NORTH EAST ATLANTIC - ATLANTICO NORDORIENTAL

THE DECLINE IN HERRING AND MACKEREL AND THE ASSOCIATED INCREASE IN OTHER SPECIES IN THE NORTH SEA by

Rodney Jones

DAFS Marine Laboratory
P.O. Box 101
Victoria Road
Torry, Aberdeen
Scotland, U.K.

Resumen

Han habido cambios marcados en la composición por especies en el Mar del Norte, asociados con una disminución de los desembarques de sardineta, anguila, platija y varias especies de gadidos. Hay también una evidencia de cambios en la abundancia de varias especies de invertebrados, y en particular de una extensa disminución de la biomasa total de copépodos en el período de 1948 a 1970.

En conclusión parece que:

a)    La disminución de caballa estuvo asociada con el desarrollo de la pesquería Noruega con red de cerco.

b)    La disminución de arenque pudo ser debida inicialmente a factores naturales, pero es probable que la pesquería también haya jugado un rol importante partícularmente en años recientes.

c)    La distribución de varias especies de peces durante sus períodos de alimentación pelágica son tales que las sobreposiciones principales ocurren entre anguilas y caballa por un lado, y entre arenques y platijas, y arenque y sardineta por el otro lado.

d)    La disminución de biomasa de caballa y el aumento de los desembarques de anguila son ambos comparativamente eventos recientes. La caballa disminuyó principalmente entre 1965 y 1970 y los desembarques de anguila aumentaron después de 1970. Parece que hubo un aumento de la dominancia de anguila, y que esto puede haber estado asociado con una reducción de la predación por parte de caballa.

e)    La disminución de arenque y el total de copépodos y el aumento de platija son eventos de largo tiempo que comenzaron como a fines de los años 1940s sino antes. No es fácil ver como estos cambios pudieron todos ser debidos a la pesca inicialmente, aun cuando la pesca ha jugado probablemente un rol importante en reducir la biomasa de arenque a un nivel relativamente muy bajo hacia los años 1970s.

f)    En el caso de sardineta y gadoidos es difícil decir hasta que punto el aumento de los desembarques fue debido a la disminución de arenque y caballa.

INTRODUCITON

During the 1960s there was a decline in the landings of herring and mackerel, and an associated increase in the landings of various other species. The object of this paper is to examine the nature of these changes and the factors which might have caused them.

Changes in the species composition of the landings

Table 1 shows the landings of selected species from the North Sea for the period 1960-1980. Over this period, herring and mackerel declined in importance, while sprat, sandeel, Norway pout, cod, haddock, whiting, saithe and plaice landings all increased.

Species that declined

The principal species that declined in the North Sea were herring and mackerel. Herring, started to decline in the 1940s, due to a decline in the Downs Stock which is the most southerly of the three N. Sea herring stocks. This stock has remained at a very low level until very recently. In the 1950s there was a decline in the central area (the Dogger stock). Finally, the northerly stock (the Buchan stock) declined in the 1960s (Burd, 1978).

Regarding mackerel, there was a very considerable decline in stock size between 1965 and 1970, associated with the development of the Norwegian purse-seine fishery. Since then the stock has declined further due to poor recruitment.

Species that increased

Species for which landings increased, were sprat, sandeel, Norway pout, cod, haddock, whiting and saithe and plaice. For all these species, landings increased to relatively high levels by the end of the 1960s, or at various times during the 1970s (Tables 1 and 2).

TABLE 2

Years in which landings attained maximum values

The largest increases were in the landings of sprat, sandeels, Norway pout and saithe. Sprat landings increased from less than 31,000 tonnes in 1962, to a maximum of 418,000 tonnes in 1975. Sandeel landings increased from about 100,000 tonnes at the beginning of 1960s to a maximum of 789,000 tonnes in 1978. Norway pout landings increased from 43,000 tonnes in 1974. Saithe landings increased from less than 50,000 tonnes at the beginning of 1960s to 253,000 tonnes in 1976. Lesser increases also occurred in cod, haddock and whiting during the 1960s. Cod and whiting landings increased by about a factor of 2 and haddock landings by a factor of 3-4. All eight species attained high levels of landings at various times between 1969 and 1976, and all, except cod, sandeels, and possibly plaice, showed a tendency for landings to decline during the second half of the 1970s.

Magnitudes of the changes in stock size

Table 3 contains estimates of stock biomass for those species for which useful estimates can be made. For herring, total North Sea spawning stock biomass declined from just under 2 × 10⁶ tonnes at the beginning of the 1960s to 350 × 10³ tonnes at the end of the 1960s.

(1) From ICES Working Group Reports and Anon, 1974
(2) Includes Division VIId
(3) Includes Division IIIa
(4) Includes discards and landings by industrial fishing (not available for cod prior to 1963)
(5) Excluding by catch from industrial fishing - includes Division IIIa

Table 3
The adult stock sizes of major species in the North Sea

Species	Biomass of Stock 000's of tonnes
	Early 60's (period)	Late 60's (period)	Age Groups	M(1)
Herring	1840	350	2-8	0.1
	(1959-61)	(1968-70)
Mackerel	about 3000	800	2-7	-
	(early 1960)	(1968-70)
Cod(3)	192	367	2-12	0.2
	(1963)	(1968-70)
Haddock(3)	118	597	2-8	0.2
	(1959-61)	(1968-70)
Whiting(3)	102	217	2-8	0.2
	(1959-61)	(1968-70)
Saithe	193	917	2-14	0.2
	(1960-62)	(1971-73)
Plaice	390	477	2-20	0.1-0.25(2)
	(1959-61)	(1970-72)

(1) Value of M assumed for determining stock size
(2) Values of M taken at 0.25 and 0.20 for 2 year old males and females and 0.15 and0.10 for older males and females respectively
(3) Based on data in Anon, 1976

It is interesting to note that recently, there have been indications of a recovery of the Downs stock.

For mackerel, spawning stock biomass declined from about 3 × 10⁶ tonnes to about 800 × 10³ tonnes over the same period.

Haddock, whiting and cod stocks appeared to reach high levels at the end of the 1960s and beginning of the 1970s, and stock sizes have been compared for the periods 1959-61 and 1968-70. For these three species, estimates have been taken from Jones, 1982, based on data from an early report of the North Sea Roundfish Working Group (Anon 1976). More recent reports of this Working Group have reviewed these biomass estimates upwards to take account of fish discarded at sea. Due to uncertainties about rates of discarding prior to the late 1970s however, these more recent data are more valuable for estimating current levels of biomass, then for comparing changes in biomass in the long term. Based on the results in Table 3, cod and whiting stocks increased by about 100% and the haddock stock by about 400% during the 1960s.

Estimates for saithe and plaice have been taken from more recent reports of the ICES Saithe and Flatfish Working Groups. Saithe biomass reached a maximum early in the 1970's. Comparison of stock sizes for the periods 1959-61 and 1971-73 showed an increase of about 375% (Table 3). The plaice story differs somewhat from that of the gadoid species in that increases in landings and stock size first became evident as early as the 1950s. By the 1960s and early 1970s, stock size was levelling out and the increase between the periods 1959-61 and 1970-72 was only 22%.

Variations in recruitment

For all species for which data is available, increases in stock size and/or landings have been associated with strong recruit year classes. Table 4 shows recruitment indices for those species for which data are available.

(1) Millions of fishing at age 1 year from VPA
(2) Abundance indices from Scottish autumn surveys in north western North Sea
(3) Abundance indices from International Young Fish Surveys

For example, in the case of cod, haddock, whiting and saithe, the increases in stock sizes were all associated with the appearance of strong year classes. For cod there were strong year classes in 1969, 1970 and 1974. For haddock, the year class of 1962 and 1967 were the two strongest on record and those of 1973 and 1974 were also strong. For whiting there were strong year classes in 1962, 1967 and 1972. For saithe there were strong year classes in 1966, 1967, 1968 and 1973.

For plaice, the recruitment indices have remained at a consistently high level from 1963 to 1980.

Recruitment indices for Norway pout (Table 4) appear to be less reliable than those for the other species above. Available data however, suggests strong year classes in 1967, 1971 and 1973. There is only a short series of recruitment indices for sandeels, but this does provide evidence of strong year classes in 1976 and 1977, immediately prior to the very high landings in 1977 and 1978.

In summary, cod, haddock, whiting, saithe, plaice, sandeels and Norway pout all show evidence of increases in landings associated with strong year classes.

There are no recruitment indices for sprat over the period when landings first increased in the early 1970s and all that can be said for this species is that there is no reason to suppose that the increase in landings was not due to the same factors as were associated with the increases in other species.

Regarding the two species that showed decreases in stock size, both herring and mackerel exhibited relatively low levels of recruitment in the 1970s, at a time when the adult stock sizes were relatively very low.

Geographical distributions of the principal species

Opportunities for species interaction should be greatest on those occasions when species occupy the same geographical area, at the same time, and when they are feeding on the same kind of food. Although some of the species concerned are pelagic, whereas others are demersal, all are pelagic feeders at some time or other in their lives. Herring, mackerel, sprat, sandeels and Norway pout are pelagic feeders throughout their lives, whereas plaice and gadoid species feed pelagically as 0-group fish during the summer of their first year of life. It is appropriate therefore to consider the geographical distributions of herring, mackerel, sprat, sandeels, Norway pout and 0-group gadoids during the summer months. This has been done, based on data in Anon (1978) and Walsh (1974), and a summary diagram showing the main centres of density of these groups is given in Figure 1.

During the summer, the main centres of density of herring are in the southern North Sea, south of latitude 54° north, and between latitudes 54° and 60° north, west of 2° east longitude. Mackerel on the other hand are found mainly in the northern North Sea, and in the central North Sea to the east of 2° east longitude. Although there is an overlap in the distributions of these two species, there is a clear distinction between their main centres of density.

Sprat and sandeels also show a separation of their main centres of density. In general, it appears that sprat distribution primarily overlaps with that of herring, whereas sandeel distribution overlaps more extensively with that of mackerel and juvenile herring.

Norway pout are found principally in the northern North Sea, north of latitude 58°30' north, and overlap with the northerly limits of the herring and mackerel summer feeding areas.

0-group haddock are found principally north of latitude 59° north, but extended a little further to the south in the eastern part of the North Sea.

0-group whiting are found in the northern North Sea and in the south-eastern North Sea and also inshore along the east coast of the UK.

Figure 1 Locations of centres of density of principal pelagic feeding stage during the summer.

Figure 2. Annual variations in herring spawning biomass and sprat landings in the North Sea.

Figure 3. Annual variations in mackerel spawning biomass and sandeel landing in the North Sea

Opportunities for species interactions

Species that overlap, and hence that have an opportunity for interaction, can be grouped as follows:

(a)   Herring and sprat. It can be noted that maximum sprat catches in the mid 1970s coincided with the period of minimum herring stock size (Fig. 2)

(b)   Mackerel and sandeel. Here, it can be noted that there was a significant increase in sandeel landings the whole of the 1970s, at a time when the mackerel stock size was at a very low level (Fig. 3).

(c)   Plaice and herring. The main concentrations of plaice, in the southern and south-eastern North Sea overlap with a major herring nursery area and with the region occupied by the Downs stock of herring. It can be noted that the increase in plaice landings became particularly noticeable during the 1950s at a time when the Downs herring had dropped to a relatively low level.

(d)   Gadoids, herring and mackerel. The young stages of cod overlap principally with mackerel, those of whiting and Norway pout principally with herring, and those of haddock with both herring and mackerel. Out of the four gadoid species, the largest increase was in haddock, ie with a species that might have benefitted from decreases both in herring and in mackerel. Saithe also showed a large increase, but detailed information about its distribution is not available.

CAUSES OF THE CHANGES IN SPECIES COMPOSITION

There are two principal factors that could have accounted for the changes in species composition. One is exploitation and the other is natural variation.

Exploitation

Exploitation has almost certainly played a large part in the decline of the herring and mackerel stocks. In the case of mackerel, in particular, the considerable decline in stock size between 1965 and 1970 was clearly associated with the development of the Norwegian purse-seine fisheries.

In the case of herring, the decline has taken place over a relatively long period, starting with a decline in the Downs stock in the 1940s followed eventually by a decline in the Buchan stock in the 1960s. For both species, recruitment was exceptionally low during the 1960s and this was the direct cause of the very low stock levels attained by the end of the 1970s. What is not clear is whether this poor recruitment was due to natural causes, or whether it was due to the low levels of spawning stock brought about by exploitation.

In the case of sprat, sandeel and Norway pout, landings could have been influenced by the diversion of effort from the industrial fisheries for juvenile herring and whiting.

Regarding the species that increased in stock size, consideration of the possible effects of exploitation depends on the trophic interactions between those species/stocks that happen to overlap in space and time. Figure 4(a) and (b), for examples, shows the simplest kind of situation in which the herbivores are grazed by two groups of species, A and B. Here, exploitation is assumed to decrease the biomass of group B, leading to a reduction in grazing pressure on the herbivores. This, in turn, could make more food available for other species and might be expected to lead to an increase rather than decrease and this, therefore might be representative of trophic interactions in the north western North Sea, where copepod biomass does not appear to have declined. This has been an important summer feeding area for herring, and the decline in this species might have made more food available for other pelagic feeders in the same area ie principally for sprats and for the young, pelagic feeding stages of gadoids.

Figure 4 Some TRophic Interactions - I
(a) without exploitation - Two groups, "A" and "B" assumed to be grazing on herbivores;
(b) suggests that with exploitation of group "B", more food should be available for group "A". Linkely outcome should be increase in biomass of both herbivores and of group "A".

Figure 5 Some Trophic Interactions - II
(a) without exploitation - Two groups, "A" and "B" assumed to be grazing on herbivores. In addition, group "B" species prey on group "A"
(b) suggests that with explotation of group "B". Group "A" should benefit from reduced predation. Posible outcome could be increase in group "A", but decrease in biomass of herbivores

In the central and southern North Sea, however, there has been decline in total copepods and to account for this, something along the lines of the scheme in Figures 5(a) and (b) would seem to be more appropriate. Here it is assumed that the group B species prey on the group A species as well as on the herbivores. In this situation, it is possible for a reduction in the biomass of group B to lead not only to an increase in group A but also to a decrease in the herbivore biomass. This model is representative of a situation in which the principal exploited species is an important predator on other primary carnivores. It might be applicable therefore to mackerel (group B) and sandeels (group A) or to mackerel and euphausids. (In this connection one can note that there was an increase in euphausids in the 1970s in the eastern part of the central North Sea, at a time when mackerel biomass was low (Lindley 1979) as well as the increase in sandeels noted in connection with Fig. 3).

An over-exploitation of herring and mackerel might therefore have contributed to increases in sandeels, sprats and gadoids. All of these could have benefitted from an increased availability of food.

The Contribution of natural processes to the change in species composition

The principal difficulty with exploitation as the primary causal factor for all of the changes in the North Sea is that it does not explain the time sequence of events. In particular, the decline in total copepod biomass appeared to start in the 1950s, whereas the decline in mackerel biomass, which might have caused it, did not occur until after the mid 1960s. The post-war decline in total copepods is paralleled much more closely in fact by the decline of herring. According to Glover et al. (1972), the decline in copepods was particularly marked in the case of Pseudocalanus and, according to Cushing (1980), this is taken in preference to Calanus by larval and post larval herring. This, suggests therefore that the decline in herring was more likely to have been due to the decline in copepods, rather than viceversa.

The possibility of natural variability cannot be ruled out, and there is considerable evidence to show that there have been long term changes in many components of the ecosystem. For example, continuous plankton records, reported in various papers from the IMER, Plymouth laboratory, provide considerable evidence of long term changes in the zooplankton and phytoplankton of the North Sea. According to Glover et al. (1972), the total number of copepods and the zooplankton biomass in the North Sea have declined over the period 1948-69. Also the duration of the zooplankton season has fluctuated considerably, and has become progressively shorter over the period 1952-64. Regarding particular species, Colebrook (1978) concludes that Pseudocalanus declined from 1948 to about 1970 in all parts of the North Sea except for the northwestern North Sea, where it fluctuated about a constant level. To the west of the British Isles also it tended to decline.

Between 1970 and 1972, Calanus finmarchicus continued to decline in abundance, in contrast to Calanus helgolandicus in the southern North Sea, which did not show any systematic change in abundance (Colebrook et al., 1978).

From 1972 to 1977 in the eastern North Sea, euphausids increased in abundance in continuous plankton records (Lindley, 1979). The increase affected all five species, which normally occur in the area, but was not paralleled in other areas of the North Sea.

There have also been changes in the abundance of some fish larvae in the plankton. For example, over the period 1948-72 there has been a decline in the abundance of whiting, Norway pout, dabs and plaice larvae. The downward trend of whiting larvae started in 1964 whereas that for plaice has been general over the whole 25 year period (Coombs 1975). For whiting and sandeels there has been a shortening of the season during which the larvae have been caught and a delay in the time of the seasonal maximum abundance (Coombs 1975). According to Coombs and Mitchell (1981) there was a southerly shift in the distribution of mackerel larve in the North Sea over the period 1948-1977. Over a similar period, the abundance of larvae increased to reach high numbers by the late 1950s and subsequently declined after the mid 1960s.

According to Robinson & Jonas (1981), zooplankton abundance was still below average in most areas of the North Sea in 1980. In particular, numbers of Pseudocalanus elongatus, Paracalanusssp., Temora longicornis, Evadne nordmani and pelagic molluscs were well below the long-term mean.

There is also evidence, from the same source, of changes in the phytoplankton. Reid (1977), for example, has concluded that in most areas of the northeast Atlantic Ocean, diatoms have declined in abundance in the last decade, while at the same time there has been an increase in phytoplankton colour. This refers to an unidentified green colouration of phytoplankton silks that may be due to the presence of organisms such as micro-flagellates, which may partially disintegrate in formalin leaving their chloroplasts in the silks. These would not be identifiable in the continuous plankton recorder samples. In particular, Reid (1975), referring to phytoplankton in the western part of the central North Sea, notes that before 1965, phytoplankton colour showed distinct spring and autumn blooms. After that date the productive season became longer, the distinction between the two blooms became less marked, and the autumn bloom declined in intensity.

The diatoms decreased markedly after 1965 and this was associated with a drastic decline in the autumn bloom. Winter and summer diatoms almost disappeared and there was also a decline in the spring bloom.

The time of the spring bloom has also changed. Glover et al. (1972) state that for the period 1948-69, the spring bloom of phytoplankton has occurred progressively later.

Regarding particular species of phytoplankton, Colebrook et al. (1978) examined fluctuations in the abundance of seven species of Ceratium that occur regularly in samples from the North Sea. All but one species showed an advance in seasonal timing in the south and eastern parts of the North Sea. None of the species showed any consistent change in seasonal timing in the northwestern part of the North Sea.

Some long term trends have also been noted in adult fish characteristics. According to Jones & Hislop (1978) for example, the mean lengths of two and three year old haddock in the North Sea, increased from the mid 1920s to reach relatively high levels by about 1960-1962. Jones (in press) shows that this increased growth rate appeared to be associated with the period when haddock first become demersal and prior to the attainment of a length of about 28 cm. An implication is that feeding conditions for juvenile haddock at the first demersal feeding stages, improved progressively for nearly 40 years, but that this was not associated with unusually strong recruitment, until the very good year classes of 1962 and 1967 appeared.

There is abundant evidence therefore to show that variability has occurred at various trophic levels, and often over considerable time periods.

REFERENCES

Anon. 1976. Report of the North Sea Roundfish Working Group. ICES CM/F:9 mimeo.

Anon. 1978. The biology, distribution and state of exploitation of shared stocks in the North Sea. ICES Coop.Res.Rep. 74: 81 p.

Burd, A.C. 1978. Long-term changes in the North Sea herring stocks. Rapp.P.-v.Réun.Cons.int. Explor.Mer. 172:137-153.

Colebrook, J.M. 1978. Continuous plankton records: zooplankton and environment, north-east Atlantic and North Sea. Oceanol. Acta 1:9-23.

Colebrook, J.M., P.C. Reid and S.H.Coombs. 1978. Continuous plankton records: a change in the plankton of the southern North Sea between 1970 and 1972. Mar.Biol.45:209-213.

Coombs, S.H. 1975. Continuous plankton records show fluctuations in larval fish abundance during 1948-1972. Nature, Lond. 258:134-136.

Coombs, S.H. and E.E.Mitchell. 1981. Long-term trends in the distribution, abundance and seasonal occurrence of larvae of mackerel (Scomber scombrus L.) around the British Isles, 1948-1978. J.Mar.Biol.Ass.UK. 61:343-358.

Cushing, D.H. 1980. The decline of the herring stocks and the gadoid outburst. J.Cons.int.Explor. Mer. 39(1):70-81.

Glover, R.S., G.A. Robinson and J.M. Colebrook. 1972. Plankton in the North Atlantic - an example of the problem of analysing variability in the environment. pp. 439-445. In Marine pollution and sea life (M. Ruivo, ed.). Fishing News Books, London.

Jones, R. 1982. Species interactions in the North Sea. In Multispecies approaches to fisheries management advice. (M.C.Mercer, ed.). Can.Spec.Publ.Fish.Aquat.Sci. 59.

Jones, R. and J-R-G- Hislop. 1978. Changes in North Sea haddock and whiting. Rapp.p.-v.Réun.Cons.int.Explor.Mer. 172:58-71.

Lindley, J.A. 1979. Continuous plankton records: an increase in the abundance of euphausiids in eastern central North Sea. ICES CM/L:25. mimeo.

Reid, P.C. 1975. Large-scale changes in the North Sea phytoplankton. Nature, Lond. 257:217-219.

Reid, P.C. 1977. Continuous plankton records: changes in the composition and abundance of the phytoplankton of the north-eastern Atlantic Ocean and North Sea 1958-1974. Mar.Biol. 40:337-339.

Robinson, G.A. and T.D. Jonas. 1981. The continuous plankton recorder survey: plankton around the British Isles in 1980. Annls. Biol.

Walsh, M. 1974. The distribution of adolescent mackerel in the North Sea. ICES CM/H:32. mimeo 15p.

NORTH EAST ATLANTIC - ATLANTICO NORDORIENTAL

RECENT HISTORY OF ATLANTO-SCANDIAN HERRING STOCKS

by

Erling Bakken

Institute of Marine Research
Bergen, Norway

Resumen

El término "Arenque Atlanto-Escandinavo" (Clupea harengus L.) cubre tres stocks: "los desovantes de primavera de Noruega", "los desovantes de Primavera de Islandia" y los desovantes de verano de Islandia". Los desovantes de primavera de Noruega constituyen el stock más grande. El presente trabajo por lo tanto se concentra en este stock, pero también se refiere a los stocks de Islandia en un intanto de resumir los cambios que siguieron la interacción entre la biología y la pesquería durante los últimos treinta años.

En general los tres stocks de arenque Atlanto-Escandinavo han tenido historias pararelas desde 1950 hasta 1980. La primera década fue caracterizada por stocks estables o en aumento. La biomasa de los stocks desovantes de la población que desova en primavera en Noruega fue en promedio de 8.5 millones de toneladas, mientras que la biomasa de los desovantes de invierno y de verano de Islandia era de alrededor de 500 000 y 135 000 toneladas, respectivamente. Estos tres stocks en total rindieron capturas anuales de 1 millón de toneladas approximadamente. Durante la década siguiente todos los stocks declinaron rápidamente. Las capturas altas fueron mantenidas en los primeros años por el incremento en el ritmo de pesca, pero a inicios de los años 1970s los stocks y las capturas habían alcanzado su nivel más bajo. En esta época se introdujeron vedas de pesca y sistemas estrictos de cuotas de captura en un intento para lograr la recuperación de los stocks.

En la última década se ha observado una recuperación de dos de estos stocks, pero el ritmo de incremento ha sido diferente. En el momento, la biomasa del stock desovante de primavera de Noruega llega solamente a un 5% del promedio que ese stock tenía en los años 1950s, mientras que los desovantes de verano de Islandia han alcanzado los niveles altos observados entre 1958-1962. Los desovantes de Islandia de primavera, por otro lado no han mostrado signo de recuperación.

Las estadísticas de captura y registros de la pesquería de arenque Atlanto-Escandinavo a partir de los años 1800s indican que previamente hubieron amplias fluctuaciones de los stocks. Esto puede sugerir que la reducción de los stocks en los años 1960s fue inducida por factores bióticos. Sin embargo, se ha demostrado claramente que la rápida reducción de los stocks fue causada por una pesquería intensa, la que directa o indirectamente ha resultado en una reducción de reclutamiento y de aquí el total colapso de los stocks.

Este impacto de la pesquería fue el resultado de las grandes mejoras tecnológicas en la flota de los barcos pesqueros con redes de cerco a fines de los años 1950s y 1960s. Algunas innovaciones como por ejemplo los equipos acústicos para detección de peces, redes más grandes de fibra sintética y absorbentes para el pescado, combinadas con un cambio del método de pesca con dos pequeños botes hacia el uso de poleas hidraúlicas y la técnica de pesca con garetas hicieron que la flota fuera extremadamente eficiente inclusive en mar abierto.

A inicios de los años 1960s eran alrededor de 700 los barcos con redes de cerco que pescaban casi exclusivamente arenque Atlanto-Escandinavo. Hacia el final de la década la flota consistía de alrededor de 450 barcos con redes de gareta totalmente modernizados los cuales tuvieron que transferirse sucesivamente de la pesca arenque Altlanto-Escandinavo, que ya casi no existía, hacia la pesca del arenque del Mar del Norte, la caballa y el capelin.

Durante los años 1970s la mortalidad por pesca de los adultos en todos los stocks aumentó desde un nivel de 0.15 hasta 1.0 aproximadamente. El incremento casi exponencial de la mortalidad por pesca fue el resultado del incremento de la capturabilidad asociado con la reducción del tamaño de los stocks, un fenómeno típico de pesquerías dirigidas hacía peces que forman cardúmenes.

Algunas características biológicas de los stocks tales como la distribución, migración, desove y crecimiento también cambiaron, en conexión con los cambios de la abundancia. En los años 1950s los individuos adultos del stock desovante de primavera de Noruega se alimentaban entre Islandia y Jan Mayen, pasaban el invierno al este de Islandia y migraban hacia el este hacia las extensas áreas de desove a lo largo de la costa oeste de Nouruega. A medida que el stock adulto se redujo en los años 1960s el área de distribución se desplazó hacia el norte con nuevas áreas de alimentación y de invernación, y áreas de desove más reducidas hacia el norte. A inicios de los años 1970s no se observó desove, y en los últimos años los desovantes de primavera de Noruega se dividieron en dos componentes que ocurrían solo a lo largo de la costa con ninguna migración oceánica.

También se han observado cambios en el patrón de migración de los desovantes de verano de Islandia, particularmente en el tiempo en que ocurre la migración entre las áreas de desove al sur y sudeste de las costas de Islandia y las áreas de alimentación más al norte. Antes, los desovantes de primavera de Islanda ocurrían en las mismas áreas que el otro stock de Islandia, pero tenían una época de alimentación. En años recientes este stock ha sido mínimo, y solamente se le ha registrado como parte de una mezcla insignificante en las capturas de los desovantes de verano de Islandia.

Bajo la situación actual de los stocks, las autoridades responsables de ordenación de la pesca no tienen otra opción que tratar de balancear el interés en un rápido incremento de los stocks y las demandas de abrir la pesquería por razones socioeconómicas.

La presente revisión se ha basado mayormente en los trabajos de Dragesund, Hamre y Ulltang (1980), Jakobsen (1980 a,b), Bakken y Dragesund (1971) y los informes de grupos de trabajo de (ICES) sobre arenque Atlanto-Escandinavo.

INTRODUCTION

The present review aims at describing changes in the stocks of Atlanto-Scandian herring (Clupea harengus L.) during the last thirty years and attempts to illustrate the interrelationship between stock dynamics and fisheries development.

Atlanto-Scandian herring must be ranked among the most important fish resources of the Northeast Atlantic with peak landings of more than 1.5 million tons in the mid 1950's. Today landings amount to merely 60,000 tons, and the herring fisheries are of minor importance. The remarkable decline in yield reflects the changes in the stocks, and the present situation can be considered transitional. The Atlanto-Scandian herring, therefore, remain important due to the potential for the future.

The term "Atlanto-Scandian herring" is now generally accepted to cover three stocks:

"Norwegian spring spawners"
"Icelandic spring spawners" and
"Icelandic summer spawners".

Since the Norwegian spring spawners constitute by far the largest stock, the term is frequently applied to that stock alone.

The present outline concentrates on the stock of Norwegian spring spawners, but refers also to the two Icelandic stocks as the recent history of all the stocks shows a series of similar characteristic events. The fluctuations in abundance and distribution of the stocks are well documented by a number of studies, primarily by Norwegian and Icelandic fisheries scientists organized in part as joint investigations. In later years assessments of the stocks are carried out under the auspices of the International Council for the Exploration of the Sea (ICES).

The review presented here is based mainly on papers by Dragesund, Hamre and Ulltang (1980), Jakobsson (1980a,b), Bakken and Dragesund (1971) and reports of the Atlanto-Scandian Herring Working Group published through annual reports by the ICES Advisory Committee on Fishery Management (Anon. 1982).

GENERAL BIOLOGY

During the later decades the distribution and migrations of the Atlanto-Scandian herring have changed. The general features will, however, best be described by the situation observed prior to 1960 when the stocks were at a relatively high level. Fig. 1 gives a schematic illustration of the areas of spawning, feeding and overwintering with indications of migration.

Norwegian Spring Spawners

Spawning takes place in February - March along the coast of Western Norway. After spawning the herring migrate westwards into the Norwegian Sea. During summer they feed heavily on zooplankton in the area northeast of Iceland along the front between cold and warm water masses. In autumn the herring concentrate in cold water east of Iceland and by December start on a spawning migration towards the Norwegian coast.

Herring larvae are transported by the current northwards along the Norwegian coast. Juveniles are found mainly in the fjords while adolescent herring occur both in the fjords of Northern Norway and in the western part of the Barents Sea and the northeastern Norwegian Sea. Maturity is normally reached at age 6-7 by a length of about 33 cm. Maximum age and length are 18-20 years and about 33 cm.

Fig. 1. Generalized illustration of the distribution and migrations of Atlanto-Scandian herring during a period of high stock level. (Modified from Anon. 1970 and Anon. 1974).

Icelandic Spring Spawners

Herring of this stock spawn in March - April at various localities off the south coast of Iceland. The spent herring feed off the south coast and migrate by June northwards on either side of Iceland continuing feeding off the northeast and northwest coast, usually mixing with Norwegian spring spawners feeding in the same area. In late autumn the adult population starts to migrate west and south around West-Iceland to spawn (Jakobsson 1980a).

Herring larvae are carried with currents clockwise along the coast, and the main nursery grounds are in fjords and coastal areas of Northwest - and North-Iceland. At about 2 years of age they mix with the adult population. Maturation normally takes place at the age of 4.

Icelandic Summer Spawners

Spawning takes place in July - August in the same general area as that of the Icelandic spring spawning herring off southern Iceland, but extends somewhat further west. During autumn the spent herring migrate northwards to Northeast- and Northwest-Iceland. At this time the distribution area overlap that of the two spring spawning stocks, although the summer spawners are generally found further south. In early winter the summer spawners migrate west of Iceland and remain during winter off the south coast near the spawning grounds.

As for the spring spawners, the larvae are transported westwards and north to the nursery grounds along the northwestern and northern coast. The summer spawners mature at 3-5 years of age.

The general biology of the three Atlanto-Scandian herring stocks are rather similar. The main difference is related to the dimensions of the distribution area and migrations. At high stock levels the Norwegian spring spawners utilize a major part of the Norwegian Sea for feeding while the Icelandic stocks are confined to the shelf area around Iceland. In general, the distribution area of each stock of Atlanto-Scandian herring corresponds to the relative size of the three stocks.

STOCK SIZE AND CATCH

During the last 30 years the yield and stock size of the three Atlanto-Scandian herring stocks have changed dramatically. Although differences exist among the stocks the histories include a system of parallel events. In the following this is described by an examination of year to year changes in landings, in stock size and in exploitation. For all stocks reliable catch statistics and stock assessments are available, and this makes it possible to describe the development in quantitative terms. It is, however, more difficult to explain the causes of the observed changes. Still, it seems clear that the fisheries have played a major role. For this reason a later section attempts to quantify the changes in the fleet which exploited the stocks, particularly during the period of stock decline.

Norwegian Spring Spawners

Fig. 2A shows the catches of Norwegian spring spawning herring during the period 1950-1980.

In the early 1950's about 70 per cent of the annual increased. This was in part the coast during the spawning period. These catches declined sharply in the late 1950's from about 1 million tons to 400,000 tons by 1960. During the 1950's catches taken off shore by drift nets by the Soviet Union gradually increased, while catches of immature herring fluctuated around 200,000 tons annually.

In the first half of the 1960's the total catch again increased. This was in part due to rich year-classes of 1959 and 1960 (Dragesund, Hamre and Ulltang 1980), but was also a result of the extensive fishery on the open ocean, particularly in the northern part of the Norwegian Sea by purse seiners of Norway and Iceland. In 1966 and 1967 record catches of about 2.0 and 1.7 million tons were landed. This was followed by an immediate drop to only 70,000 tons in 1969.

During the 1970's catches have been governed by various fishing restrictions and less than 10,000 tons have been taken annually.

Fig. 2B shows the stock changes during the same period with an inserted graph illustrating the development of the fishing mortality rate 1955-1968. The data are taken from the comprehensive study by Dragesund, Hamre and Ulltang (1980) updated by stock estimates provided by Hamre for the ICES Advisory Committee on Fishery Management (Anon. 1982).

In the first half of the 1950's the spawning stock was at a level of about 8 million tons. The 1950 year class was strong and contributed to an increase in the spawning stock in the mid 1950's. In 1957 the stock was estimated at 10 million tons. From then on the stock declined until 1963. The decline was caused by reduced recruitment and the fishing mortality on the adults remained at about 0.15. Recruitment improved and the stock increased to 3.7 million tons in 1965. Catches, however, went up rapidly at the same time (Fig. 2A), and the fishing mortality increased steadily, also on the younger age groups.

Fig. 2. Norwegian spring spawning herring
A. Annual catch, total and by major fishing nations, demonstrating the increased catch in the late 1960's and the collapse of the fishery by 1970.
B. Biomass of spawning stock (age > 5 years) and fishing mortality rate (age > 7 years) 1955-1968, showing the stock decline and increased exploitation during the 1960's
(Data and fgure adapted from Dragesund, Hamre and Ulltang 1980, Anon. 1972 and Anon. 1982.)

During the last half of the 1960's the spawning stock declined due to an almost complete lack of recruitment and a continued high exploitation resulting in fishing mortalities above 1.0. By 1970 the stock was reduced to less than 50,000 tons. In the 1970's the stock has increased slowly, and in 1980 it was estimated at 450,000 tons (Anon, 1982). This is only about 5 percent of the stock size during the early 1950's.

The stock of Norwegian spring spawners was near extinction in the early 1970's, and no effective measures of management were introduced prior to this. Analyses by Dragesund, Hamre and Ulltang (1980) have shown that the collapse of the fishery and the stock occurred as a combined effect of increased effort in the adult fishery and a continuation of high exploitation on the immature stock. An introduction of minimum landing size or other means of protecting 0- and 1-group herring in the 1960's, preferrably also combined with a catch quota on adult fish determined by an optimum fishing mortality, might have prevented overexploitation. It is estimated that by such regulations of the fishery the spawning stock could have remained at a level of 2.4 to 4.7 million tons in 1970.

Icelandic Spring Spawners

The history of this stock is illustrated by the graphs of Fig. 3 based on data by Jakobsson (1980a).

The Icelandic spring spawning herring were caught by Icelandic vessels both on the north and the south coast of Iceland. As indicated in the previous section the Icelandic spring spawners occurred at times together with the Icelandic summer spawners and the Norwegian spring spawners. Mixed catches were, however, split on stocks on the basis of biological criteria, mainly by scale typing.

A major part of the catch of Icelandic spring spawners was taken on the north coast in summer.

Fig. 3A shows the catch for the years 1950-1980.

During the 1950's the annual catch increased from 20,000 tons to a peak of 270,000 tons in 1962. From then on catches declined, and near zero catches were taken throughout the 1970's.

Fig. 3B shows the changes in the size of the Icelandic spring spawning herring stock and the fishing mortality during the years of stock decline as reported by Jakobsson (1980a).

The spawning stock increased by good recruitment in the early 1950's and reached a peak level of about 800,000 tons in 1957. During the 1960's the stock decreased sharply, and a linear relationship between stock and recruitment was observed (Jakobsson 1980a). By 1970 the stock was virtually extinct and has shown no sign of recovery.

As shown in Fig. 3 increased catches on the declining stock resulted in a marked increase in the fishing mortality during the early 1960's. Prior to 1960 fishing mortalities were very low, in most years less than 0.1. From this level the fishing mortality rose to 1.0-1.7 in the period 1964-1968, and fell again in the late 1960's as there were no herring left to fish.

Jakobsson (1980a) analysed the development of the fishery and the impact on the stock, and evaluated various management strategies which could have been followed. No fishing of 0- and 1-year old herring and catch quotas corresponding to low fishing mortalities would have postponed or prevented the collapse of the stock. It is, however, evident that drastic changes in the environment contributed to the sharp reduction in recruitment in the late 1960's.

Fig. 3. Icelandic spring spawing herring
A. Annual catch.
B. Biomass of spawning stock (age < 4 years) and fishing mortality (weighted mean, age 4-15 years) 1955-1972 illustrating the relationship between stock decline and fishing mortality increase.
(Data from Jakobsson 1980a.)

Icelandic Summer Spawners

The development of the fishery and assessments of the stock have been described by Jakobsson (1980a). For the most recent years his data are reported to ICES (Anon. 1982) and estimates of stock size from echo surveys 1973-1982 are available (Jakobsson 1982).

Fig. 4A shows the changes in catch during the period 1950-1980. In the 1950's most of the catch was taken on the south coast of Iceland by drift nets in autumn. Purse seining became increasingly important, and from 1960 purse seiners equipped with sonar and power blocks revolutionized the fishery. Catches increased rapidly and exceeded 100,000 tons in the early 1960's.

After 1965 catches declined and regulations in the fishery were introduced. Following a near total ban on fishing in 1972-1974 catches have again increased, and in 1980 about 50,000 tons were allowed to be taken.

Fig. 4B shows the stock changes: a build up period in the 1950's, a decline in the 1960's and new increase during the 1970's. On the figure is also shown the varying rate of exploitation.

Fishing mortalities in the 1950's were low and generally fluctuated between 0.1 and 0.2. In 1960 the situation changed. Fishing mortality increased, particularly on the younger age groups. In 1965 and 1967 the fishing mortality on the mature herring reached 1.2 and 1.5. During the years 1972-1974 purse seining was prohibited and fishing mortalities were negligible. From 1975 the fishing mortality was predetermined and governed the catch quotas in order to allow a rebuilding of the stock.

Stock changes are, of course, not only determined by fishing, but depend also on recruitment. From 1954 to 1963 recruitment was stable and high resulting in an increasing stock. This stock, however, produced small year classes during 1964-1970. In 1971 and 1974 when the stock was at a very low level, relatively good year classes occurred. These year classes, protected by regulations of the fishery, contributed significantly to the spawning stock. Continued good recruitment has resulted in a rapid rebuilding of the stock. At the onset of the 1980's the stock is again at the level of that found in the mid 1960's.

It will be recognized from the descriptions given above that the three stocks of Atlanto-Scandian herring have passed through a rather parallel history from 1950 to 1980. The first decade was characterized by stable or increasing stocks. They were lightly fished by traditional gear near the coast predominately during the spawning period. The overall biomass of the stocks was about 10 million tons yielding catches of about 1 million tons. During the 1960's all stocks declined at a rapid rate. Catches remained high as the stocks were now fished extensively. A new technologically advanced fishing fleet of seiners followed the shoals of herring migrating between spawning areas near the coast and feeding areas in the open sea. At the same time the exploitation changed to include a larger proportion of immature herring. By the beginning of the 1970's all stocks were in a depleted stage. Catches had fallen to a minimum, and management in the way of fisheries restrictions seemed to have come too late. During the 1970's management has aimed at minimizing the fisheries in an attempt to allow the stocks to recover. Such a recovery has been seen for two of the stocks, but the recruitment and the rate of recovery have been different. At present, the total biomass of adult Atlanto-Scandian herring is only about 700,000 tons, and the allowed catch 60,000 tons, i.e. less than 10 per cent of the stock and yield in the early 1950's.

FISHING FLEETS

The histories of the Atlanto-Scandian herring stocks are linked to the development of the fleet of vessels exploiting the stocks. The escalated exploitation and the decreasing stocks in the 1960's were in general terms a result of technical improvements and increased efficiency of the purse seine fleet.

Fig. 4. Icelandic summer spawning herring
A. Annual catch showing the fluctuating yield 1955-1980.
B. Biomass of spawning stock (age > 3 years at the time of spawning) and fishing mortalities (weighted mean, 1950- 1959 age 3-10, later age 3-13) reflecting the effects of the unrestricted fishery and the fisheries managment of the 1970's.
(Data from Jakobsson 1980a and Anon. 1982.)

The fishing effort in a fishery by purse seiners is difficult to define and measure. For this reason no effort data, suitable to quantify changes in the fishing power, are available. Evidence of such changes can, however, be found in records of the number of vessels participating in the herring fishery together with indicators of altered catching capabilities.

Purse seining has been the most important fishing method in terms of landed catch of Atlanto-Scandian herring over the period 1950-1980. Other gears, mainly nets, accounted for about 40 per cent of the landings in the early 1950's, but fell to less than 10 per cent by 1960.

Fig. 5 shows the changes in number of Norwegian purse seiners during the recent 30 years. Up to 1957 the number increased steadily, and catches of Norwegian spring spawning herring went up at the same rate. In the late 1950's the stock declined (Fig. 2) and the seiners lost the main source of income and ran into an economic crisis. Consequently, the number of seiners in operation fell considerably. At this stage a new development started. Encouraged by the success of technical improvements in Iceland the fleet went through a modernization process and started to fish on the open sea. Fishing shifted from the depleted stock of Norwegian spring spawners in succession to North Sea herring, mackerel and capelin.

By the end of the late 1960's the fleet consisted of about 400 seiners having a cargo capacity of about 100,000 tons (Mietle 1969). In 1969 the fleet landed a total of about 1.5 million tons of herring, mackerel and capelin. The catching capacity was, however, not fully utilized. Given sufficient fish resources, adequate shore facilities and acceptable market conditions landings could possibly be doubled (Bakken and Dragesund 1971).

The evolution of the fishery was a result of a remarkable technological development of the purse seine fishing in the late 1950's and early 1960's. This is illustrated in Fig. 6.

In the 1950's purse seine fishing was carried out by a two-dory method. Half of the net was loaded in each of two dories which shot the net in semi circles. In the beginning of the 1960's the fishing changed to a ring net method. The net was operated from the aft section of the seiner and hauled back by a power block (Hamre and Nakken 1971).

The new system had a series of advantages. Time and labour were saved, larger and deeper nets could be operated, and fishing was less affected by rough weather allowing off shore operation.

Along with this improvement acoustic fish finding instruments, particularly the sonar, came into use (Fig. 6). This greatly increased the ability of the vessels to locate herring shoals. In addition a number of other innovations further increased fishing power; fish pumps, synthetic fibre nets, transverse thrust propellers, lengthened vessels, lifted decks, etc. This resulted in improved earnings which due to the taxation system were reinvested in new vessels, improvements and gears (Mietle 1969).

It seems now clear that the increased fishing power of the purse seine fleet had a determining effect on the stock of Norwegian spring spawning herring. The stock decline came about as a result of increased fishing effort particularly on the younger age groups in the summer and autumn. Ulltang (1976 and 1980) has shown that although the stock decreased the catchability actually increased, since for a schooling fish like the herring a constant fishing effort will generate a constant catch instead of a constant fishing mortality as is usually assumed.

In brief it may seem appropriate to characterize the relationship between the fishing fleet and the stock during the period 1950-1980 thus: in the beginning the livelihood of the fleet was determined by the stock, later the fleet determined the livelihood of the stock.

The development of the purse seine fleet in Iceland was very similar to that of Norway, only that the technological changes occurred earlier and faster. In the late 1950's the Icelandic fleet consisted of close to 250 purse seiners (Jakobsson 1980b). During the 1960's the number was reduced, but as for the Norwegian vessels the fishing power increased due to the technological improvements. The sonar guided ring net method by power block had replaced the two-dory fishing by about 1960 (Jakobsson 1980b). At the end of the 1960's the number of purse seiners were reduced to about 100, and during the next 10 years further to about 50.

Fig. 5. Changes in the number of Norwegian purse seiners (1.o.a. <90 feet) with indications of main target species. (Data by Mietle 1969 and Fishery statistics of Norway.)

Fig. 6. Technological development in the Norwegian fleet of purse seiners in the 1960's illustrated by the adoption of sonar and the change over from two-dory fishing to the ring net technique in the herring fishery. (Adapted from Bakken and Dragesund 1971.)

It may be concluded that also for the Icelandic stocks the fishery played a decisive role in reducing the stock size. The Icelandic spring spawners, however, could be considered as a stock of Atlanto-Scandian herring living at the outer limits of its distribution (Jakobsson 1980a), and shifts in environmental conditions seem to have contributed significantly to stock changes. In the stock of Icelandic summer spawners the rate of decline was reduced, but not halted, due to restrictions in the purse seine fishery.

BIOLOGICAL CHANGES

In association to changes in the abundance of Atlanto-Scandian herring the general biology of the stocks has also changed, e.g. distribution, migration, spawning and growth.

In the 1950's the Norwegian spring spawners were feeding in the Norwegian Sea north-east of Iceland towards the island Jan Mayen, overwintered east of Iceland and migrated from this area eastwards to extensive spawning grounds along the west coast of Norway (Fig. 1). As the adult stock was reduced in the 1960's, the area of distribution shifted northwards, north of Jan Mayen, with a new overwintering area off northern Norway. The area of spawning diminished, and the most important grounds were found on the coast of northern Norway.

In the early 1970's the adult stock was minimal, and spawning was not observed. Restrictions in the fishery protected the 1969 year class. This year class spawned at an early age, in 1973, mostly on the coast of northern Norway. After spawning the herring did not migrated to the open sea, but remained along the coast during the feeding period in summer. In this way the life history and migration pattern of the stock was totally changed (Dragesund, Hamre and Ulltang 1980).

In later years, the stock has remained in coastal waters split in a northerly and a southerly component having separate spawning grounds. Recruitment has been low, and the slight increase in stock size (Fig. 2) is mainly due to an increase in the southern component (Anon. 1982).

The observed individual growth rate in the 1970's has been significantly higher than that of the 1950's. As example, an 8 year old herring of the 1969 year class reached a length of 36.5 cm while the average length at the same age during the 1950's and 1960's was about 34 cm. Along with this the mean age of maturity has been reduced from 6-7 in the 1950's to 4 years in the 1970's (Hamre, pers. comm.).

Changes have also been observed in the migration pattern of the Icelandic summer spawners, particularly in the timing of the southward migration from the feeding area to the overwintering area off southern Iceland. The migration starts earlier, and the stock now remains during winter in dense shoals close to the shore in restricted localities on the south coast (Jakobsson 1982).

In the previous section it was pointed out that the disappearance of the Icelandic spring spawners seem to have been associated with adverse environmental conditions: low sea temperatures and absence of zooplankton to feed on. Conditions improved after the collapse of the stock (Jakobsson 1980b), but Icelandic spring spawners are now only recorded as an insignificant admixture to the Icelandic summer spawners.

Catch statistics and records of the fisheries on Norwegian spring spawners from early 1800 indicate wide fluctuations, and suggests that the recent situation of extremely low stock level is not unique. An hypothesis of alternation in spawning grounds and natural cycles of fluctuating abundance caused by biotic factors was put forward by

Devold (1963). The investigation and analyses of later years indicate that the hypothesis must be rejected, but the concept of stock components seems verified.

MANAGEMENT

Saetersdal (1980) has reviewed the management history of the Norwegian spring spawning herring along with other pelagic fish stocks.

The collapse of the Norwegian spring spawning stock (Fig. 2B) was a very dramatic event, but it attracted surprisingly little attention from the scientific community at the time. In 1965 the Liaison Committee of ICES reported to the Northeast Atlantic Fisheries Commission (NEAFC) as the responsible international management body that the stock was declining. The decline was, however, assumed to have natural causes and no action was taken. Even as late as 1970, after the fishery had completely collapsed, the scientific advice was insubstantial: "reduce the mortality rate on immature herring and avoid further increase in fishing rate on the adult stock". Not until 1975 was a complete ban on fishing recommended.

It seems that the stock events were, at least in part, recorded by the scientists during this period, but the significance was not realized, and no effective management action was taken until too late. To some extent this can be attributed to the system of fisheries management in the 1960's, since a consensus of several fishing nations was required in order to obtain effective regulations.

The management of the Icelandic stocks, meaning restrictions in the fisheries, came in force at a somewhat earlier stage of the stock decline. In 1966 a minimum legal size of landed herring was introduced, advice (Jakobsson 1980a) was given on a maximum catch quota and closed seasons (Fig. 4). The recommendations were gradually accepted, but in general, they were enforced too late to halt the stock decline, and the Icelandic spring spawners were depleted before effective action was taken.

After 1970 the biological monitoring and, based on this, the established system of strict management has been of prime importance for the rebuilding of the Icelandic summer spawning stock.

In the present situation the management objective is to rebuild the herring stocks. To achieve this the authorities have no option but to attempt to balance the interest in a rapid rate of stock increase to the demands of allowing a fishery for socio-economic reasons. This is particularly the case for the Norwegian spring spawners for which the long term annual yield of a rebuilt stock would be at least 1.5 million tons.

The future of the Icelandic spring spawner cannot be expected to be determined by any form of management. A re-establishment of this herring appear to be dependent on the development of the Norwegian spring spawning stock as Jakobsson (1980a) has suggested that the stock will not recover without immigration.

ACKNOWLEDGEMENT

I am greatly indepted to colleagues at the Institute of Marine Research, Bergen, particularly Mr. Johannes Hamre, for providing data on the Norwegian spring spawners and to Mr. Jakob Jakobsson at the Marine Research Institute, Reykjavik, for allowing me to cite unpublished results.

REFERENCES

Anon. 1970. Report of the working group on Atlanto-Scandian herring. Coop.Res.Rep.Int.Coun. Explor.Sea. Ser.A 17:1-48.

Anon. 1972. Report of the meeting of the working group on Atlanto-Scandian herring. Coop.Res. Rep.Int.Coun.Explor.Sea. 30:1-27.

Anon. 1974. Survey of fish resources in the North-east Atlantic. Coop.Res.Rep.int.Coun.Explor.Sea. 37:1-111.

Anon. 1982. Report of the advisory committee on fishery management, 1982. Int.Coun.Explor.Sea. 1982:1-267.

Bakken, E. and O. Dragesund. 1971. Fluctuations of pelagic fish stocks in the northeast Atlantic and their technological and economic effects on the fisheries. International Symposium on fisheries economics. OECD symp., Paris 1980 (FI/T 1/6):1-26.

Devold, F. 1963. The life history of the Atlanto-Scandian herring. Rapp.p.-v.Réun.Cons.int.Explor. Mer. 154:98-108.

Dragesund, O. and Ø. Ulltang. 1978. Stock size fluctuations and rate of exploitation of the Norwegian spring spawning herring 1950-1974. Fisk Dir.Skr.Ser.Havunders. 16:315-337.

Dragesund, O., J. Hamre and Ø. Ulltang. 1980. Biology and population dynamics of the Norwegian spring spawning herring. Rapp.P.-v.Reun.Cons.int.Explor.Mer. 177:43-71.

Hamre, J. and O. Nakken. 1971. Technological aspects of the modern Norwegian purse seine fisheries. pp. 225-231. In Modern Fishing Gear of the World: 3. Fish Finding, Purse Seining and Aimed Trawling. Fishing News (Books) Ltd., London.

Jakobsson, J. 1980a. Exploitation of the Icelandic spring and summer spawning herring in relation to fisheries management 1947-1977. Rapp.P.-v.Réun.Cons.int.Explor.Mer 177:23-42.

Jakobsson, J. 1980b. The north Icelandic herring fishery and environmental conditions 1960-1968. Rapp.P.-v.Réun.Cons.int.Explor.Mer 177:23-42.

Jakobsson, J. 1982. Echo surveying of the Icelandic summer spawning herring 1973-1982. Symposium on fisheries acoustics. ICES/FAO Symp., Bergen, 1980(22):1-36.

Mietle, P.L. 1969. Developments and investments in recent years in Norwegian purse seine fisheries for herring, capelin and mackerel. International conference on investment in fisheries. FAO conf., Rome, 1969 (Bp 43):1-21.

Ulltang, Ø. 1976. Catch per unit of effort in the Norwegian purse seine fishery for Atlanto-Scandian (Norwegian spring spawning) herring. FAO Fish.Tech.Pap. (155):91-101.

Ulltang, Ø. 1980. Factors affecting the reaction of pelagic fish stocks to exploitation and requiring a new approach to assessment and management. Rapp.P.-v.Reun.Cons.int.Explor.Mer. 177:489- 504.

Saetersdal, G. 1980. A review of past management of some pelagic stocks and its effectiveness. Rapp.P.-v.Réun.Cons.int.Explor.Mer. 177:505-512.