance du Conseil International pour l Exploration de la Mer, 89: Hessle, C Sprat and sprat-fishery on the Baltic coast of Sweden. Meddelande från Kungliga Lantbruksstyrelsen (Communications from the Royal Agricultural Administration), 262(1): (In Swedish). Hoek, P. P. C., and Hjort, J Réunions d Édinbourg des Commissions A et B. (4-9 Septembre 1902). Commission A: Pour les problèmes de la migration des poissons. Rapports et Procès-Verbaux des Réunions du Conseil International pour l'exploration de la Mer, 1, Commissions Spéciales: Hoek, P. P. C., and Garstang, W Réunions d Édinbourg des Commissions A et B. (4-9 Septembre 1902). Commission B: Pour les questions de la diminution du rendement de la pêche. Rapports et Procès-Verbaux des Réunions du Conseil International pour l Exploration de la Mer, 1, Commissions Spéciales: Hoek, P. P. C., and Trybom. F Draft-answer to the questions regarding the salmon fisheries of the Baltic, presented by the Governments of Denmark, Finland, Russia and Sweden. Rapports et Procès-Verbaux des Réunions du Conseil International pour l Exploration de la Mer, 7, Appendix D: ICES Report of the Working Group on Assessment of Demersal Stocks in the Baltic. ICES CM 1978/J:3. 31 pp. ICES Baltic Salmon Assessment Working Group. ICES CM 1980/M:3. 35 pp. Ikonen, E., and Auvinen, H Results of stocking with Baltic salmon in Finland in ICES CM 1982/ M: pp. Ikonen, E., and Auvinen, H. 1984a. Migration of salmon in the Baltic Sea based on Finnish tagging experiments. ICES CM 1984/M:4. 32 pp. Ikonen, E., and Auvinen, H. 1984b. Migration of sea trout stocks in the Baltic Sea on the basis of Finnish tagging experiments. ICES CM 1983/M:5. 16 pp. Ikonen, E., and Auvinen. H Migration of salmon postsmolts (Salmo salar L.) in the Baltic Sea. ICES CM 1985/ M: pp. Johnson, T Dispersal area of perch, Perea fluviatilis, tagged in a stream flowing into the Bothnian Sea. Aquilo Serie Zoologica, 18: Järvi, T. H Fluctuations in the Baltic stock of salmon Rapports et Procès-Verbaux des Réunions du Conseil International pour l Exploration de la Mer, 106: Jönsson, N., and Biester, E Results of tagging experiments on the Rügen spring herring 1977/78. ICES CM 1979/J: pp. Khan. N. van, Drzycimski, I., and Chojnacki, J A contribution to the biology of sprat (Sprattus sprattus balticus Schneider) from the Bornholm Basin. ICES CM 1972/H:23. 9 pp. Khoziosky, S., Shvetsov, F.. and Uzars, D The dynamics of mortality components in Baltic sprat in Sub-divisions 26 and 28. ICES CM 1983/J:7. 6 pp. Kühlmorgen-Hille, G Infestation with larvae of Anisakis spec, as a biological tag of herring in Subdivision 22, Western Baltic Sea. ICES CM 1983/J:11. 7 pp. Kandier, R Untersuchungen über den Ostseedorsch während der Forschungsfahrten mit dem R.F.D. "Poseidon" in den Jahren (Research on Baltic cod during the survey of R.F.D. "Poseidon" in ). Ber. Deutsches Wissenchaftliches Kommission, Meeresforschung (Reports from the German Scientific Marine Research Commission), 11(2): (In German). Lamp, F., and Tiews, K Progress report of comparative tagging experiments on the Baltic cod. Rapports et Procès- Verbaux des Réunions du Conseil International pour l Exploration de la Mer, 166: Larsson, P. O., and Atheskar, S Laxsmoltens vandring från Lule-älven (Migration of salmon smolts from the Lule River). Fiskeritidskrift for Finland (Journal of Finnish Research), 23( 1): 8-9. (In Swedish). Lehtonen, H Gäddans vandringar i våra kustvatten (Migration of pike in our coastal waters ). Fiskeritidskrift for Finland (Journal of Finnish Fisheries), 17(3): (In Swedish). Lehtonen, H Biology and stock assessments of coregonids by the Baltic coast of Finland. Finnish Fisheries Research, 3: Lehtonen, H Stocks of pike-perch (Stizostedion lucioperca L.) and their management in the Archipelago Sea and the Gulf of Finland. Finnish Fisheries Research, 5: Lehtonen, H., and Toivonen, J Fresh-water fishes. In The Baltic Sea. Elsevier Oceanographic Series, 30: Ed. by A. Voipio. Elsevier Scientific Publishing Company, Amsterdam. 418 pp. Lehtonen, H., and Toivonen, J Migration of pike-perch, Stizostedion lucioperca L., in different coastal waters in the Baltic. Finnish Fisheries Research, 7: Lindquist, A Meristic and morphometric characters, year classes and races of the sprat (Sprattus sprattus). Report of the Institute of Marine Research, Lysekil, Series Biology, 17: Lindquist, A Contribution to the knowledge of the Baltic sprat, (Sprattus sprattus). ICES CM 1971 /H: pp. Lindroth, A., Larsson, P. O., and Bertmar, G Where does the Baltic salmon go? In Coastal Research in the Gulf of Bothnia, pp. 387^114. Ed. by K. Müller. Dr W. Junk Publishers, The Hague. 462 pp. Neill, S., and Cullen, J Experiments on whether schooling by their prey affects the hunting behavior of cephalopods and fish predators. Journal of Zoology, 172:

10 370 E. Aro Netzel, J Polish cod tagging experiments in the region of Slupsk Furrow in the years 1957/1963. ICES CM 1968/ F:7. 13 pp. Netzel, J Polish cod tagging experiments in the Baltic in 1969 and Rapports et Procès-Verbaux des Réunions du Conseil International pour l Exploration de la Mer, 166: Neuman, E Species composition and seasonal migrations of the coastal fish fauna in the southern Bothnian Sea. In Coastal Research in the Gulf of Bothnia, pp Ed. by K. Müller. Dr W. Junk Publishers, The Hague. 462 pp. Nordqvist, O Ein Vorschlag zum Programm der Fischereiuntersuchungen in der Ostsee. Annexe A: Rapports et Procès-Verbaux des Réunions du Conseil International pour l Exploration de la Mer, 1: (In German). Nordqvist, O Die Längenmasse von in der südlichen Ostsee gefangenen Lachsen und Meerforellen als Vorbereitung einer eventuellen Einführung von vereinbarten Mindestmassen dieser Fische. Rapports et Procès-Verbaux des Réunions du Conseil International pour l Exploration de la Mer, 7, II Spezielle Teil, Report 1: (In German). Otterlind, G Frän strömmingsmärkningarna pä ostkusten (Tagging of herring in the East coast). Ostkusten (East Coast), 3: (In Swedish). Otterlind, G Migration of flounder in the central Baltic. ICES CM 1965/Baltic Belt Seas Committee pp. Otterlind, G Migration of plaice and flounder in the southern Baltic. ICES CM 1967/F: pp. Otterlind, G Fish stocks and fish migration in the Baltic Sea environment. Ambio Special Report, 4: Otterlind, G Cod migration and transplantation experiments. ICES CM 1984/J: pp. Otterlind, G Cod migration and transplantation experiments in the Baltic. Zeitschrift für Angewandte Zoologie. (Journal of Applied Zoology), 1(1): Parmanne, R., and Sjöblom, V Recaptures of Baltic herring tagged off the coast of Finland in ICES CM 1982/Jr pp. Parmanne, R., and Sjöblom, V Recaptures of Baltic herring tagged off the coast of Finland in ICES CM 1986/J: pp. Popiel, J On the biology of the Baltic herring. Reports of the Sea Fisheries Institute, Gdynia, 19: Rannak, L., Arman, J., and Kangur, M Lohe ja meriforell (Salmon and sea trout). Valgus, Tallinn. 152 pp. (In Estonian). Rechlin. O Remarks on Baltic Sea sprat assessment in relation to stock identity. ICES CM 1986/J: pp. Schmidt. J Contributions to the life-history of the eel (Anguilla vulgaris Flem.). Rapports et Procès-Verbaux des Réunions du Conseil International pour l'exploration de la Mer, 5, Bericht über die Tätigkeit der Kommission C2 in der Periode Februar 1903-Märtz Spezielle Teil, Report 4: Seber, G. A. F The Estimation of Animal Abundance and Related Parameters. Charles Griffin and Co. Ltd., London. 506 pp. Segerstråle, C Återfångade märkta Ingä-gäddor ( Recaptures of pike tagged in Ingå). Finlands Jakt och Fisketidskrifter (Journal of Finnish Game and Fisheries), 26: (In Swedish). Sjöblom, V, Aro, E., and Suuronen, P Migrations, mortality and growth of cod in the northern Baltic Sea. ICES CM 1980/J:8. 15 pp. Strandman, M Monika SF 1517 (Tagged pike named Monika, tag number SF 1517). Fiskeritidskrift for Finland, Ny Serie (Journal of Finnish Fisheries, New Series), 8: (In Swedish). Strzyzewska. K.. and Popiel, J Characteristic of herring infested by Anisakis larvae caught in the Gulf of Gdansk. ICES CM'l974/H:15. 4 pp. Svärdson, G Spéciation of Scandinavian Coregonus. Reports of the Institute of Freshwater Research, Drottningholm, 57: Svärdson, G., and Fagerström, A Adaptive differences in the long distance migration of some trout, Salmo trutta. stocks. Reports of the Institute of Freshwater Research, Drottningholm, 60: Thurow, F Local movements of salmon in the Baltic Sea. ICES CM 1973/M:7. 5 pp. Toivonen, J Kuhan (Lucioperca lucioperca L.) vaelluksista, kasvusta ja kuolleisuudesta Suomenlahden saaristossa, Saaristome-ressä ja Ahvenanmaalla (On the migration, growth and mortality of pike-perch in the archipelago of the Gulf of Finland, Archipelago Sea and Åland Sea). Manuscript, University of Helsinki, Department of Zoology. 203 pp. (In Finnish). Trybom, F. 1907a. Proposal regarding an international treaty between the Baltic Countries. Rapports et Procès-Verbaux des Réunions du Conseil International pour l Exploration de la Mer, 7. Appendix D: Trybom, F. 1907b. Markierungen von Lachsen und Meerforellen im Ostseegebiete. Rapports et Procès-Verbaux des Réunions du Conseil International pour l Exploration de la Mer, 7, II Spezielle Teil, Report 2: (In German). Trybom, F. and Schneider, G Die Markierungsversuche mit aalen und die Wanderungen gekennzeichneter Aale in der Ostsee (Eel tagging experiments and migration of tagged eel in the Baltic Sea). Rapports et Procès-Verbaux des Réunions du Conseil International pour l Exploration de la Mer, 7, II Spezielle Teil, Report 3: (In German). Veldre.I Kilu (Sprat). Valgus, Tallinn. 199 pp. (In Estonian). Vitinsh. M Migration of flounder in the eastern Baltic. ICES CM 1972/F: pp. Vitinsh, M Some regularities of flounder (Platichthys flesu s L.) distribution and migrations in the eastern and north-eastern Baltic. Fischerei-Forschung, 1: Vitinsh, M Flounder stocks and fishery in the eastern and northeastern Baltic in Annales Biologiques du Conseil International pour l Exploration de la Mer, 32: Zarnecki, S., and Duszynski, J Migrations of sea trout from Pomeranian rivers (Results of tagging in 1960). ICES CM 1961/53. 5 pp.