Species or stocks of fish can be classified in several ways according to the nature of their movements, and to how their movements relate to the boundaries between national EEZs, and between EEZs and the high sea areas.
Although they are not virtually exclusive, three basic types of movements during the life of an individual fish can be usefully distinguished:
(1) Regular seasonal migrations
(2) Regular movements in accordance with the growth and development of the fish
(3) Random movements or dispersions
A further type of movement should also be borne in mind. This is the possible medium to long-term changes in the general location of the stock. As more series of data of individual stocks over long periods become available it is clear that the geographical position of a stock is not fixed. Sometimes the change can be easily related to changes in climate, e.g., a general warming may allow a stock to extend its distribution further towards the poles (the impact of climatic changes has been reviewed by Cushing and Dickson, 1976). Other changes, e.g., alteration in the dominant clupeoid species in the English Channel between herring and pilchard, may be more subtle. An important case is where a reduction in the population abundance is accompanied by a shrinking in its area of distribution. The California sardine, which at its peak was found as far north as southern British Columbia (Murphy, 1966) is a good example. In any case, though the changes may be easy or difficult to explain they are almost impossible to predict. The practical point is that the pattern in which a stock is shared between countries (or indeed whether or not it is shared at all) is not fixed, but can change over the years. For example, until recently the sardines off north-west Africa were confined to Morocco, but has now extended their distribution southward so that at least the more southerly stock is shared between Morocco and Mauritania.
This type of change in distribution is not considered further here, but must be borne in mind in setting up arrangements to deal with shared stocks.
The term migration has been used in many ways; a representative dictionary definition (Concise Oxford Dictionary) is “(of birds and fishes), come and go with the seasons”. This generally implies a systematic shift in the main concentrations of fish, often to the extent that there is little or no overlap between where the fish are in summer and in winter. The best known examples are of birds, both because they are easily seen and because they can move faster and further - the record being held by the Arctic terns that migrate between the Arctic and Antarctic. In the sea the movements of whales between the Antarctic, where they feed, and the warm tropical and sub-tropical waters, where they breed, provide good examples of clearly defined migration.
Another, if somewhat extreme example of long migrations, in this sense, as well as of the seasonal type of movements (those associated with the growth and development of the individual fish) is provided by the Atlantic bluefin tuna. This spawns in tropical areas, the juveniles grow up in sub-tropical and temperate waters, and the adults perform long-distance migrations, moving to summer feeding areas as far north as Norway and Newfound land, returning to spawn in the Caribbean and the Mediterranean.
Migrations require energy to perform. There must therefore be significant advantages in terms of, e.g., food supply, overall survival or recruitment success, to make the movements worthwhile. This probably implies marked seasonal changes in the food supply or other conditions in at least part of the migration path (though the movements of the fish might well be such as to keep itself in some preferred, and reasonably constant, conditions).
One might therefore expect that migrations would be longer and more marked where the seasonal differences are greatest, i.e., in temperate rather than in tropical or sub-tropical waters. One might further expect that the seasonal migrations would be predominantly north and south. To some extent this is true, e.g., the sardinella off north-west Africa move up and down the coast between southern Mauritania and Guinea-Bissau, but there are several other patterns of movement, among which inshore-offshore patterns are particularly common. In temperate regions the inshore waters can be highly productive in summer, and fish go there to feed, but are cold and unattractive in winter, and the fish retreat to deeper water. In the Indian Ocean the monsoon winds can in some places and seasons cause the bottom waters to become deficient in oxygen, so that the demersal fish move, either well offshore, or to very shallow waters close inshore. Other seasonal changes, e.g., in the depth of the thermocline, can cause similar inshore-offshore migrations.
Movements in accordance with the growth and development of the individual fish are at least as important as the seasonal migrations. Fish pass through a number of life-stages that are extremely different in their nature and requirements. In the first few months of life, as eggs and small larvae, most marine fish generally drift somewhat passively with the prevailing currents; the larvae of many species then settle in a nursery area (often in shallow coastal waters) until they reach a size at which they join the main adult fish stocks. For species with pelagic eggs or larvae this pattern must involve some active directed movement to counter-balance the drift in the larval stage (for example, adult cod move from the Spitzbergen/Barents Sea shelf to the spawning area round the Lofoten Islands, and the eggs and larvae drift back north in the extension of the Gulf Stream), but there may be more complicated patterns of movement.
All types of fish (including crustaceans and molluscs, other than those firmly attached to the bottom) make movements of the third type. The differences between species or stocks lie in the distances covered, and the degree to which the individual fish has some type of home territory, to which it will tend to return, or whether any directed movements tend merely to maintain the fish in favourable environmental conditions, irrespective of the precise geographical location.
The extent of the movements range from the scallops which may make no more than an occasional jump of less than a metre, to the tunas, whose continual active swimming may take them a hundred kilometres away from their starting point in a day or so. Compared with the knowledge of terrestial animals, little is known about the territorial habits of most fish. One might expect it to be better developed for fish living in close contact with permanent features such as are provided by rocky bottoms or coral reefs than for fish living in the open sea. However, even for these the sea is less featureless than might appear from a ship, so that the extent of their movements may be smaller than that of suitable conditions. The implication is that if, for example, a species is distributed continually along an extensive coast, the movements of any individual fish may well cover much less than the whole extent.
So far as management policies and the needs for collaboration between countries are concerned, the several patterns of movement can have important differences. In the simplest form of the second pattern each fish will, apart from the return of adult fish to the spawning grounds to complete the cycle, move through each area (nursery grounds, feeding area, spawning area) once, and only once, during its life. The fisheries in different areas interact in a clearly hierarchical manner. Fishing in any one area will (so far as a given brood of fish is concerned) only be affected by fishing “up stream” (i.e., on younger fish), and only affect those “downstream” on older fish. Interactions will be effective in the opposite direction (i.e., fisheries on older fish will affect those on younger fish) only to the extent that recruitment is affected by the abundance of adult fish. This is not always apparent, and (which is perhaps more significant in terms of getting agreement) is in virtually all cases difficult to demonstrate, at least until the recruitment has been seriously reduced. Nevertheless, prudent management of any stock should involve the maintenance of the minimum abundance of spawning fish, whether or not a relation with recruitment has been clearly demonstrated.
In contrast, seasonal migrations in their simplest form, for relatively long-lived fish, result in a more balanced type of interaction. Fishing on the wintering grounds may affect the summer fishery in the subsequent year, but they in turn will be affected by summer fishing in previous years. The incentives for an agreement for coordinated management of the resources are therefore greater here than in the other pattern, in which the country whose fisheries act “upstream” of the others may be less immediately inclined to seek agreement.
The impacts of the third type of movement - random, or at least a poorly defined dispersion - are less easy to describe simply. Where the movements are large compared with the distribution of the species or stock, then the movements achieve a good mixing within the species or stock. This means that a given level of removals by one group of fishermen will have an equal effect on all fishermen, wherever and whenever, within the distribution of that stock, the fishing is carried out. On the other hand, if the movements are small compared with the range of the species, then there will be little or no interaction between fisheries at the opposite extremes of the range, even though between them there is no barrier or point at which one can say there is a separation between stocks. For example, many species of demersal fish occur off the coasts of all the countries in the Gulf of Guinea, and move back and forth along the coasts (Champagnat and Domain, 1978). The distances covered in these movements are not well known, except for a few species and locations where tagging experiments have been done, but probably average not more than a hundred miles. This means that many fish will move between say western Ghana and eastern Ivory Coast, but not many across the whole width of the two countries, between eastern Ghana and western Ivory Coast.
The distinction between ordered migration and dispersion is not sharp. There are likely to be variations in the timing, distances and directions of migration between individuals so that an initially compact group of fish will become dispersed over a wide area. For example, in the Pacific skipjack tuna exhibit some clear patterns, e.g., as exhibited by the results of the tagging programmes of the Inter-American Tropical Tuna Commission and the South Pacific Commission, but some individual fish may move to quite different areas than the main bulk of fish tagged at the same time and place.
This pattern of dispersion raises problems in the definition and interpretation of the term ‘stock’. This term, unlike related terms such as ‘species’ or ‘population’, which have reasonably clear biological definitions, has a practical and operational basis related to the particular problem being faced - a ‘unit stock’ (or ‘stock’) is a group of fish that can or has to be treated as a ‘unit stock’ for some definite purpose. That is, there is no objective definition, valid under all circumstances, of what is meant by a unit stock - certainly no real definition will be attempted here. Rather, a unit stock is an empirical grouping of fish that is sufficiently large that, when analysing the data concerning it, or taking policy decisions about its exploitation and management, events on adjacent ‘stocks’ can be ignored, or at least treated in a different way to events within the stock. Equally, it should be sufficiently small to be for practical purposes homogeneous, and for possible divisions within the stock, into smaller groupings, to be ignored.
It should be clear from this that what is considered a stock is not an absolute, but can change, depending on the purposes, for convenience of analysis or policy making. Also there can be situations when there is no group of fish that can satisfy the conflicting requirements described in the previous paragraph. In these situations, the methods of analysis and approaches to policy making will have to be modified - usually in the direction of being more complicated and difficult to apply - to take account of the biological realities and particular conditions of the fisheries.
For some fish the concept of unit stocks works well. For example, the fact that individual Pacific salmon returns to spawn in the stream in which they were born means that the fish spawning in particular streams can be used as the basic bricks, to be taken together in larger or smaller groups to form stocks for different purposes. In international discussions between Japan on one side, and the U.S.A. and Canada on the other, it may be sufficient to treat salmon spawning in Asia and in North America as two stocks. Between the U.S.A. and Canada, stocks may be considered as groups of fish originating in each major river system, but at the national level, when applying detailed management measures, the separation of stocks may be taken down much finer, possibly to the individual stream.
The same may be true for other species which have a well-defined homing instinct (which may be few) and the more numerous species which have distinct and well-separated spawning areas, even when it is not clear that all fish return to spawn in their original areas. For example, the herring spawning in the Downs, at the southern extremity of the North Sea, form a distinct group, which mixes with the ‘Bank’ group, spawning further north, on the summer feeding grounds in the central and northern North Sea. Neither group seems to mix much with the herring spawning outside the North Sea. For different purposes the Downs herring, or all North Sea herring, can be treated as unit stocks.
For species with dispersed spawning, possibly throughout its range, the situation is much less clear. For skipjack in the Pacific, or croaker off West Africa, there does not appear any distinct groupings. For these species there may be for many purposes no group that can be treated as a simple unit stock. Rather there may be a group of fish or fishermen of primary interest, and other groups at increasing distances which will be of decreasing relevance. For example, in considering the croaker fishery of Ghana, considerable account should be taken of what is happening off Togo or eastern Ivory Coast, but very little about the fishery off Cameroon or Sierra Leone. Similarly anyone analysing the skipjack fishery of Papua New Guinea or considering its management needs to take great account of events in the Solomons, and probably some note of those around the Carolines orNew Hebrides, but can ignore anything happening off South America.
Species and stocks of fish can also be classified according to the relation of their movements to national boundaries. This clearly depends as much on the pattern of the boundaries, as on that of the movements - long movements may still lie wholly within a large nation's EEZ, while elsewhere a short movement may be enough to cross a boundary between one EEZ and another. The classification of stocks in relation to boundaries will therefore be somewhat different to that based on purely biological considerations. A possible classification is as follows:
(1) Stocks lying wholly within a single national jurisdiction.
(2) Stocks occurring within two or more EEZs, with movement across boundaries, but no clear migratory pattern.
(3) Stocks occurring within two or more EEZs, and a clear pattern of movement (seasonal migration or movement according to developmental stages) between one zone and another.
(4) Stocks occurring wholly on the high seas beyond national jurisdiction.
(5) Stocks occurring both on the high seas and in one or more national jurisdiction.
These categories are not absolute, and groupings and sub-divisions can readily be suggested. For example, group (5) might be divided, in the same manner as groups (2) and (3) are distinguished, between stocks with clear migrations, either seasonal or according to developmental stages (e.g., southern bluefin tuna, with its spawning grounds in the southeast part of the tropical Indian Ocean; nursery grounds in the coastal waters off Australia; and feeding area in the temperate oceanic waters from southern Africa to New Zealand), and those whose movements, so far as they are known, are in the nature of random dispersal (possibly skipjack). However, the present grouping is convenient; some characteristics of each group are discussed below.
From the point of view of this report these are the least interesting stocks. The study and management of these stocks is purely a matter for the state in whose waters the stock occurs. Nevertheless, it must be recognized that this group is extremely heterogeneous. It contains many small resident stocks, but also many stocks whose movements are by no means negligible but happen to be less extensive than the national limits (e.g., most fish stocks off Australia, haddock around Iceland, etc.). It should also not be assumed that a stock necessarily falls into this group merely because present catches are limited to a single national EEZ. Many of the species, especially molluscs and crustaceans, which are partially or wholly sedentary in the exploited phase are mobile at early stages of their life history, even if it is a matter of drifting in the currents. The success of a fishery for scallops or lobster may therefore depend on the conditions on spawning grounds some distance “up-stream”, including the abundance of spawners, and be affected by the amount of fishing on those grounds. In such cases the stocks concerned should be more properly considered in group (3).
Some movement across boundaries is certain to occur except when the boundary coincides with a natural barrier. For example, between island states, or between countries facing each other across straits, there is often deep water that is an effective barrier to many adult fish, especially bottom-living species. In such cases the fish round each island may form separate stocks. On the other hand, if two countries lie adjacent to each other along the edge of the continent, and the same species occurs in both countries, then it can be taken as virtually certain that there is some interchange across the boundary, even in the absence of any systematic migratory movement.
This does not mean that the two countries share a single stock. Because of their lack of systematic movement many of the species falling in this category are likely to form a number of stocks. Often there are no well-defined distinction between one ‘stock’, e.g., in one bay, and the next. Many inshore species of fish and crustaceans will fall into this category. For example, throughout the whole of south and southeast Asia, the small-scale fishermen exploit very much the same group of species, the differences between one district and another being mainly of a matter of local ecological conditions (nature of the bottom, etc.). Across most boundaries of national EEZs, e.g., between Thailand and Malaysia, or India and Bangladesh, there will be interchange, but the fish concerned will probably not move far. Thus the effects of events on one side of the boundary are not likely to extend more than perhaps a few kilometers into the adjoining EEZ. It is perhaps useful to use a distinct term for this category of stocks; the term ‘transboundary stock’ is suggested. A feature of these stocks is that the relations tend to be symmetrical, i.e., conditions tend to be similar either side of the boundary, and movements of fish across the boundary in one direction are likely to be about equal to movements of fish, in similar numbers and of similar sizes, across the boundary in the opposite direction. An exact balance of movements across the boundary will depend on the fisheries on each side being similar, as well as the fish. This will often be the case, but sometimes there may be an abrupt change in social or economic factors between adjacent countries sharing a marine boundary. This may give rise to differences in fishing methods and/or intensity, and hence a net flow of fish from the lightly fished area into the more heavily fished. This general and biological symmetry may be contrasted with the lack of symmetry that often occurs in the following category.
This category includes a large number of stocks, possibly even the majority of the stocks supporting large-scale industrial fisheries. Within this large group the details vary greatly, particularly in regard to the proportion of the total stock that moves across the boundary, and the extent of the differences in the composition or behaviour of the stock on either side of the boundary.
The ordered movements of this category imply almost inevitably some differences in conditions at either end of the migration path. Hence the effects of fishing on the stock at one place rather than another, and the benefits that can be gained from such fisheries, are not likely to be equal. This is clearly the case for movements associated with growth and development of the fish – from nursery area, to adult feeding area, to spawning ground. The literature on the dynamics and management of fish stocks is full of studies showing the effects of varying the ages at which fish are caught. These show that there is a loss in the yield by catching fish before the growth rates begins to slow and to be balanced by losses due to natural mortality. This is likely to cause problems when the nursery grounds lie in one EEZ and the adult stock in another. For example, in the northeast Atlantic part of the nursery areas of the stock of cod spawning at the Lofoten Islands in northern Norway lie in the EEZ of the Soviet Union (Dannevig, 1954; Maslov, 1944). Though older fish also occur in this zone the catches there tend to be of fish smaller than the optimum size at first capture. Again Champagnat and Domain (1978) have shown that along the West African coast from Mauritania to Guinea several of the demersal species have distinct areas of distribution of adults and juveniles, often in different countries.
Some distinction between the areas inhabited by different stages in the life history exists in nearly all fish. Very often the main difference is that the younger stages are found inshore in shallow water (often brackish waters). The movements from the nursery or juvenile grounds to those inhabited by the adults thus tends to be right angles to the coast. They are therefore generally parallel to the boundaries between natural EEZs, rather than across them. On the other hand, river estuaries are important nursery areas for many species, and are also frequently the boundaries between states. This increases the likelihood that movement offshore from the nursery grounds will take the fish across the boundary between national EEZs. On the whole, the impact of this type of movement on the degree of interaction between fisheries in different EEZs is somewhat less than its frequency might suggest, but it certainly is significant for some stocks. For example, when the shrimp stock that has its nursery grounds in the Casamance area of Senegal moves offshore to the adult grounds it enters the EEZ of Guinea-Bissau. Often this has the same effect as some degree of random dispersal; that is, if both inshore nursery areas and off-shore adult grounds occur in an EEZ of two adjacent coastal states, it cannot be expected that, in their movement offshore, the fish will remain precisely in their original EEZ, but that there will be some mixture.
Seasonal movements of adult fish (“migration” in the stricter sense) is somewhat less general, being more common in temperate waters than in tropical waters with less marked seasonal changes. Movements are often north and south, tending to keep in somewhat similar conditions, e.g., of water temperature, and are commonly parallel to the coast. A higher proportion of the stocks moving in this way will therefore cross national boundaries. For example, off northwest Africa most of the important pelagic species (sardinella, mackerel, etc.) move across two or three EEZs during the year. During these movements there can be big changes in the behaviour and condition of the fish. They may disperse during the feeding season, and then congregate in dense - and easily caught - schools just before and during spawning. At the end of the feeding season they will be fat, in good condition and likely to fetch a high price, while after spawning they may be thin and worth little. Thus the attractions for the fishermen will not be the same in all parts of the range; in the extreme it may only be worth fishing at certain times and places. For example, in the northeast Atlantic there is a large stock of blue whiting to the west and north of the British Isles. During the feeding season it moves north and occurs widely in the EEZs of Norway and the Faroe Islands, but at this time the fish are so dispersed that the catch rates are not high enough to support an economic fishery. During the spawning season the fish move south into the British EEZ, and form concentrations that can support fisheries for both consumption and meal.
