The population biology and exploitation of capelin (Mallotus villosus) in the Barents Sea

Abstract

The life history of the Barents Sea capelin stock through the various growth, mortality at the different life stages. The ecological role of the capelin is discussed, as well as its population dynamics. The stock history, its abundance and exploitation is dealt with, together with the history of stock assessment and management. The main aim of the review is to compile and bring to light many not readily available sources of knowledge concerning the Barents Sea capelin stock. These include Russian literature, cruise reports, theses, various kinds of working documents. ER -

Full text

many years also been studied by scientists from the
Polar Institute of Fisheries and Oceanography PINRO
in Murmansk, Russia. Unlike for the Icelandic capelin,
where a detailed review was recently published
(VILHJÁLMSSON 1994), and despite the ecological impor-
tance of capelin and its key role as target for the fishing
industry, no comprehensive review of its biology and
ecological role in the Barents Sea has been compiled.
Much of the existing information can only be found in
unpublished cruise reports, unpublished papers pre-
sented to various meetings and symposia, and in theses
and other kinds of ‘grey’ literature. The aim of this article
is to present a synopsis of the knowledge of the Barents
Sea capelin stock, based on the information found in
these sources. In addition, I will present some results of
ongoing studies, utilising the steadily growing capelin
data base at the Institute of Marine Research in Bergen.
1 INTRODUCTION
The Barents Sea capelin (Mallotus villosus Müller) stock
is potentially the largest capelin stock in the world, its
biomass in some years reaching 6-8 million tonnes. It is
the largest stock of pelagic fish in the Barents Sea, with
a key role as an intermediary of energy conversion from
zooplankton production to higher trophic levels, annu-
ally producing more biomass than the weight of the stand-
ing stock. It serves as a forage fish for other fish species
as well as marine mammals and sea birds, and has pro-
vided an annual fishery harvest of up to 3 million tonnes.
The stock became a special object of interest to the fish-
ing industry when the fishery on the Norwegian spring-
spawning herring was banned in the early 1970s. A com-
prehensive research program for studying the capelin
stock was initiated by the Institute of Marine Research
in Bergen, Norway around 1960, and this species has for
INVITED REVIEW
SARSIA THE POPULATION BIOLOGY AND EXPLOITATION OF
CAPELIN (MALLOTUS VILLOSUS) IN THE BARENTS SEA
HARALD GJØSÆTER
GJØSÆTER, HARALD 1998 12 30. The population biology and exploitation of capelin (Mallotus
villosus) in the Barents Sea. – Sarsia 83:453-496. Bergen. ISSN 0036-4827.
The life history of the Barents Sea capelin stock through the various phases from egg to maturity
is reviewed, including distribution, feeding, growth, mortality at the different life stages. The
ecological role of the capelin is discussed, as well as its population dynamics. The stock history,
its abundance and exploitation is dealt with, together with the history of stock assessment and
management. The main aim of the review is to compile and bring to light many not readily
available sources of knowledge concerning the Barents Sea capelin stock. These include Russian
literature, cruise reports, theses, various kinds of working documents.
Harald Gjøsæter, Institute of Marine Research, PO Box 1870 Nordnes, N-5024 Bergen, Norway.
E-mail: harald.gjoesaeter@imr.no
KEYWORDS: Capelin; Mallotus villosus; Barents Sea; biology; ecology; stock history; fishery; stock
assessment; fisheries management.

454 Sarsia 83:453-496 – 1998
68° 60° 70°50°40°
30°
20°10°0°
70°
72°
74°
78°
76°
80°
Novaya Zemlya
Barents Sea
Franz Josef Land
Spitsbergen
Finnmark
Kola
Bear Island
Troms
Fig. 1. The Barents Sea and adjacent areas, with main ocean currents, bathymetry (200 m and 500 m depth
contours) and names of places mentioned in the text. The currents entering the Barents Sea from the south-
west are the North Atlantic current carrying warm high-salinity water and the Norwegian Coastal Current
carrying warm low-salinity water. The currents entering from the north and east are carrying cold low-
salinity Arctic water.
2 SHORT DESCRIPTION OF THE AREA
Fig. 1 is a map of the Barents Sea, showing some topo-
graphical and hydrographical features and names of places
mentioned in the text. The Barents Sea is a high-latitude,
shallow continental shelf area. It is bounded in the north
by the archipelagos of Spitsbergen and Franz Josef Land,
in the east by Novaya Zemlya, and in the south by the
coasts of northern Norway and Russia (Fig. 1). In the
west, the boundary between the Barents Sea and the
Norwegian Sea is usually drawn along the continental
edge at about 10° to 15°E. More than 20 % of the area is
shallower than 100 m, but troughs deeper than 400 m
enter the area from the west and north-east.
The Norwegian Coastal Current flows along the coast of
Norway and Russia, given the name Murman Coastal Cur-
rent when it crosses the border between the two countries.
The Norwegian Atlantic Current flows into the Barents Sea
from south-west, dividing into two branches flowing east-
wards and north-eastwards. Arctic water enters the Barents
Sea through the channel between Spitsbergen and Franz
Josef Land and, more important, between Franz Josef Land
and Novaya Zemlya (LOENG 1991).
The three main water masses of the Barents Sea,
Coastal Water, Atlantic Water and Arctic Water, are
linked to these current systems. In addition, locally
formed water masses resulting from processes taking place
inside the area, e.g. seasonal freezing and melting of ice,
can be found. Where the Atlantic and Arctic water meet,
a well-defined Polar Front is formed. Its position is rather
stable in the area south of Spitsbergen, where it is
governed by the bottom topography, but is more variable
in the eastern parts of the Barents Sea.
3 STOCK DISCRIMINATION
RASS (1933) divided the Barents Sea capelin into three
‘forms’ or ‘races’ which he called the Finnmarken, the
Murman and the Novaya Zemlya capelin, after their
spawning places. These groups spawned in spring, sum-
mer and autumn respectively. However, PROKHOROV
(1965) and LUKA (1978) were of the opinion that spring
and summer-autumn spawning capelin were not ecologi-
cally isolated groups. COLLETT (1903) mentions one oce-
anic stock of capelin and several fjord stocks living in
Norwegian fjords in Finnmark, Troms, Nordland and
Trøndelag counties. He argued that the fjord stocks are
not completely isolated from the oceanic stock, although

Gjøsæter – The population biology of capelin in the Barents Sea 455
ABC
DEF
GHI
J
Fig. 2. Development of the capelin egg at 4 °C. A: 5 hours after fertilisation. B: About 12 hours after fertilisation. C: About 24
hours after fertilisation. D: Age 4 days. E: Age 7 days. F: Same age, frontal view. G: Age 12 days. H: Age 20 days. I: Age 25 days.
Embryo is dissected out of the egg). J: Newly hatched capelin larva. From GJØSÆTER & GJØSÆTER (1986).
they mainly spawn within the fjords. DUSHCHENKO (1985),
who used electrophoretic studies of variability of
myogens, non-specific esterases and malic enzyme, found
no reasons to distinguish any reproductively isolated
groups. He concluded that his results confirmed the opin-
ion, already existing, that early and late spawning capelin
were not independent reproductive groups. In Balsfjorden,
Troms, Northern Norway, there is what is normally
considered a local fjord stock of capelin. However, using
genetic methods MORK & FRIIS-SØRENSEN (1983) argued
that inter-sample differences in allele frequencies at four
polymorphic loci were not significant and thus did not
indicate genetic isolation between the fjord stock and the
oceanic stock. On the other hand, KENNEDY (1979), who

456 Sarsia 83:453-496 – 1998
after fertilisation a fertilised egg may be distinguished
from an unfertilised as it has a clear periviteline space.
After about five hours the blastodisc is seen as a cap on
top of the yolk (Fig. 2A).
Stage 2. Cleavage of blastodisc, morula, blastula.
Duration: from age seven hours to age two days. Ap-
pearance: At age seven hours the egg is at the two-cell
stage, and continues through the four-cell stage (Fig. 2B)
et. seq. As the cleavage progresses, the individual cells
become progressively more difficult to discern. The
morula (Fig. 2C) is visible after about 24 hours and, in
the course of the second day, the morula begins to be
hollowed out, forming the blastoderm.
Stage 3. Gastrulation, closure of blastopore. Duration:
from age two to six days. Appearance: Around day three
the blastoderm starts to grow around the yolk, a process
which can easily be observed in the egg. At day four the
rim of the blastoderm reaches about three fourths of the
distance around the yolk (Fig. 2D). Simultaneously,
gastrulation takes place. At age five days the embryo is
seen as an oval thickening of the blastoderm , which at
day six can be seen to reach about half way around the
yolk sac.
Stage 4. Organogenesis I. Formation of pre-organs.
Duration: from age six to twelve days. Appearance: On
day seven the head end of the embryo has become broader
and higher than the tail end (Fig. 2E and F) and on the
next day the optic bulbs begin to form. During this stage
there are only minor changes in the outer appearance of
the embryo. There is some growth in length, but the
embryo does not reach around the circumference of the
yolk sac (Fig. 2G). Towards the end of this stage the
inner ear can be observed to contain structures which are
probably the primordial otoliths.
Stage 5. Organogenesis II. Further organ development.
Duration: from age twelve to twenty-four days. During
this stage the embryo begins to move, the heart starts to
beat, and the eyes become pigmented. The body grows
in length, and the tail continues developing. Fig. 2H shows
the embryo 20 days after fertilisation. At day 22 a faint
pigmentation appears below the gut, and during the two
last days of this stage the pigmentation becomes more
distinct.
Stage 6. Preparation for independent feeding. Dura-
tion: from age 25 days to hatching, which may start around
day 33 and last for more than 20 days for a batch of eggs.
Appearance: At the beginning of this stage melanophores
are present both below and above the gut, and
pigmentation is also more pronounced under the tail and
on the yolk sac (Fig. 2I). The head separates from the
yolk sac. Three to four days later the segmentation reaches
the tail, and in the yolk sac the oil globules begin to
aggregate into one large sphere. About age one month the
pectoral fins appear, and the mouth starts to form. At
0
10
20
30
40
50
60
70
80
90
100
22 26 30 34 38 42 46 50 54 58 62 66 70 74 78
INCUBATION PERIOD (DAYS)
HATCHING (%)
7 deg
4 deg
2 deg
Fig. 3. Hatching curves for three batches of eggs incubated at
2, 4, and 7 °C. Redrawn after GJØSÆTER & GJØSÆTER (1986).
studied infestation by the cestode parasite Eubothrium
parvum in capelin from the Barents Sea and Balsfjorden,
concluded that the difference in frequency distribution
and the failure to find any heavily infested fish in the
Barents Sea confirm the suggestion that the capelin of
Balsfjord form a local isolated population, which does
not migrate into the Barents Sea.
It seems reasonable to conclude, for the moment, that
there is one large oceanic stock of capelin in the Barents
Sea and, in addition, one or more populations in fjords
like that in Balsfjorden, although not completely isolated
genetically from the oceanic stock, may be self-contained.
This paper deals with the Barents Sea stock.
4 THE LIFE HISTORY
4.1 THE PLANKTONIC STAGES
4.1.1 Embryonic and larval development
GJØSÆTER & GJØSÆTER (1986) kept artificially fertilised
eggs from capelin of the Barents Sea stock under control-
led temperature conditions comparable to those observed
on the spawning beds. They gave a description of the
development and the effect of temperature on the
embryonic growth, the eggs’ ability to adhere to the
substrate, and the fertilisation rate at different salinities.
The description of the embryonic development given
below is based on a temperature of 4 °C, a typical tem-
perature at the spawning beds of the Barents Sea capelin.
The embryonic stages referred to in the description of
the development are more or less identical to those used
by FRIÐGEIRSSON (1976) when describing the develop-
ment of the Icelandic capelin. The duration of each stage
at 4 °C is given for the fastest developing eggs in the
study group which hatched after 34 days (GJØSÆTER &
GJØSÆTER 1986).
Stage 1. Blastodisc formation. Duration: from fertili-
sation to age six hours. Appearance: About two hours

Gjøsæter – The population biology of capelin in the Barents Sea 457
days 33-34 the pigmentation resembles that of a newly
hatched larva (Fig. 2J). The mouth seems fully developed
and is open.
Hatching curves for three batches of eggs, incubated at
2, 4 and 7 °C (Fig. 3) show that the incubation period is
to a large degree dependent on temperature, varying from
about 20 days for the fastest developing eggs at 7 °C to
80 days for the slowest developing eggs at 2 °C. At
hatching, the mean total length was 7.55 mm (N = 102,
range 6.1-8.2 mm) and the mean yolk sac diameter was
1.15 mm (N = 102, range 0.7-2.0 mm).
POZDNJAKOV (1960) also studied the embryonic devel-
opment of the Barents Sea capelin, but used a somewhat
less detailed stage description than the one adopted here.
He reported length at hatching to be from 4.8 to 7.5 mm,
but it is not quite clear whether he measured the total
length of the larvae.
4.1.2 Growth of larvae
Feeding, growth and survival of capelin larvae from the
Barents Sea stock were studied in an outdoor basin by
MOKSNESS (1982). He sampled naturally spawned eggs
from a spawning site at the coast of Finnmark, which
hatched in the laboratory and were released in a 2000 m3
outdoor basin. Approximately 100 000 larvae were re-
leased in the basin, and 2.1 % survived after 127 days,
when the experiment was terminated.
Mean growth in length during the first 12 days was
0.29 mm day–1, but decreased to about 0.2 mm day–1
from age 40 days until the end of the experiment. The
growth rate is expected to be determined by the density
of zooplankton, and in another experiment, when two
batches of capelin larvae were given zooplankton in den-
sities more than 10 times higher than observed in the
basin experiment, they grew at rates of 0.44 mm and 0.31
mm day–1 during the first 26 and 15 days respectively
(ØIESTAD & MOKSNESS 1979). The temperature condi-
tions in the basin during these experiments (8-20 °C at
the surface and 6-12 °C near the bottom (MOKSNESS 1982))
were higher than experienced in the natural habitat in the
southern Barents Sea. This probably increased the growth
rate but it is uncertain to what extent.
A larval survey of capelin in the Barents Sea has been
conducted annually since 1981 (ALVHEIM 1985; FOSSUM
1992; ICES 1996a). The aim of that survey has been to
describe the distribution and abundance of the larvae.
The survey has normally been carried out in the last half
of June, i.e. when most of the larvae are about one month
old (Section 4.1.4). The larvae caught at each station
(Gulf III high speed plankton sampler, ZILSTRA 1971)
were length measured. In most years, the majority of the
larvae were of 5 to 15 mm standard length, while the
number of larvae > 20 mm was low. The mean length in
the period 1981 to 1990 varied from 8.9 mm to 12.9 mm.
If an age of one month and a standard length at hatching
of 6 mm are assumed for all years, these mean lengths
correspond to a mean daily growth rate of 0.10-0.23 mm
day–1. Based on counts of primary rings in otoliths of
field sampled 0-group capelin, GJØSÆTER & MONSTAD
(1982) calculated a mean growth rate of 0.174 mm day–1.
Annual 0-group surveys have been carried out in the
Barents Sea in August since 1965. The main aim of this
survey has been to describe the distribution of the 0-
group of various species and to calculate abundance indi-
ces. LOENG & GJØSÆTER (1990) analysed the growth of
various 0-group species in relation to temperature condi-
tions based on data from 1965 to 1989. The mean total
length of capelin varied from 35-58 mm, with a mean for
all years of 45 mm. As pointed out by the authors,
offspring from summer spawning capelin (see section
4.2.3) may have influenced the mean length in some years.
However, in only 6 out of the 32 years of data, capelin
smaller than 20 mm were included in the measurements
and then in very low numbers. A length of 20 mm in late
August would, if these specimens derived from the main
spawning in spring, correspond to a mean growth rate in
the order of 0.15 mm day–1. Assuming an age of three
months for the 0-group capelin with mean length of 45
mm, measured in August, gives a mean growth rate over
the period of 0.4 mm day–1. These results indicate that
the growth rate in terms of length is higher in the period
July-August than it is in the period May-June. LOENG &
GJØSÆTER (1990) found some evidence for a positive
relationship between mean length in August and variations
of temperature conditions in the Barents Sea.
4.1.3 Larval feeding
Larvae kept in the basin at Flødevigen (MOKSNESS 1982)
were observed to reach the end of the yolk sac stage
(EYS) at age 10 days (at 8 °C). They began to feed at age
4 days (laboratory) and 5 days (basin) while the yolk sac
volume was 0.020 mm3. In the basin, the feeding inci-
dence was low (< 10 %) during the first 25 days, but had
increased to 70 % on day 40. The length of the longest
prey organisms increased from 300 to 1230 µm at a lar-
val length from 7 to 20 mm, and further to 1400 µm for
larval lengths up to 40 mm. The smallest prey organisms
found in the larval guts consisted of various
phytoplankton organisms of 9-50 µm in length. The
zooplankton in the basin was dominated by larvae of
Spionidae spp. (10 organisms l–1) during the first part of
the experiment while veligers of Littorina spp. (5 organ-
isms l–1) dominated during the remainder of the period.
The gut content of the larvae reflected the composition
of plankton in the basin. Thus, the larvae were appar-
ently preying upon the dominant organisms of suitable
size in their surroundings.
MOKSNESS (1982) also reported on a field study of

458 Sarsia 83:453-496 – 1998
coid and calanoid copepods.
BJØRKE (1976) studied feeding of larval capelin near
the coast of Finnmark in May 1971. The food items
eaten by larvae 4.8-21.0 mm in length, mainly consisted
of Calanus eggs (52 %) and Calanus nauplii (42 %). By
comparing the gut content with the composition of plank-
ton in the sampling area (Table 1) he concluded that the
larvae preferred eggs over nauplii.
The larvae began to feed while still having large yolk
sacs, but the feeding incidence increased with decreasing
yolk sac size. Inspection of larvae, sampled during a 24
hour cycle, led to the conclusion that feeding started shortly
after sunrise and declined at nightfall.
4.1.4 Geographical distribution of larvae and 0-group
From 1967-1980, investigations of larval capelin distri-
butions were carried out in most years, but
no abundance estimates were made
(HOGNESTAD 1969a, b, & c, 1971; BUZETA &
al. 1975; GJØSÆTER & MARTINSEN 1975; HAMRE
& RØTTINGEN 1977; DOMMASNES & al. 1978;
DOMMASNES 1978b; DOMMASNES & al. 1979a;
DOMMASNES & al. 1979b; ELLERTSEN & al. 1980;
SEREBRYAKOV & al. 1984). Since 1981, annual
surveys for the purpose of describing the
geographical distribution and abundance of
capelin larvae have been carried out in June
(ALVHEIM 1985; FOSSUM & BAKKEPLASS 1989;
BAKKEPLASS & LAUVÅS 1992; GUNDERSEN
1993a, 1993b; KRYSSOV & TORESEN 1993;
HAMRE & KRYSSOV 1994; TANGEN 1995;
TANGEN & BAKKEPLASS 1996). From 1965, an
international 0-group survey of the Barents
Sea has been carried out annually in August-
September (ICES 1965, 1966, 1967, 1968,
1969, 1970, 1971, 1973a, 1973b, 1974, 1975,
1976, 1977, 1978, 1979, 1980, 1981, 1982b,
1983, 1984a, 1985a, 1986, 1987, 1988, 1989,
1990, 1991b, 1992, 1994, 1995, 1996b,
1996c). Based on the distribution maps and
textual information presented in these reports,
the approximate western, northern and eastern
boundaries as well as the characteristics
(types) of the larval and 0-group distribution
in April-June and August-September are given
in Table 2. Distribution maps of larvae in
May-June, together with spawning areas (See
section 4.2.1.4), have also been constructed
(Figs 4-6). Before 1981, the total distribution
area of the capelin larvae was not always
covered. Consequently, maps for earlier years
do not show the northern extension of the
distribution area. In general, larvae are found
east to about 36-37°E in May-June, while
Table 1. Comparison of gut content of larvae from a station
with mean length 7.97 mm, 64 % without yolk sac, and of
surrounding plankton. From BJØRKE (1976).
Food items In plankton In diet
Number per m3% Number %
Calanus eggs 100 3 21 54
Calanus nauplii 1100 30 17 44
Copepods 2300 64 - -
Other food 100 3 1 2
Table 2. Geographical distribution of capelin larvae in April-June (larval
surveys), and in August-September (0-group surveys), shown by its west-
ern, northern and eastern limits. The distribution type is characterised ac-
cording to the main distribution areas. See text for data sources.
Larval survey 0-group survey
Year Western Eastern Northern Western Eastern Distribution
limit (°E) limit (°E) limit (°N) limit (°E) limit (°E) type
1965 15 41 central
1966 22 46 central-east
1967 18 unknown unknown 22 48 central-east
1968 16 31 unknown 16 50 west-east
1969 14 31 unknown 5 45 west-east
1970 22 31 unknown 22 50 central-east
1971 15 40 73 10 52 west-east
1972 15 40 73 10 50 west-east
1973 27 unknown 73 15 50 central-east
1974 32 unknown 70 20 45 central-east
1975 30 40 unknown 28 50 east
1976 25 38 73 18 50 central-east
1977 27 35 72 20 45 central-east
1978 30 37 72 26 43 east
1979 25 37 73 20 55 central-east
1980 25 38 73 15 50 central-east
1981 16 34 73 5 55 west-east
1982 16 33 73 5 N.A. west-east
1983 16 36 74 <5 50 west-east
1984 18 36 73 <2 55 west-east
1985 17 34 73 5 46 west-east
1986 29 31 70 26 50 east
1987 30 33 71 25 50 east
1988 22 33 73 20 50 west-east
1989 18 34 74 7 42 west-east
1990 21 35 74 18 55 west-east
1991 17 36 74 5 55 west-east
1992 19 37 73 20 55 west-east
1993 18 38 74 25 56 central-east
1994 31 36 71 30 50 east
1995 30 35 70 30 50 east
1996 18 37 73 5 56 west-east
capelin feeding in spring 1971. The number of food items
in the gut of larvae caught in the field was at the same
level as that in the basin and no particular prey group
dominated. Larvae caught in the field (yolk sac larvae
with yolk sacs from 0.03 mm3 to EYS, and larvae from 6
to 15 mm) mostly fed on copepod nauplii and harpacti-

Gjøsæter – The population biology of capelin in the Barents Sea 459
Fig. 4. Spawning areas and spring larval distribution during the period 1967-1976. See text for data sources. Open stars: Assumed
spawning areas, filled stars: known spawning areas, Norwegian surveys. Circles: Known spawning areas, Russian surveys.

460 Sarsia 83:453-496 – 1998
Fig. 5. Spawning areas and spring larval distribution during the period 1977-1986. See text for data sources. Open stars: Assumed
spawning areas, filled stars: known spawning areas, Norwegian surveys. Circles: Known spawning areas, Russian surveys.

Gjøsæter – The population biology of capelin in the Barents Sea 461
Fig. 6. Spawning areas and spring larval distribution during the period 1987-1996. See text for data sources. Open stars: Assumed
spawning areas, Norwegian surveys. Circles: Known spawning areas, Russian surveys.

462 Sarsia 83:453-496 – 1998
could any difference be detected between larval length
groups of 6-9 mm and 10-14 mm.
BELTESTAD, NAKKEN & SMEDSTAD (1975) found that in
August the 0-group capelin descended down to the
thermocline during night while they partly stayed in the
surface layer during daytime.
4.2 THE IMMATURE AND ADULT PHASE
The capelin undergo metamorphosis when they are about
7.5 cm long (VESIN & al. 1981). The changes from a typical
larval appearance, (e.g. slender body, sparse pigmenta-
tion) to a more adult appearance are gradual, and
individuals which are not fully pigmented at lengths up
to 8-10 cm may be found. The metamorphosis normally
takes place in spring/summer in the second year of life,
(i.e. when the offspring from the main spawning season
are about 12 months old).
The immature phase lasts from metamorphosis until
first maturation, which normally takes place in the third
or fourth year of life. Since most capelin spawn only
once and then die (see section 6.5.2), practically all growth
takes place during this stage. If the life history prior to
maturity is classified in this way, the adult phase only
lasts for a relatively short time interval, i.e. from maturity
is reached until spawning.
4.2.1 Distribution and migrations
4.2.1.1 General distribution
Usually, the capelin stock stays in the Barents Sea dur-
ing all life stages, but perform extensive seasonal migra-
tions. During winter and early spring, there is an ‘up-
stream’ spawning migration towards the coast of north-
ern Norway (Troms and Finnmark counties) and Russia
(Kola county) (Fig. 1), while during summer and autumn
there is a north- and north-eastward feeding migration.
During autumn, the adult capelin are found in both Atlantic
and Arctic water, with ambient temperature from –1 °C
to 2 °C, (GJØSÆTER & LOENG 1987). The fry, upon hatching
on the spawning sites at the coast, drift offshore with the
ocean currents, and spread out into the central and eastern
parts of the Barents Sea where the young capelin mainly
stay during the first months of their life.
The position of both spawning areas, nursery areas
and feeding areas vary with hydrographic conditions
(LOENG 1981, 1989a, 1989b; OZHIGIN & LUKA 1985;
OZHIGING & USHAKOV 1985; GJØSÆTER & LOENG 1987;
USHAKOV & OZHIGIN 1987). In ‘warm years’, character-
ised by strong inflow of Atlantic water from the west
and high temperatures in the Barents Sea, the distribu-
tion of capelin is displaced north- and eastwards. In 1973
and 1974, typical warm years, the capelin reached the
extremity of their distribution area off Franz Josef Land
the western extension of the distribution is quite variable.
In some years, the western limit is at 14-16°E, i.e. at
Vesterålen. In other years, the western limit is at about
30°E, at the Varanger Peninsula in Eastern Finnmark.
The northern extension at this time is normally at 73-
74°N. However, in years when distribution is easterly,
the northern limit is often displaced as far south as 71-
72°N.
In August-September, the 0-group capelin has a much
wider distribution, in most years extending eastwards
beyond 50°E. The western boundary is much more vari-
able. Thus, in some years, 0-group capelin is found to
the west of Spitsbergen, i.e. at about 3-5°E, while in
other years no 0-group capelin is found west of 25-30°E.
In all years except 1992, there is a fairly close correlation
between larval and 0-group distribution. In years with a
western larval distribution there will also be a western 0-
group distribution and an eastern larval distribution will
lead to an eastern 0-group distribution.
4.1.5 Depth distribution
The vertical distribution of larvae along the coast of Troms
and Finnmark in April-June was described by HOGNESTAD
(1969a, 1969b, 1969c, 1971). He used Clarke-Bumpus
plankton samplers to monitor the horizontal and vertical
distribution of capelin larvae. Hognestad’s observations
indicate that the newly hatched larvae reside in the upper-
most 25 m, but gradually disappear, to be subsequently
found in deeper layers. In 1967, the proportion of larvae
found in the uppermost 25 m decreased from 62 % to
20 % over a period of three weeks. In 1968, the corre-
sponding values were 56 % to 29 % over a period of 14
days. In 1969, however, 93 % of the larvae were found in
the uppermost 25 m in late April, and 56 % were still
found there in the beginning of June. A similar trend was
observed in 1970. The larvae found at depths greater than
25 m seemed to be more or less evenly distributed in the
layers 30-50 m and 50-75 m. It is not clear whether the
changed depth distribution was caused by depth-selective
mortality or if there was an active vertical migration of the
larvae (HOGNESTAD 1969a, 1969b, 1969c, 1971). Diurnal
changes in vertical distribution were not discussed in these
reports. SALVANES (1984) analysed the depth distribution
of capelin larvae in the years 1972-1975. She showed that
when the material from April, May and June was pooled,
all length groups seemed to be found at somewhat shal-
lower depths at night than during the day. The depth
distribution of larvae was studied during a capelin larval
survey in 1989 (FOSSUM & BAKKEPLASS 1989), using a sub-
mersible pump. Fifty litres of water were filtered through
a plankton net from 10, 20, 30, 40, 50, and 60 m depth at
0800, 1100, 1400, and 1700 UTC. The larvae were mainly
found from 20-40 m, and there was no sign of any vertical
migration in the twilight hours (from 1700 UTC). Neither

Gjøsæter – The population biology of capelin in the Barents Sea 463
4.2.1.2 Winter distribution
During winter (December-February), the capelin are
normally found south of the ice edge in the central parts
of the Barents Sea. In warm years, the overwintering
areas extend further to the east (Fig. 7A) than in cold
years (Fig. 7B). During January the maturing part of the
stock gradually segregates from the immature part,
occupying the southern part of the common distribution
area.
4.2.1.3 Spawning migration
During February, the maturing part of the stock begins
to move towards the coast. The migration routes and the
time and place where the spawning stock approaches the
coast are determined by hydrographic factors (MARTINSEN
1933; PENIN 1971, LUKA & PONOMARENKO 1983;
SHEVCHENKO & GALKIN 1983; OZHIGIN & LUKA 1985). In
most years, the migration follows two or even three dif-
ferent routes towards the coast. In warm years, the ma-
Fig. 7. Wintering migrations (arrows) of capelin in October
and wintering areas (hatched) in November-December in a typi-
cal warm year (A), and a typical cold year (B). The position of
the polar front is indicated by a continuous black line. Redrawn
from OZHIGIN & LUKA (1985).
Fig. 8. Wintering areas of immature (hatched) and mature
(cross-hatched) capelin and main routes of spawning migra-
tions (arrows) in January in a typical warm year (A) and a typi-
cal cold year (B). The position of the polar front is indicated by
a continuous black line. Redrawn from OZHIGIN & LUKA (1985).
and the northern coast of Novaya Zemlya. In ‘cold
years’, characterised by weak inflow and low
temperatures, such as in the period 1979-1982, the capelin
are found further to the south and west. Under such
hydrographic conditions, a part of the capelin stock is
also found west of Bear Island and along the west coast
of Spitsbergen.
LOENG (1981) compared the northern extension of the
capelin distribution area with temperature conditions at
100 m depth, and found linear correlation coefficients r
of 0.85-0.90. Similarly, OZHIGIN & USHAKOV (1985) com-
pared the northern limit of the feeding areas of capelin
(measured along a series of southwest-northeast transects)
with a number of different hydro-meteorological indices,
and found high correlations. On the basis of multiple
regression analysis they were able to forecast the position
of the main capelin concentrations with a fairly high
precision two months in advance.

464 Sarsia 83:453-496 – 1998
turing capelin mostly approaches the coast of Finnmark
and the Kola peninsula from the north-east (Fig. 8A),
while in cold years there may be additional spawning
migrations from the areas south of Bear Island to the
west coast of Troms and Finnmark (Fig. 8B). USHAKOV
& OZHIGIN 1987 showed that the capelin do not immedi-
ately respond to thermal changes in the water. There
appears to be a certain inertial, delaying responses with
respect to changes of temperature conditions. After a
series of cold years (1965-1969 and 1977-1982) the
spawning of capelin in warm years (1970-1971 and 1983-
1984) still continued to be restricted to areas near the
Norwegian coast.
4.2.1.4 Spawning
The location of capelin spawning areas have been de-
scribed on a general basis by several authors, e.g. RASS
(1933), PROKHOROV (1968), SÆTRE & GJØSÆTER (1975)
USHAKOV & OZHIGIN 1987, as well as in numerous cruise
reports and other documents dealing with capelin spawn-
ing in particular years. Based on the information con-
tained in these reports, and on material provided by N.G.
Ushakov at PINRO, Murmansk, charts have been pro-
duced where the spawning areas are indicated, together
with the resulting larval distribution in May-June (Figs
4-6). In the years from 1971 to 1984 the spawning areas
were located by sampling eggs with a Petersen grab. In
other years, the most probable spawning areas have been
more subjectively determined, e.g. from sampling of
spawning or newly spent capelin, observations of capelin
eggs in fish stomachs, and by observations of diving ducks
feeding on capelin eggs.
Before 1967, only sporadic information exists on the
location and extent of spawning areas. MØLLER & al.
(1961) describe the spawning migration in 1961 as con-
sisting of two separate approaches, one towards western
Finnmark and one towards eastern Finnmark and the
Kola coast. In 1966, the capelin migrated to the spawning
areas from the east, along the Kola coast towards eastern
Finnmark (LAHN-JOHANNESEN & al. 1966).
Apparently, the spawning in 1967-1970 took place
along the Norwegian coast from about 18-22°E to 32°E
(STRØM & VESTNES 1967; STRØM; & al. 1968; STRØM &
MONSTAD 1969; LAHN-JOHANNESEN & MONSTAD 1970).
Nothing is known about spawning on the Russian side of
the border in these years. According to the larval distri-
bution in 1968-1970 (Fig. 4) spawning has probably also
taken place further west than 18°E. In 1971, and in par-
ticular in 1972, spawning occurred along a wide area at
the Troms, Finnmark and Kola coasts, while in 1973-
1976 a more typical eastern spawning took place
(DRAGESUND & al. 1971; BJØRKE & al. 1972; GJØSÆTER &
SÆTRE 1973a; GJØSÆTER & al. 1974; GJØSÆTER &
MARTINSEN 1976; HAMRE & SÆTRE 1976; N.G. Ushakov,
PINRO, pers. commn). In 1972-1974 no information on
larval distribution exists, but for the other years the lar-
val distribution confirms the position of spawning. In
1977, spawning began near Vardø on 18 March and at
Fruholmen on 29 March. These were the main spawning
areas, but there was ‘occasional spawning on a smaller
scale along the coast’ (DOMMASNES & HAMRE 1977). Al-
though an extensive survey was carried out in 1978, no
spawning areas were located (DOMMASNES & al. 1979a).
Nonetheless, capelin larvae were detected off eastern
Finnmark and Kola in June, and some spawning must
have taken place in these areas (Fig. 5). In 1979, three
spawning invasions were detected (HAMRE & MONSTAD
1979), but only at the Varanger peninsula was spawning
confirmed by the detection of eggs. However, the larval
distribution (Fig. 5) shows that additional spawning must
have taken place further west. In 1980 the main spawn-
ing area was also near Vardø, but additional spawning
areas were found at Magerøy, Sørøy and Arnøy (HAMRE
& MONSTAD 1980). In 1981, 1982 and in particular in
1983, the main spawning areas were displaced westwards
(ALVHEIM & al. 1983a; ALVHEIM & al. 1983b; GJØSÆTER
1983). From 1984 onwards, spawning areas were no
longer detected by grab surveys on the Norwegian side
of the border. Based on information from other surveys
along the coast, spawning was found to take place off the
coast of Troms and Finnmark in 1984 (DOMMASNES 1984),
and along the Troms, Finnmark and Kola coasts in 1985
(GJØSÆTER 1985d). In 1986, mature capelin were only
found in the Varanger fjord on the Norwegian side of the
border, and observations of newly hatched larvae there
in late June show that some spawning took place in these
localities (SOLEMDAL & BRATLAND 1986), even if no larvae
were detected during the annual larval survey in June.
Some spawning was observed along the Rybachi peninsula
and further east (N.G. Ushakov, PINRO, pers. commn).
In 1987 no spawning was observed off the Norwegian
coast in spring, but on 31 July spawning was observed
outside Berlevåg (28°E) (G. Sangolt, Norwegian
Directorate of Fisheries, pers. commn). Furthermore, in
1987 and subsequent years spawning took place along
the Rybachi peninsula (N.G. Ushakov, PINRO, pers.
commn). In 1988, GJØSÆTER (1988) found indications of
spawning only off eastern Finnmark and in the Varanger
fjord in mid-April. However, observations of larvae all
along the Finnmark coast in June (Fig. 6) show that some
spawning must have taken place over a wider area.
In 1989, spawning seemingly took place from 17°E to
34°E (SANGOLT 1989; N.G. Ushakov, PINRO, pers.
commn). Judging from the larval distribution in June (Fig.
6), spawning also occurred over a large area in 1990, but
no surveys were carried out off the Norwegian coast in

Gjøsæter – The population biology of capelin in the Barents Sea 465
that year. GJØSÆTER (1991) found spawning and spent
capelin along the coast of Troms and Finnmark in March
1991, and SANGOLT (1992) observed spawning and spent
capelin along the coast, east of 24°E, in March 1992 (Fig.
6). In 1992, a spawning area was also detected near the
island Dolgiy (69°21'N, 58°57'E) on 22 July (S. Dahle,
Akvaplan AS, Tromsø, pers. commn). In 1993, spawning
capelin were observed along the coast of Finnmark, east of
Hjelmsøy (Fig. 6) (ANTHONYPILLAI & al. 1993). During
spring 1994, only scattered concentrations of capelin were
detected, except for one single concentration to the north-
east of the Varanger peninsula (GJØSÆTER 1994). The
distribution of the larvae found in June (Fig. 6) also indicates
an easterly spawning in 1994. During 1995 and 1996, no
surveys were carried out to locate capelin spawning off
the coast on the Norwegian side of the border.
4.2.1.5 Feeding migration, summer and autumn distri-
bution
The immature fish will generally move towards the south
from the area of overwintering and are found not far from
Fig. 9. Main capelin concentrations in June (hatched) in a typi-
cal warm year (A) and a typical cold year (B). The position of
the polar front is indicated by a continuous black line. Redrawn
from OZHIGIN & LUKA (1985).
Fig. 10. Main feeding migration routes of capelin in July-August
(arrows) and concentrations in September (hatched) in a typical
warm year (A) and a typical cold year (B). The position of the
polar front is indicated. Redrawn from OZHIGIN & LUKA (1985).
the coast in late spring. The spring bloom starts earlier in
coastal areas and on the banks than further offshore, and
the capelin utilise the food base in these areas in spring and
early summer. Spent fish that have survived the spawning
will probably join the immatures in these areas. In June
these concentrations are found further to the north (Fig
9A and B). When the ice starts to melt and the ice edge
recedes northwards, the capelin migrate northwards as
well. Following the receding ice edge is a phytoplankton
and then a zooplankton bloom, resulting from the
stabilisation of the relatively nutrient rich water masses
(SKJOLDAL & REY 1989). The capelin feed on this
zooplankton bloom, moving with it until the northern-
most feeding areas have been reached in September-
October. GJØSÆTER & al. (1983) presented a conceptual
model of the development of the processes linked to the
ice edge, where the processes taking place behind the
receding ice edge are conceived as a continuous ‘spring
bloom’ moving with the ice. These feeding areas will
change according to the hydrographic situation as shown
in Fig. 10A and B. In late October and November, the

466 Sarsia 83:453-496 – 1998
there are large, old individuals, but any systematic inves-
tigation of this bottom dwelling component has not been
undertaken. Therefore, it is unknown whether there is a
separate component of the stock mostly staying at near
bottom depth, or these are just individual fish staying
there for shorter periods.
4.2.2 Growth
The growth of capelin is extremely flexible with large
variations within and between years. Various authors
have studied the growth of Barents Sea capelin, OLSEN
(1968), PROKHOROV (1968), MONSTAD (1971), SHULGA &
BELUSOV (1976), MONSTAD & GJØSÆTER (1977), GJØSÆTER
(1985c, 1986), GJØSÆTER & LOENG (1987), SKJOLDAL & al.
(1992). The capelin grow to a maximum length of about
20 cm (males) and 18 cm (females), and the weight sel-
dom exceeds 50 grams (PROKHOROV 1968).
The growth has been found to vary with stock size
(ULLTANG 1975; GJØSÆTER 1986), with water tempera-
ture (SHULGA & BELUSOV 1976; GJØSÆTER & LOENG 1987)
and with geographical distribution (GJØSÆTER 1985c,
1986). The length- and weight-at-age of two year old
capelin, as measured during the annual acoustic surveys
carried out jointly by PINRO, Murmansk and Institute
of Marine Research (IMR), Bergen, have varied substan-
tially in the period 1972-1996 (Fig. 11). The general trend
is an increase in length and weight over this period. How-
ever, the last half of the 1980s and the period 1995-96
are characterised by high values while 1978-1979, 1984-
1985 and 1991-1993 are periods of low growth. The
decrease in mean length and weight, observed from 1990
to 1991, and the increase observed from 1993 to 1996
coincide with a sudden increase and decrease in the stock
size during these periods respectively. The general trend
of increasing mean lengths and weights during the period
1973-1996 also coincides with a general trend of
decreasing stock size in this period. Although mean
length and weight of two years old fish reflects the
accumulated growth over three growth seasons and,
therefore, cannot be directly compared to stock
abundance in one particular year, this indicates that the
growth is density dependent, or more precisely, stock
abundance dependent. There are, however, no clear-cut
relationships between stock size and individual growth
when analysed on a yearly basis. MONSTAD & GJØSÆTER
(1977), studying the growth of the year classes 1967-
1969, noted that their data showed no correlation between
growth and year class strength. GJØSÆTER (1986) came to
the same conclusion regarding growth of the year classes
1974-1985. He was not able to demonstrate density or
abundance dependent growth, neither between growth
and density within geographical sub-areas nor between
growth and abundance of the total stock in each year.
Both of these investigations were undertaken before the
capelin concentrations move back south- and south-west-
wards, and eventually overwinter south of the ice edge in
the areas indicated in Fig. 7A and B.
4.2.1.6 Vertical distribution
The vertical distribution of capelin larvae was discussed
in section 4.1.5. The vertical distribution and migration
of immature and adult capelin was studied by LUKA &
PONOMARENKO (1983) and LUKA (1984). The vertical
migrations of capelin change during the year. In spring
(March to April), when light reappears after the polar
night, the capelin descend into the near bottom layers at
sunrise, but ascend from these layers at the onset of
twilight in the evening. In summer (May-August) when
the light endures during 24 hours, the vertical migrations
become less distinct. However, some changes in vertical
distribution are still evident, but the migration rhythms
are not clearly diurnal. During September, when the
changes in light intensity between day and night become
more clear-cut, the diurnal rhythm of vertical migrations
reappears, but is most evident among the older age groups.
Apparently, the immature capelin remain in the upper
water layers both during day and at night. In late autumn
(October-November), with the onset of the polar night,
the amplitude of vertical migrations is reduced as the
light intensity decreases. At this time of the year, the
mature capelin descend to near bottom depths, disperse,
and start migrating south towards the spawning areas. In
December, mature capelin are mainly observed near the
bottom. In January, the pre-spawning capelin more often
form schools in intermediate and upper layer during their
migration to the spawning areas, especially at night. As
the light intensity increases in February, the diurnal
vertical migrations become more evident. Young capelin
(age groups 1 and 2) are often observed in the upper
layers during the winter period.
Although it is generally considered a pelagic species,
capelin is quite commonly caught in small numbers in
bottom trawl, both during day and night and throughout
the year. The general impression is that the capelin found
10
11
12
13
14
15
16
19731975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995
YEAR
MEAN LENGTH (cm)
4
6
8
10
12
14
16
18
20
MEAN WEIGHT (g)
Length
Weight
Fig. 11. Mean length- and weight-at-age of 2 years old Barents
Sea capelin measured during the annual autumn surveys in the
Barents Sea.

Gjøsæter – The population biology of capelin in the Barents Sea 467
dramatic stock collapses in the 1980s and 1990s and
similar analyses, including the year classes from these
periods, which are now being made, may produce different
results.
GJØSÆTER (1985c, 1986) found clear differences be-
tween growth of capelin in different parts of the Barents
Sea. He compared estimated growth rates in the current
growth season (based on back-calculation of length from
otoliths) for seven subareas of the Barents Sea, and found
that growth was always more rapid in the southern and
western parts than in the eastern and northern areas.
These differences persisted regardless of whether the
growth was generally high or low in one particular year.
These differences should probably be attributed either
to temperature conditions, to food abundance, or both.
GJØSÆTER & LOENG (1987) found correlation coefficients
r of 0.70 and 0.53 between capelin growth and ambient
temperature for two- and three-year-olds respectively,
when all the material from 1974-1985 was considered,
and r between 0.85 and 0.91 for within-year data. They
concluded that there is a general pattern of increased
growth in length with increasing temperature within the
observed temperature interval, but that any growth dif-
ferences observed and ascribed to temperature variations
will be a combination of direct, physiological effects and
indirect effects through increased availability of food.
SHULGA & BELUSOV (1976) found a negative correlation
between the length of two and three years old capelin
and temperature, using the mean temperature of the 0-
200 m layer along the Kola section in July in as an indi-
cator of temperature conditions in the Barents Sea. How-
ever, the relevance of using temperatures along the Kola
section as an indicator of the temperature conditions in
the various feeding areas of capelin, and to compare such
an indicator with accumulated growth during three to
four growth seasons, is questionable.
4.2.3 Maturation
MONSTAD (1971) established a maturity classification for
both sexes of Barents Sea capelin (Table 3) based on
macroscopic criteria. The classification was modified from
that presented by NIKOLSKY (1963). Monstad stated that
the classification was difficult, especially for males.
FORBERG (1982, 1983) made a histological study of
the capelin ovaries and established an alternative matu-
rity scale with 10 stages. This scale is currently used at
the IMR, Bergen for maturity classification of female
capelin, while the scale described in Table 3 is still in
use for the classification of males.
FORBERG (1982) classified oocytes in two growth
phases, first (FGP) and second (SGP) growth phase.
The FGP was further divided into three stages; the chro-
matin nucleolus stage (oocyte diameter – OD 5-15 µm),
the early perinucleolus stage (OD 15-150 µm), and the
late perinucleolus stage (OD 100-190 µm). This third
stage can be found in capelin larger than 10 cm throughout
the year in the Barents Sea, and is a ‘resting stage’. The
SGP was classified into five stages; yolk vesicle stage I
(OD 180-280 µm), yolk vesicle stage II (fat vesicle stage)
(OD 250-450 µm), primary yolk stage (OD 430-550
µm), secondary yolk stage (550-650 µm) and, finally,
the tertiary yolk stage (OD 650-1020 µm).
Table 3. Maturity scale used for both sexes prior to 1982, but after that only for males. From MONSTAD (1971).
Code Stage Description Females Males
1Juvenile (a) Gonads threadlike, sexes
difficult to separate
2 Juvenile (b) Gonads increasing in volume. Ovaries transparent, Testes transparent,
Sex can be determined without colour without colour
3 Maturing (a) Gonads opaque, blood vessels Ovaries with yellow/white Testes white or
can be seen grains with white spots
4 Maturing (b) Gonads increasing in volume. Ovaries pink or yellowish Testes light gray or white.
Blood vessels distinct white filling up 2/3 or more No milt-drops appear under
of body cavity pressure
5 Maturing (c) Ovaries occupy whole of body Testes gray. Milt runs with
cavity. Most eggs transparent some pressure applied
6 Spawning Running gonads
7 Spent Gonads emptyied. Some residual
eggs and sperm may occur
8 Spent/ Gonads small and collapsed
Recovering

468 Sarsia 83:453-496 – 1998
tion of spawners, while four-year-olds dominated in the
spawning stock in the other years during this period.
The age distribution in the spawning stock will obvi-
ously reflect the strength of the year classes taking part
in the spawning. However, since maturation is closely
linked to fish size (TJELMELAND 1985), the growth rate of
the immature stock will also affect the age distribution of
the spawning stock. In periods with a high growth rate of
the immatures, the year classes will mature and spawn at
a young age, while in periods of slow growth the spawning
will be postponed to an older age.
It is difficult to discriminate between early and late
spawners by visual inspection of the gonads during the
main capelin investigations in the autumn. However,
TJELMELAND & FORBERG (1984) developed a model for
that purpose. Because of the difficulties in obtaining
acoustic measurement of the amount of capelin spawn-
ing in the different seasons, it has not been possible to
test the predictive reliability of this model. Therefore,
FORBERG & TJELMELAND (1985) and TJELMELAND (1996)
have modelled the maturation of capelin as a
monotonically increasing function of fish length, ac-
cording to the equation
()
ml
ePP l
()=+−
1
1412
where m(l) is the proportion of fish in length group l,
measured during the autumn survey, that will mature and
spawn next spring, P2 is the length at 50 % maturity and
P1 is the shape parameter, the change of maturation with
length at P2. The shape parameter was determined from
a fit to the empirical maturation data according to the
maturation scale described in section 4.2.3, while the
length at 50 % maturity was determined by comparing
the immature stock in one year to the total stock in the
following year, assuming total spawning mortality.
TJELMELAND (1996) found that the estimated maturation
function fitted the maturation data remarkably well, for
P2 values in the range 13.5-14.5 cm. The parameter val-
ues varied both with maturity stage and age. It was found
that the most likely values of 50 % maturing length was
13.8 cm and 14.6 cm for females and males respectively.
FORBERG & TJELMELAND (1985) studied the spatial and
temporal variation of the maturing length P2 of Barents
Sea capelin during the period 1978-1983. They found a
significant variation in P2 between subareas of the Barents
Sea, but the variation was not consistent from year to
year. They estimated P2 for the different maturity stages
according to FORBERG (1982), and found that there was a
significant variation of P2 between years when ‘mature’
was defined as all female capelin in SGP. However, when
only those individuals classified in yolk vesicle stage II
and above were classified as mature, the corresponding
FORBERG (1982) found that the number of FGP oocytes
of various sizes always exceeded the number of synchro-
nously growing SGP oocytes, indicating that female
capelin have a potential for repeated spawning. It was
also found that the SGP lasted less than one year, and
consequently it was concluded that the presence of a
significant number of yolk vesicle oocytes or more mature
SGP oocytes was a good indication that the fish was
going to spawn within one year.
It has been observed in many years that the Barents
Sea capelin may have a prolonged spawning season. The
main spawning takes place in spring, while parts of the
stock may spawn in early summer and even in late sum-
mer (RASS 1933; MØLLER & OLSEN 1962a).
The age distribution of the spawning stock in differ-
ent years has been described by many authors.
Dommasnes (1985) reviewed the literature and presented
a synopsis of this information for the period 1954-1983
(Table 4). In a few years (1954, 1956-1959, 1965-1967)
the three years old fish represented the highest propor-
Table 4. Percentage age distribution of maturing and spawning
capelin during the period 1954-1983. From DOMMASNES
(1984b).
Year Age Number
23456
1954 1.6 78.1 19.9 0.4 0 238
1955 0 1.7 56.3 41.4 0.6 174
1956 0 52.4 42.7 4.9 0 61
1957 5.1 77.2 17.2 0.5 0 611
1958 0 88.8 11.2 0 0 98
1959 2.2 68.8 29.0 0 0 224
1960 0 40.5 58.8 0.7 0 973
1961 0.4 14.3 83.6 1.7 0 699
1962 0 12.4 67.0 20.4 0.2 917
1963 0 4.9 90.0 5.1 0 752
1964 0.2 4.2 52.6 43.1 0 -
1965 0.9 91.0 7.8 0.3 0 -
1966 32.3 64.3 4.4 0 0 300
1967 33.8 57.4 8.8 0 0 591
1968 2.6 35.7 61.7 0 0 863
1969 0 25.9 73.8 0.3 0 3380
1970 0 29.2 70.2 0.6 0 5304
1971 0 4.3 91.1 4.7 0 6215
1972 0 9.6 65.1 25.4 0 2450
1973 0 5.8 74.2 20.0 0 1837
1974 0.3 10.0 65.1 24.2 0.4 -
1975 0.1 9.9 79.3 10.4 0.3 -
1976 0.1 4.8 57.8 37.0 0.3 -
1977 0 5.5 58.5 32.2 3.9 -
1978 0 17.5 53.9 23.9 4.7 -
1979 0 22.4 62.9 13.7 1.1 -
1980 0 4.0 87.4 8.3 0.4 -
1981 2.6 6.0 61.7 27.7 2.4 -
1982 3.0 37.2 46.5 13.1 0.2 -
1983 0 21.2 63.9 14.1 0.9 -

Gjøsæter – The population biology of capelin in the Barents Sea 469
P2 was quite stable from year to year. They argued that
this group probably consisted of spring spawners only.
The observed variability between years could reflect ac-
tual differences in average maturing length or, more prob-
ably, differences in timing of sexual maturation. These
authors also compared P2 for two and three years old
capelin and found that in most years the maturing length
was significantly larger for the two-year-olds. A two
year old fish must have invested much more energy per
unit time in somatic growth than a three year old fish of
the same length. Therefore, competing energy
requirements may explain differences in maturity rates
and in length at maturing between fishes of different age
(FORBERG & TJELMELAND 1985).
Using the maturation model just described, and ac-
counting for natural mortality, growth, and fishing mor-
tality during the period from measurement to spawning,
the maturing stock in spring can be forecast based on the
measurements of the total stock in autumn.
4.2.4 Behaviour
The behaviour of capelin associated with the spawning
process was studied at the coast of Finnmark in the pe-
riod 1971-1975 and reported on by BAKKE & BJØRKE
1973; BJØRKE & al. 1972; GJØSÆTER & al. 1974; GJØSÆTER
& MARTINSEN 1976; GJØSÆTER & SÆTRE 1973a; SÆTRE &
GJØSÆTER 1975. Scuba divers observed the capelin at the
spawning grounds, and echo sounders were used to ob-
serve the capelin schools approaching the spawning
grounds.
The capelin approached the spawning grounds in dense
pelagic schools, consisting of several hundred tonnes of
fish. Males and females often formed separate schools.
When the schools reached the spawning grounds, they
settled to the bottom, and often formed a dense layer
above the sea floor. At the spawning grounds, the divers
observed two kinds of schools. The first type consisted
of more or less regularly oriented capelin swimming for-
ward or in circles. The distance between the individual
capelin in these kinds of schools was usually between 15
and 30 cm. The second type of schools was often
pyramidal and consisted of irregularly oriented fish. The
lower part of the schools, with a diameter of 3-5 m, was
close to the bottom. The mean distance between indi-
viduals was only 5 cm or less. Pre-spawning and spawn-
ing males were totally dominant in both types of schools,
and females were seldom observed on the spawning
grounds. The spawning act was never observed by divers,
possibly because it mainly takes place during night.
Spent males were occasionally seen at the spawning
grounds. They seemed to be in very bad condition, and
many were found dead on the bottom.
SEREBROV (1985) studied the schooling behaviour of
capelin in the Barents Sea, by means of underwater pho-
tography. He found that during daytime the average den-
sity of fish in the schools was 1.4 specimens m–3 (SD =
2.3) but could, at maximum densities, reach 15-20 speci-
mens per m3. At night, the density never exceeded 3.5
specimens m–3, and the average density was 0.8 speci-
mens m–3 (SD=0.74). The schools normally consisted of
sub-groups of 3-4 individuals and 5-7 individuals during
daylight and at night respectively, separated by an average
distance of about three body lengths. The average fish
density in these subgroups was about 4 and 20 times (for
day and night respectively) higher than for the school as
a whole. While moving quickly, schools condense because
these subgroups move towards each other, and eventually
the average density of the whole school increases to that
of the subgroups. In exceptional cases, e.g. when attacked
by predators, the school may become very dense with
150 specimens m–3.
Schooling by size was observed by OLSEN (1965), who
noted that the length distribution of capelin caught in
purse seines varied considerably from one catch to an-
other in the same area. This phenomenon was also re-
ported by GJØSÆTER & KORSBREKKE (1990), who ana-
lysed the age- and length-distribution of trawl catches of
capelin. The mean length of two years old capelin caught
together with one-year-olds was significantly lower than
the mean length of those caught together with older
capelin. Their interpretation was that two-year-olds
choose to school with fish of their own size. Conse-
quently small two-year-olds school together with smaller
fish (one-year-olds) while large two-year-olds school with
older, (and larger) capelin. The tendency to school by
size is probably linked to swimming speed, which is
known to increase with increasing body length (e.g.
HARDEN JONES 1968).
5 THE ECOLOGICAL ROLE OF THE CAPELIN
5.1 FOOD AND FEEDING
The capelin play a key role in the transportation of en-
ergy upwards through the food web in the Barents Sea. It
is the only fish stock capable of utilising the zooplankton
production in the central and northern areas including the
marginal ice zone. Young herring (Clupea harengus), in
some years present in large quantities in the Barents Sea,
seldom penetrates to the north of the 74th parallel. Polar
cod (Boreogadus saida) feed in the northern areas, but
this species is not a specialised plankton feeder and is
probably not able to compete with the capelin or take its
place in the food web.
Various authors have studied the feeding of capelin in
the Barents Sea (PANASENKO & SOBOLOVA 1980; LUND
1981; PANASENKO 1981, 1984; PANASENKO & NESTEROVA
1983; HASSEL 1984; HASSEL & al. 1991, AJIAD &

470 Sarsia 83:453-496 – 1998
PUSHCHAEVA 1992). Three groups of planktonic crusta-
ceans are important in the diet of capelin: copepods,
euphausiids and amphipods (Table 5). Their relative
importance varies with season, area, predator size and
year. Other important food items are Chaetognatha,
Limacina and Oikopleura.
The most intensive feeding takes place during the pe-
riod July to October. Young capelin do not feed during
the winter but, among capelin larger than 15 cm, more
than 50 % were found to have food in their stomachs.
The feeding activity increases during a short period prior
to spawning, but decreases again among that part of the
stock which actually spawns (LUND 1981).
The relative importance of different kinds of prey var-
ies with the length of the capelin. The importance of
copepods decreases with increasing capelin length, while
euphausiids and amphipods are most important for adult
capelin (Table 6).
Among the copepods, the most important species are
Calanus finmarchicus, Pseudocalanus elongatus,
Euchaeta norvegica and Metridia longa, which contrib-
ute to the diet throughout the year, while Calanus
hyperboreus, Microcalanus pusillus, Oithona similis and
Oncaea borealis may play an important role during one
or more quarters of the year.
Two species of Euphausiids, Thysanoessa inermis and
T. raschii and two amphipods, Themisto libellula and T.
Table 7. Seasonal variation of the index of stomach fullness (in ‰ of body weight), based on data from 1976-1982. After
PANASENKO (1984).
Month: 123456789101112
Length (cm)
9-10 - 2.97 - - - - - 2.83 9.76 15.97 - -
10-11 - 5.85 - - - - - 9.74 34.70 33.65 - -
11-12 - 16.46 - - 16.00 - - 12.38 29.36 29.57 - -
12-13 18.16 24.88 9.60 15.29 29.68 4.82 12.71 19.97 32.17 58.47 - -
13-14 11.21 14.18 10.48 8.41 17.19 11.95 24.81 21.78 26.36 11.35 - -
14-15 7.46 23.09 74.9 3.29 19.43 15.42 22.42 22.69 31.88 22.52 1.90 -
15-16 10.85 23.82 5.27 1.07 21.84 13.73 22.56 20.93 24.08 25.19 2.21 -
16-17 4.98 18.19 2.39 2.63 8.65 19.43 26.95 17.72 14.57 14.00 2.62 -
17-18 6.99 6.54 3.14 - - 3.97 - 18.74 22.80 19.70 1.06 -
18-19 2.60 1.16 - - - - - 14.00 2.13 - - -
abyssorum, play important roles in the diet of capelin.
The Euphausiids are mostly found in the southern parts
of the Barents Sea and are mainly eaten in spring and
summer, while the two Amphipods are associated with
arctic waters in the north-eastern parts and are mainly
preyed upon during autumn, when the capelin stock is
distributed farthest to the north (Table 5).
There is a considerable variation in feeding intensity
with season (PANASENKO 1984). There are two annual
peaks in stomach fullness (Table 7), one in February and
one in July-October. The first peak is associated with
the onset of krill predation in coastal areas, but the sec-
ond with predation on Calanus, krill and Amphipoda
during the main feeding season in the northern areas.
There seems to be no prominent daily variation in
food consumption. AJIAD & PUSHCHAEVA (1992) showed
that feeding peaked in the evening and a minimum occurred
around noon among all length groups, while LUND (1981)
found that during autumn, with limited light at night,
feeding was most intensive during the day. In summer,
when there is sufficient light both during day and night at
these high latitudes, feeding seemed to continue also at
night. PANASENKO (1981) found that in the second half
of August 1980, capelin fed intensively on concentra-
tions of large plankton (Euphausiacea, Copepoda,
Amphipoda) which at that time kept mainly to the near-
bottom layers during the twenty-four hours. The capelin
Table 5. Relative importance (% energy contribution) of prey
for 13-16 cm capelin sampled during the four quarters of a
year. After LUND (1981).
Spring Summer Autumn Winter
Copepods 0.6 8.9 28.5 24.2
Euphausiids 98.4 88.3 27.6 62.8
Amphipods 0.8 2.1 41.8 11.2
Others 0.2 0.7 2.1 1.8
Table 6. Relative importance (% of weight) of the three main
prey groups for Barents Sea capelin, for two length groups and
two seasons. After PANASENKO (1984).
Capelin below Capelin above
13 cm length 13 cm length
Food February- July- February- July-
components May October May October
Copepods 7.2 64.9 3.9 41.0
Euphausiids 87.1 22.7 89.7 43.3
Amphipods 2.1 7.0 4.0 12.4

Gjøsæter – The population biology of capelin in the Barents Sea 471
performed diurnal vertical migrations; descending during
daytime to the plankton rich water near bottom depth.
The stomach fullness increased during that period, and
was at a maximum between 04 PM and 08 PM. At night
the capelin ascended to midwater and scattered. During
this period, the stomach fullness decreased. The main
food objects were Euphausiids and Amphipods. There
are reasons to believe that the feeding behaviour of capelin
depends on the behaviour of the available prey. In August
1981, another three sets of twenty-four hour stations
were worked, and reported on by PANASENKO (1984). In
this case, a decline in the stomach fullness at night could
be seen at two of the stations, while the opposite was
true in one case (Fig. 12). The diet consisted of 50-70 %
Copepoda and 30-40 % Euphausiacea and Amphipoda,
but it is not clear at what depths the main food
concentrations were found in this case.
HASSEL (1984) compared capelin prey selection to the
available zooplankton in May and August 1981, collect-
ing samples along north-south transects. In May, food
supplies for the capelin seemed restricted, since only 1 g
per m2 was recorded in the upper 200 m. Of the capelin
stomachs investigated 30 % were empty and the mean
stomach fullness was 0.18 %. Euphausiids were the most
important food item, constituting 40 % by weight, while
the copepods ranged next with a contribution of about
32 %. In August, the stomach fullness had changed dras-
tically from May. At the northernmost station the mean
fullness index exceeded 8 %. The biomass of plankton
was also much higher in August than earlier, and was
increasing from south to north and reached about 20 g m–
2 at the northern limit of the capelin distribution area.
Comparison between stomach contents and plankton
composition in the sea indicated that capelin feed on the
available food without a strong selection of any particu-
lar food item. There was, however, a selection of prey
item by size. Small capelin preferred small copepods,
while larger fishes selected larger copepods.
AJIAD & PUSHCHAEVA (1992) studied capelin feeding in
the Goose Bank area in August 1989 and estimated the
daily ration during that period to be between 1.3 and 2.2
% of fish body weight.
5.2 FEEDBACK FROM CAPELIN FEEDING ON THE
ZOOPLANKTON COMMUNITY
The capelin stock is the key link between zooplankton
and higher trophic levels in the Barents Sea and the pre-
dation pressure from capelin seems to affect the biomass
of zooplankton in the feeding area (SKJOLDAL & al. 1992).
DALPADADO & SKJOLDAL (1996) found that following the
large reduction of the capelin stock between 1984 and
1987, there was a subsequent increase in the abundance
and biomass of the two euphausiid species Thysanoessa
inermis and T. longicaudata, while a decrease in krill
abundance and biomass was observed to follow the rapid
recovery and growth of the capelin stock from then until
1991. This suggests a predator-prey interrelationship
between capelin and krill and that the krill populations
are controlled by predation. In August 1985, the impact
of grazing from capelin on zooplankton was studied by
HASSEL & al. (1991). They found that the biomass of
zooplankton in the uppermost 100 m was much lower in
areas where capelin were present as compared to areas
without capelin, suggesting rapid depletion of the major
prey items by feeding capelin. The investigation showed
that when the capelin moved northwards during the sum-
mer feeding migration (cf. section 4.2.1.5) the ‘capelin
front’ had a capelin biomass more than three times that
of zooplankton in areas without capelin, and would have
the potential to graze down the available prey in 3-4 days.
Another feedback mechanism was discussed by
TIMOFEEV (1988). He found that organic substances of
sperm and seminal fluid, ejaculated by capelin males into
the water, could influence plankton productivity in the
spawning areas. His calculations showed that the increased
production of phyto- and zooplankton caused by these
substances, constituted from 0.3-3 % of the actual
productivity of coastal Barents Sea waters.
5.3 PRODUCTION
The stock size estimates, obtained during the various sur-
veys, do not give any direct information on the biomass
production of the capelin stock. The annual production
may be estimated as the weight of the catch taken during
the year, the weight of the remaining spawning stock
(which is lost through the spawning mortality), and the
output due to natural mortality other than spawning mor-
tality. GJØSÆTER (1996) gave estimates for this quantity
for the period 1973-1996, calculated by using the model
‘Capstock’ (cf. section 10.1). As shown in Fig. 13, the
annual biomass production is generally higher than the
0
10
20
30
40
50
60
12:00 16:00 20:00 00:00 04:00 08:00
TIME OF THE DAY
STOMACH FILLING DEGREE
St.no.1 St.no. 2 St.no. 3
Fig. 12. Daily variations of capelin stomach filling degree (stom-
ach weight in ‰ of fish weight) at three stations sampled every
fourth hour in August 1981. Based on data in PANASENKO (1984).

472 Sarsia 83:453-496 – 1998
0
1
2
3
4
5
6
7
8
9
1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996
YEAR
BIOMASS (MILL. TONNES)
"M output biomass"
Standing stock, autumn
Fig. 13. Stock and production in the Barents Sea capelin stock.
From GJØSÆTER (1996).
0
5
10
15
20
25
30
35
40
45
50
1947 1951 1955 1959 1963 1967 19 71 1975
YEAR
FREQUENCY OF
OCCURRENCE (%)
FO Cod
FO Haddock
Fig. 14. Frequency of occurrence (FO) of capelin in cod and
haddock stomachs during the period 1947-1977. Data from
PONOMARENKO & YARAGINA (1984).
standing stock as measured during the autumn surveys. In
periods with a high natural mortality, the removal from
the stock is larger than the production, and the stock size
decreases, while the stock size increases when natural
mortality is low and production exceeds the losses.
5.4 PREDATION
The capelin may be classified as a forage fish (GJØSÆTER
1996) with a key ecological role as an intermediary be-
tween the zooplankton level and higher trophic levels.
Its important role as prey for numerous predators, e.g.
other fish, seals, whales and birds has been known for
centuries (SARS 1879; HJORT 1914). The consumption of
capelin by some of these predators has been quantified,
while the knowledge of the consumption by others is
purely qualitative.
5.4.1 Fish predators
Cod (Gadus morhua L.) is the main predator of capelin.
The Polar Research Institute of Marine Fisheries and
Oceanography in Murmansk, Russia (PINRO) has run a
stomach sampling program on cod, dating back to the
beginning of this century (PONOMARENKO & PONOMARENKO
1975; PONOMARENKO, PONOMARENKO & YARAGINA 1978;
PONOMARENKO & YARAGINA 1978; PONOMARENKO &
YARAGINA 1984). The reports on the results of this sam-
pling program are mainly concerned with qualitative as-
pects, while consumption estimates are presented for the
period 1971-1973 and 1975-1981. The frequency of oc-
currence (the number of cod stomachs with capelin as
percent of the total number of analysed stomachs) is pre-
sented in Fig. 14 for the period 1947-1977. The observed
variation is considerable; the lowest value of 7 is recorded
in 1964 and the highest of 48 in 1969. The frequency of
occurrence varies in a cyclic fashion, with approximately
10 years between each peak. In addition to the frequency
of occurrence of various prey organisms, a degree of
stomach fullness on a 1-5 scale (prior to 1958 1-4) is
recorded (PONOMARENKO & YARAGINA 1978). The esti-
mates of capelin consumed by cod (PONOMARENKO &
PONOMARENKO 1975; PONOMARENKO & YARAGINA 1984)
are based on a daily ration estimated by NOVIKOVA (1962)
for a period of intensive feeding on capelin. Recalculated
for age groups of cod (PONOMARENKO & YARAGINA 1984),
this daily ration was 5.4 % of predator body weight for
3-4 year old cod, 3.4 % for 5-8 year old cod, and 1.7 %
for age groups 9-15 years. The results of these calcula-
tions are given in Table 8. These results are based on the
assumption that the daily rations cited above were
applicable during two months of intensive feeding (in
spring), while these rations were halved during the other
10 months of the year. Cod biomass was obtained from
Virtual population analyses (VPA) runs made by Arctic
Fisheries Working Group of ICES (ICES 1982a).
In 1984, the Institute of Marine Research in Bergen,
Norway launched a program to monitor the consumption
of various prey by cod. From 1987 the stomach sampling
program has been carried out in co-operation with the
Polar Research Institute of Marine Fisheries and
Oceanography in Murmansk, Russia, and in the period
1984-1995 nearly 79000 cod stomachs have been ana-
lysed. This analysis is quantitative, i.e. the weight of each
prey type in individual cod stomachs is recorded (MEHL
& YARAGINA 1992). BOGSTAD & MEHL (1997) presented
estimates of capelin consumption by cod, based on this
joint database for the period 1984-1995 (Table 8).
Larval capelin have been found in the stomachs of 0-
group cod (BELTESTAD & al. 1975; GJØSÆTER 1995). The
extent of this predation, and whether it is of any signifi-
cance for capelin recruitment is unknown.
Haddock (Melanogrammus aeglefinus L.) is a preda-
tor on capelin during parts of the year, mainly during
the capelin spawning period, when both the capelin and
their eggs are consumed by haddock (COLLETT 1903;
ZENKEVITZH 1963; SÆTRE & GJØSÆTER 1975; ANTIPOVA &
al. 1980; KOVTSOVA 1988). The frequency of occurrence
(FO) of capelin in haddock stomachs over the period

Gjøsæter – The population biology of capelin in the Barents Sea 473
1947 to 1977 is shown in Fig. 14. The FO is generally
smaller for haddock than for cod, but shows a similar
cyclic variation with maxima in the early 1950s, 1960s,
and 1970s. Although haddock was regularly observed to
feed on capelin eggs during the period 1971-1973, when
extensive investigations on the spawning grounds of
capelin were undertaken (SÆTRE & GJØSÆTER 1975), it
was not possible to assess the quanta of capelin eggs
consumed by haddock. However, these authors stated
that a significant part of the eggs spawned at depths
exceeding 100-150 m may be consumed.
The herring (Clupea harengus) occur as juveniles in
the Barents Sea, and while there, consume considerable
amounts of capelin larvae (HUSE & TORESEN 1995). The
resulting mortality of capelin larvae is thought to be a
major cause of the recruitment failure of capelin associ-
ated with years when abundant herring year classes are
growing up in the area (HAMRE 1988, 1991; HUSE 1994;
GJØSÆTER 1995; GJØSÆTER & BOGSTAD 1998). Based on
the contents of herring stomachs, sampled in the field in
1992 and 1993, and a simple predation model, HUSE &
TORESEN (1995) estimated the number of larvae consumed
by herring in the period 1984-1993 and the resulting
impact on capelin recruitment (Table 9). In 1992, the
frequency of occurrence of capelin larvae in herring stom-
achs was 5.6, while in 1993 it was 3.0. The average num-
bers of larvae per stomach were 3.1 and 1.8, respec-
tively. The length of the ingested larvae ranged from 8 to
25 mm. In both years there was a large horizontal and
vertical overlap between the juvenile herring and capelin
larvae. Applying the average frequency of occurrence
found in 1992 and 1993, the same model parameters, and
available abundance estimates of herring and capelin
larvae, the model was run for each year in the period
1984-1993 (Table 9).
Even if some of the underlying assumptions , as well
as the abundance estimates of herring and capelin larvae,
may be questioned, these results show that a substantial
part of the capelin larvae that survive until June-July,
may be removed by herring predation during the summer.
This is in good agreement with the observations in the
period 1983-1985 (FOSSUM 1992) and 1991-1993
(GJØSÆTER 1995), i.e. that in spite of the large amount of
capelin larvae recorded in early summer in these years,
few larvae survived until the 0-group investigations in
August, and the acoustic estimate of the one-year-olds in
autumn og the following year. The consequences of these
mechanisms for the recruitment of capelin are discussed
in section 6.1.
Other fish predators include Greenland shark
(Somniosus microcephalus) (COLLETT 1903), Greenland
halibut (Reinhardtius hippoglossoides) (COLLETT 1903),
Esmarks eelpout (Lycodes esmarki) (COLLETT 1903),
Thorny skate (Raja radiata) (ANTIPOVA & NIKIFOROVA
1990), Long rough dab (Hippoglossoides platessoides)
(BERESTOVSKI 1989), Deep sea Redfish (Sebastes
mentella) (DOLGOV & DREVETNYAK 1993) and various
rockfishes (Anarchicadidae) (COLLETT 1903). These are
Table 8. Consumption of Barents Sea capelin by cod (million
tonnes). Data sources: 1971-1973; PONOMARENKO & PONO-
MARENKO (1975). 1975-1981; PONOMARENKO & YARAGINA
(1984). 1984-1995; BOGSTAD & MEHL (1996).
Capelin consumed by cod (million tonnes)
Year 3-4 years 5-8 years 9-15 years Total
1971 4.76
1972 4.22
1973 5.44
1974 No data
1975 2.508 3.901 0.127 6.536
1976 3.087 3.432 0.125 6.644
1977 1.753 2.362 0.097 4.212
1978 2.138 1.816 0.116 4.070
1979 2.608 1.343 0.170 4.121
1980 1.515 2.577 0.148 4.240
1981 0.788 3.228 0.161 4.177
1982-1983 No data
1984 0.505 0.216 0.014 0.734
1985 1.179 0.423 0.017 1.618
1986 0.577 0.246 0.005 0.828
1987 0.183 0.043 0.000 0.225
1988 0.101 0.233 0.003 0.336
1989 0.169 0.422 0.001 0.593
1990 0.750 0.919 0.010 1.679
1991 1.879 1.163 0.053 3.093
1992 2.191 0.516 0.142 2.849
1993 2.434 1.058 0.151 3.644
1994 0.796 0.569 0.025 1.390
1995 0.470 0.330 0.003 0.801
Table 9. Estimated number of capelin larvae (1012) consumed
in the years 1984-1993 in three scenarios of different time-
length of predation (TP80 = 80, TP100 = 100, TP120 = 120
days). Proportion of estimated capelin larval abundance re-
moved by herring in the same number of days is also given.
After HUSE & TORESEN (1995)
Numbers Percentage
Year TP80 TP100 TP120 TP80 TP100 TP120
1984 0.53 0.67 0.85 7 8 10
1985 0.35 0.45 0.54 4 5 6
1986 0.10 0.12 0.15 - - -
1987 - - - 0 0 0
1988 - - - 0 0 0
1989 0.06 0.07 0.08 1 1 1
1990 0.12 0.14 0.17 1 1 1
1991 0.74 0.93 1.11 25 31 37
1992 2.16 2.70 3.24 30 37 44
1993 1.47 1.83 2.20 44 56 67

474 Sarsia 83:453-496 – 1998
mostly deep water fishes feeding on capelin when found
at near bottom depths, or following the capelin to the
coast during the spawning migration.
5.4.2 Seal predators
Among the seals, the main capelin predator is the harp
seal (Phoca groenlandica). NORDØY & al. (1995a) give an
estimate of the harp seal stock’s food consumption, based
on an analysis of stomach contents in 1993 and the
observed food intake of captive seals. Of a total annual
food consumption of 1 million tonnes, about 700 000
tonnes consisted of various fish species, of which about
250 000 tonnes consisted of capelin. This estimate is
based on a Barents Sea harp seal population of 600 000
individuals. The authors argue that the results may be
biased in that the proportion of fish in the diet is
underestimated, because most of the stomach material
was collected from seals caught in the pack ice where
amphipods dominated the diet.
The dependence of the harp seal upon capelin as food
was demonstrated during the capelin collapse in 1986-
1989 (see section 11). During those years, the harp seal
population invaded the coast of Northern Norway and
more than 100 000 seals drowned in gill nets during 1987
and 1988 (HAUG & NILSSEN 1995). Food shortage in the
traditional wintering areas was probably the main reason
for these invasions.
The ringed seals (Phoca hispida) may feed on capelin
during the autumn when capelin is distributed near ice in
the northern Barents Sea. However, there is evidence
that these seals mainly eat pelagic crustacean and polar
cod (HANSEN & al. 1996).
5.4.3 Whale predators
The minke whale is the main whale predator on capelin
(NORDØY & al. 1995b; HAUG & al. 1995a, 1995b). Based
on calculated energy requirements of growing and adult
animals, estimates of the diet composition in 1992 (HAUG
& al. 1995a), and a minke whale estimate of 87 000 ani-
mals, NORDØY & al. 1995b estimated a total consumption
of 355 000 tonnes of capelin. Since parts of the Northeast-
Atlantic minke whale stock do not enter the Barents Sea,
the use of the food composition data from the Barents
Sea on the whole stock may lead to an overestimate of
capelin consumption. On the other hand, the estimated
number of whales in this stock based on sighting surveys
in 1995, is considerably higher, 112 000 animals (TORESEN
1997).
Other whale species living in the area or visiting the
area during feeding migrations, like the blue whale
(Balaenoptera musculus), fin whale (Balaenoptera
physalus), sei whale (Balaenoptera borealis), humpback
whale (Megaptera novaeangliae), harbour porpoise
(Phocoena phocoena), killer whale (Orcinus orca),
narwhale (Monodon monoceros), white whale
(Delphinapterus leucas) and white-beaked dolphin
(Lagenorhynchus albirostris) may consume some capelin
(COLLETT 1903; HANSEN & al. 1996), but no quantitative
information is available.
5.4.4 Bird predators
The main fish feeding birds are the alcids, of which the
common guillemot (Uria aalge), and the puffin
(Fratercula arctica) have specialised in feeding on
schooling pelagic fish, while the Brünnich’s guillemot
(Uria lomvia) feeds on zooplankton organisms in addi-
tion to capelin and polar cod (SAKSHAUG & al. 1992).
MEHLUM & GABRIELSEN (1995) estimated the food re-
quirement of the total population of seabirds to be about
1.4 million tonnes annually in the Barents Sea, but the
proportion of capelin of this total food base is not avail-
able. The common guillemot represents about 10 % of
the total food requirement, and this species mostly eats
capelin. The Brünnich’s guillemot represents 55 %, but
has a much lower proportion of capelin in its diet. A
total mean capelin consumption in the order of 200 000-
300 000 tonnes could be a fair guess (GJØSÆTER 1997).
GJØSÆTER & al. (1972) and GJØSÆTER & SÆTRE (1975)
observed that diving ducks were feeding on capelin eggs
at spawning beds shallower than 50 m. At one site, where
dense concentrations of eggs (up to 200 eggs cm–2) were
found in a 600 000 m2 area, a flock of about 1300 ducks
(mostly king eider, Somateria spectabilis, together with
common eider, S. mollisima, and long-tailed duck,
Clangula hyemalis), were observed feeding during a period
of 20 days. The stomachs of 12 out of 13 ducks that
were shot were full of gravel and capelin eggs, and
practically no other food remains were found in their
stomachs. Although the ducks were quite numerous, cal-
culations indicated that the quanta of eggs consumed were
less than 2-3 % of the total egg production on the spawn-
ing ground (GJØSÆTER & SÆTRE 1975).
5.5 COMPETITION
Competition for food is most likely to take place be-
tween capelin and polar cod (Boreogadus saida), since
both are opportunistic feeders on pelagic zooplankton
(AJIAD & GJØSÆTER 1990), and there is a considerable
overlap between their distribution areas (MONSTAD &
GJØSÆTER 1987; GJØSÆTER & USHAKOV 1997). PANASENKO
& SOBOLEVA (1980) calculated similarity indices between
capelin and polar cod for different areas and seasons.
They found that in the central areas the greatest similar-
ity was observed in spring with respect to euphausiids,
while in the north-eastern areas the greatest similarity
was observed in summer with respect to copepods. If
competition between these stocks was important, a con-

Gjøsæter – The population biology of capelin in the Barents Sea 475
sequence should be a rise in the polar cod stock size in
periods when the capelin abundance was low. Although
there was an increase in recruitment to the polar cod
stock during the period of the first capelin stock collapse
in 1985-1989 (ICES 1996c), the total stock of polar cod
was not higher during this period than it was prior to or
after that period, nor did polar cod abundance increase
during the second capelin stock collapse in 1994-1996
(GJØSÆTER & USHAKOV 1997). Thus, it can be argued that
a competition for food between these two stocks is prob-
ably not strong enough to affect the size of the stocks to
any noticeable degree.
6 POPULATION DYNAMICS
6.1 RECRUITMENT
It is difficult to measure the recruitment to the capelin
stock. A series of larval abundance estimates exists, based
on annual surveys in June since 1981 (ICES 1997). A
series of 0-group trawl indices (combined indices based
on the area and density of the distribution), based on
annual surveys in August, is available from 1965 (ICES
1996c). These two time series are given in Table 10.
GUNDERSEN & GJØSÆTER (1998) compared these series to
the series of acoustic abundance estimates of one-year-
olds, and concluded that the larval estimates showed only
weak correlation with the acoustic estimates. The 0-group
index series was, however, highly correlated with the
estimates of one-year-olds (r2 = 0.75, p < 0.0001). The
larval abundance estimates probably measure the number
of larvae produced and, possibly, the spawning stock size,
but a high and variable mortality during the first months of
life renders them useless as measures of recruitment.
Based on data on larval abundance estimates, 0-group
indices and 1-group abundance estimates (ICES 1996a)
survival indices were calculated (Fig.15), showing that
during the period 1984-1986 and 1992-1993, survival
rates were low. The index series based on the 0-group
estimates and that based on the 1-group estimate are
quite consistent, apart from the years 1983-1985, when
the survival until age 1 (1 ½ year time span) was
seemingly much lower than until August in the first year
of life, indicating additional high mortality during the
second year of life. Furthermore, the indices for the year
class of 1994 deviates considerably. However, almost no
larvae were found in 1994, making the larval index uncertain
and, consequently, the calculated survival indices as well.
6.1.1 Causes of variation in recruitment
GJØSÆTER (1972) compared the observed recruitment with
spawning area and spawning time, but found no signifi-
cant correlation between either of these variables and
recruitment. In the studied period, 1951-1971, there is
Table 10. Abundance (1012) of capelin larvae in June (ICES
1997), and abundance indices of 0-group capelin in August
(ICES 1996c).
Larval 0-group
Year abundance index
1965 37
1966 119
1967 89
1968 99
1969 109
1970 51
1971 151
1972 275
1973 125
1974 359
1975 320
1976 281
1977 194
1978 40
1979 660
1980 502
Larval 0-group
Year abundance index
1981 9.7 570
1982 9.9 393
1983 9.9 589
1984 8.2 320
1985 8.6 110
1986 - 125
1987 0.3 55
1988 0.3 187
1989 7.3 1300
1990 13.0 324
1991 3.0 241
1992 7.3 26
1993 3.3 43
1994 0.1 58
1995 0.0 43
1996 2.4 291
only one example of a strong year class resulting from a
westward spawning, otherwise recruitment was poor or
medium. On the other hand, there was no example of a
poor year class resulting from an eastern spawning, but
several year classes at or above average strength. There
was a connection between early and westerly spawning
on the one hand, and late and easterly spawning on the
other. According to Gjøsæter’s data, it appeared that
recruitment was favoured by a late easterly spawning.
OLSEN (1968) also suggested a connection between east-
erly spawning and strong year classes of capelin.
BJØRKE & al. (1988) tested the hypothesis, that in years
of westerly spawning the larvae will be partly dispersed
in water masses carried to areas outside the Barents Sea,
expatriated and lost from the Barents Sea capelin stock.
They concluded that the hypothesis was probably too
simplistic, but such a mechanism could not be ruled out
as one of several factors controlling year class strength.
GUNDERSEN (1995) supported this hypothesis.
0
1
2
3
4
5
6
7
1981 1983 1985 1987 1989 1991 1993
YEAR CLASS
SURVIVAL INDEX
0-gr:larvae
1-gr:larvae
Fig. 15. Survival indices for capelin from the larval (June) stage
to the 0-group (August) stage, and from the larval stage to the
1-group (October) stage. Data from ICES (1996a).

476 Sarsia 83:453-496 – 1998
GUNDERSEN (1993a) compared some environmental
variables (temperature, salinity, inflow of Atlantic water
to the Barents Sea), experienced by the larvae to the
strength of the resulting year class. She found mainly
weak correlation between these variables and larval abun-
dance. Thus, larval abundance estimates were correlated
with the volume of water flowing into the Barents Sea in
June (r = 0.65, p = 0.08) and with the salinity in the
western part of the Barents Sea in June (r = 0.61, p =
0.06), but not with the temperature.
The relationship between stock and recruitment is es-
sential for all stock assessments on a time scale where
recruitment affects the fishable stock. For the capelin,
where most fish die after spawning and the main reason
for regulating the fishery is to allow a sufficient amount
of spawning to secure normal recruitment as far as pos-
sible, it is of vital importance.
GJØSÆTER (1972) used catch per unit of effort data
from the fishery in the spawning season as a measure of
spawning stock abundance, and calculated recruitment at
age 4 for the period 1951-1971. He found a positive
relationship between stock and recruitment, but with
considerable variation. Gjøsæter did not attempt to fit
any recruitment model to the data, but concluded that
both large and small spawning stocks appear to be able
to yield both good and poor recruitment. On the average,
small stocks seem to provide more recruits per parent
than larger spawning stocks.
HAMRE & TJELMELAND (1982) studied the stock-recruit-
ment relationship in the period 1974-1979, on the basis of
acoustic stock size estimates, using the number of two-
year-olds as a measure of recruitment. They fitted the
Beverton & Holt recruitment function R = (S · Rmax) / (S
+ S1/2) to the data, and obtained estimates of Rmax = 44.5
· 1010, and S1/2 = 0.43 million tonnes. All data points in
this period fitted this function quite well.
GJØSÆTER (1997b) investigated the stock-recruitment
relationship on the basis of estimates of spawning stock
and recruitment at age 1 in the period 1973-1995,using
the spreadsheet model ‘Capstock’ (see section 10.1) (Fig.
16). Some years show a very different recruitment/par-
ent stock ratio from the other years of the series. The
most notable exceptions are the years 1984, 1985, 1992,
1993, and 1994. In all of these years, young herring where
present in large numbers in the Barents Sea. Herring is
known to reduce the survival of capelin larvae (cf. sec-
tion 11) and, therefore, all years with more than 200 000-
300 000 tonnes of herring present in the Barents Sea
were marked on the stock-recruitment plot (Fig. 16).
These included 1991 and 1995 in addition to those men-
tioned above. A Beverton & Holt function was fitted to
the data when the herring years were excluded. The ob-
tained estimate of Rmax was 804 ·1012, and of S1/2 98 700
tonnes. The values of Rmax cannot be directly compared
in the above estimations because the difference in age at
recruitment, but the half-value of the function, S1/2, shows
that when this larger span of years is included the func-
tion rises much more sharply than it does when based on
the data from 1974-1979. It seems reasonable to con-
clude that the stock-recruitment relationship for capelin,
without herring in the Barents Sea, follows a Beverton &
Holt function reasonably well, while in years when young
herring are present recruitment fails completely. The two
years, 1991 and 1995 seem to contradict this hypothesis.
GJØSÆTER & BOGSTAD (1998) explain these seemingly
contradicting data points as being due to a small geo-
graphic overlap between young herring and capelin lar-
vae in these years.
SEREBRYAKOV & al. (1985) approached the stock-re-
cruitment problem in a slightly different way. Instead of
using the biomass of the spawning stock, they calculated
what they defined as the population fecundity (PF). The
PF was estimated on the basis of individual fecundity by
age (section 6.1.3), total number of fish, the maturity
ogive and sex ratio in the spawning stock. As an index of
recruitment, they used the number in each year class at
an age of three years. In the study period, 1972-1981,
the PF varied from 1.05 · 1015 (1974) to 3.18 · 1015 (1976),
and the recruitment from 0.18 · 1012 (1975) to 0.55 · 1012
(1972). The authors calculated survival rates SR (the ratio
between number of eggs and number of three-year-olds)
which varied from 0.008 % (1975) to 0.029 % (1972),
and defined three categories of survival conditions; fa-
vourable (SR > 0.020), average (0.014 < SR < 0.020), and
unfavourable (SR < 0.014). This approach to stock-
recruitment studies, taking into consideration not only
the total biomass of the spawners but the actual number
of eggs spawned, is interesting for two reasons. First, it
may reveal some of the mechanisms regulating the stock
recruitment relationship and, second, it could give a bet-
ter fit to the data points and thereby improve the models
Fig. 16. Stock-recruitment relationship for Barents Sea capelin.
Data from 1973 to 1996 are included. The years marked are
years with young herring present in the Barents Sea. These years
are excluded when the Beverton & Holt recruitment function
shown was fitted to the data. From GJØSÆTER & BOGSTAD (1998).
0
100
200
300
400
500
600
700
800
900
0 500 1000 1500 2000
SPAWNING STOCK SIZE (THOUSAND TONNES)
RECRUITMENT (BILLIONS ONE
-
YEAR-OLDS)
Herring not
present
Herring
present
Beverton &
Holt curve
91
92
93
83
94
85
95

Gjøsæter – The population biology of capelin in the Barents Sea 477
where this relationship is used. However, the period avail-
able to the authors was probably characterised by more
or less average survival conditions. In the following pe-
riod, the capelin larvae have probably experienced better
survival conditions (e.g. in 1988), as well as considerably
worse conditions (years with young herring in the Barents
Sea). Therefore, the different PF levels (sufficient PF to
produce an average year class under unfavourable
conditions, the critical PF which will not produce an
average year class even under favourable conditions etc.
as defined by SEREBRYAKOV & al. (1985) cannot be readily
used in management for establishing a target spawning
stock level.
6.1.2 Fecundity
The fecundity of Barents Sea capelin was studied by
POZDNJAKOV (1957), GJØSÆTER & MONSTAD (1973), GALKIN
& KOVALEV (1975), and HUSE & GJØSÆTER (1997). Fig. 17
summarises the results from these studies. There is an
approximately linear increase in fecundity with increas-
ing length. Most female spawners are in the length
interval 14 to 18 cm, and the fecundity in this length
range will vary from 6 000-10 000 to 14 000-18 000.
Whether the differences in fecundity by length, shown in
Fig. 17, are real differences between years and/or areas,
or artefacts caused by methodological differences, is
unknown.
6.2 AGE DETERMINATION
Scales do not develop in capelin during the first year of
life, and otoliths (sagittae) have been preferred for age
readings (HAMRE 1977). The annual growth zones or annuli
in capelin otoliths are relatively well pronounced, except
for the first one which may be difficult to detect,
particularly in older specimens. PROKHOROV (1963, 1965,
1968) claimed that even very small first rings should be
counted as the first winter ring because such otoliths
mainly stem from summer spawned capelin with a short
growth season before the first winter. However, some
earlier Russian authors (cited by PROKHOROV 1968) and
PITT (1958) omitted the central zone when counting an-
nuli in capelin. Growth studies were used to assess the
validity of ageing of capelin by BAILEY & al. (1977) in
Canadian waters and by HAMRE (1977) for the Barents
Sea stock.
By back-calculating lengths at deposition of the first
ring and comparing them with lengths from sampling,
BAILEY & al. (1977) reached the conclusion that the first
ring was a true winter-ring, but a second ring was found
to be deposited during the process of metamorphosis,
which normally occurred during its second year of life.
This ‘metamorphic check’ was found in 77 % of one and
two years old fish and in 44 % of adults, this decrease
0
5
10
15
20
25
30
13 14 15 16 17 18 19 20
FISH LENGTH (cm)
FECUNDITY (10
3
EGGS)
Pozdnjakov 1957
Gjøsæter&Monstad 1973
Galkin&Kovalev 1975
Huse&Gjøsæter 1997
Fig. 17. Fecundity by length for Barents Sea capelin. Data from
POZDNJAKOV (1957), GJØSÆTER & MONSTAD (1973), GALKIN &
KOVALEV (1975) and HUSE & GJØSÆTER (1997).
being interpreted as an obfuscation of the check by the
increasing opacity observed in older otoliths. The ab-
sence of the metamorphic check in some otoliths could
result from its deposition simultaneously with the first
or second annulus.
HAMRE (1977), working with Barents Sea capelin, stud-
ied this problem during the summer of 1976. Both 0- and
1-group capelin were sampled and length at deposition
of the first ring was back-calculated. He found that even
the smallest fish with one ring, a modal length of 5.8 cm
in early July and a mean length at deposition of the first
ring of 4.2 cm, were far too large to be the offspring of
that year’s spawning. It was found that even at the border
of the larval drift towards the north and east, the mean
length of the larvae was less than 2 cm. He concluded
that the first ring, even with a radius as small as 0.05 mm,
reflects the first year’s growth and should be counted as
the first winter ring. Otoliths from Barents Sea capelin
interpreted at the IMR, Bergen have, since then, been
aged according to these findings, and otoliths sampled
previously and interpreted differently have been reread.
The age reading method at PINRO, Murmansk also
complies with this interpretation, and there is a high
correlation between the interpretation of capelin otoliths
at these two institutes (GJØSÆTER 1985a).
GJØSÆTER (1985b) identified two main problems in the
age determination of capelin. The ‘first ring problem’,
i.e. the thickening of the otolith which causes the first
ring to disappear in older otoliths and the ‘false ring
problem’, false rings which may be present quite
frequently and are most pronounced in otoliths from
capelin more than three years old. He concluded that
although some individuals are probably assigned a false
age by the current method of age determination, the
practical consequences are small. He found no evidence
of a metamorphic check like that described for the capelin
in Newfoundland waters (BAILEY & al. 1977).

478 Sarsia 83:453-496 – 1998
Table 11. The parameters a and b in the equation W = a · lb calculated for immature and maturing males and females during the
four quarters of the year, based on data from 1968-1970. From MONSTAD (1971).
Winter Spring Summer Autumn
Male Female Male Female Male Female Male Female
Immatures a0.00041 0.00029 0.01623 0.03072 0.00177 0.00248 0.00154 0.00182
Immatures b3.85 4.00 2.40 2.17 3.32 3.19 3.35 3.29
Maturing a0.00240 0.00036 0.00011 0.00037 0.01411 0.00093 0.00038 0.00259
Maturing b3.25 3.89 4.33 3.88 2.64 3.61 3.89 3.17
TERESHCHENKO (1996) studied the capelin otolith struc-
ture with the aim to validate the age determination. She
confirmed that the hyaline winter ring is formed during
the winter cessation of growth from November to March-
April. In 9 % of the otoliths, the hyaline ring was present
already in September. The opaque summer zone is formed
during the period of intensive feeding and growth, from
May to September. There was evidence that in immature
fish the opaque zone begins to form earlier in spring than
it did in older fish. She also claimed that it is possible to
identify individuals which have spawned from those that
have not, based on the appearance of the otoliths. The
criteria for discrimination is a thickening of the hyaline
zone, increased contrast between hyaline and opaque
material, and a sharp reduction in fish growth rate among
those fish which have taken part in spawning. Based on
these criteria she found that in 1993, 10 % of the capelin
spawning stock were spawning for the second time (see
also section 6.5.2).
GJØSÆTER & MONSTAD (1982) studied the formation of
primary growth increments in the otoliths of capelin from
the Barents Sea. Otoliths from individuals with known
age, reared in a basin, as well as otoliths from field sam-
ples were analysed. They concluded that primary growth
increments do not form daily in adult capelin, but that
they probably do in capelin up to the age of about one
year. There was, however, evidence that fast-growing
specimens form more rings per unit time than do slow-
growing ones. The validation experiment was conducted
in a basin located in southern Norway, and the tempera-
ture in the basin reached 20 °C during the experiment.
This is much higher than the temperature experienced in
the natural habitat and the validity of the experiment is,
therefore, questionable.
6.3 GROWTH MODELS
Based on observed mean length-at-age, growth appears to
be fast until an age of three years, and then stops almost
completely. MONSTAD (1971) fitted von Bertalanffy
growth curves to such data and found the growth to be
described by the curves: Lt = 26.72 · [1 – e–0.07(t + 1.29)] for
males and Lt = 19.80 · [1 – e–0.12(t + 0.48)] for females (Fig.
18). However, the rapid decline in growth rate for older
fish is partly an artefact caused by the length dependent
maturation and high spawning mortality among capelin.
Individuals, surviving to an age of four or five years,
experienced the slowest growth rate and did not mature
and spawn at a younger age. The maturation length rep-
resents a kind of upper limit of the observed length in the
stock (GJØSÆTER 1985c, 1986). By back-calculating length
from otolith growth zones, MONSTAD & GJØSÆTER (1977)
and GJØSÆTER (1985c, 1986) found that the capelin
continues to grow at a relatively fast rate during the whole
of its life.
6.4 CONDITION
MONSTAD (1971) studied the length-weight relationship
for males, females, immatures and maturing fish sepa-
rately during the four quarters of the year in 1968-70. He
calculated the parameters a and b in the equation W = a · lb,
which are shown in Table 11. Maturing capelin exhibit the
largest weight for a given length in spring, and this is
most notable among the males. Among immature capelin
there are small differences between the sexes. In autumn
the curves for immature males and females are
approximately equal.
0
5
10
15
20
25
0 2 4 6 8 10121416182022
MONTHS
LENGTH (cm)
Males
Females
Fig. 18. vonBertalanffy growth curves for male and female
capelin. From MONSTAD 1971.

Gjøsæter – The population biology of capelin in the Barents Sea 479
Fig. 19. Estimated instantaneous natural mortality coefficients
M for the immature fish of age 2-3, 3-4 and average for all age
groups of capelin, based on annual acoustic abundance esti-
mates. Before 1983, the natural mortality was estimated only
as average values for all age groups. Data from ICES (1997).
6.5 MORTALITY
6.5.1 Mortality of immatures
The mortality of larvae is probably mainly caused by
predation, which is dealt with in section 5.4.1. The natu-
ral mortality of age one year and older capelin has been
studied by DOMMASNES (1981), TJELMELAND (1987b),
TJELMELAND & BOGSTAD (1993) and ICES (1997). DOMMASNES
(1981) found annual instantaneous natural mortality
coefficients (M) ranging from 0.35 to 1.03 for age 2-3
fish, 0.37 to 1.80 for age 3-4, and 1.16 to 2.66 for age 4-
5 fish, when the year classes 1971 to 1977 were analysed.
His estimates include spawning mortality, which explains
the high M for older age groups. An M of 2.66 will
remove 93 % of a year class during one year. TJELMELAND
(1987b) attempted to analyse the mortality of mature
capelin, caused by predation by cod, using cod stomach
data and abundance as well as distribution of cod and
capelin during the winter months (January to April). He
derived mortality coefficients of 0.65 ± 0.05 and 0.95 ±
0.10 for the years 1985 and 1986 respectively, where the
confidence intervals represent approximately 90 %
probability. By using the acoustic estimates of each age
group from the annual autumn surveys carried out jointly
by PINRO, Murmansk and IMR, Bergen in the period
1973 to 1996, taking account of catches and assuming
total spawning mortality, monthly natural mortality
coefficients have been calculated for the immature part
of each age group (ICES 1997). The results for the age
groups 1-2, 2-3 and average for all age groups are shown
in Fig. 19. Until around 1983, the natural mortality was
stable at about 0.05 month–1. From then on, M increased
sharply to 0.2-0.3 per month, corresponding to a 20-25
% removal of the individuals from the stock each month.
This was one of the main causes for the stock collapse in
1984-1987. The mortality decreased sharply during the
period 1987-1990, and was estimated at an unrealistically
low level in 1990. The same pattern was iterated in 1990-
1996; the natural mortality increased to very high levels
(partly causing the stock collapse in 1993-1995) once
more to return to values observed in the 1970s.
6.5.2 Spawning mortality
The question whether the Barents Sea capelin is
semelparous or iteroparous has been discussed by sev-
eral authors, i.e. PROKHOROV (1960, 1965, 1968),
MIGALOWSKIY (1967), MONSTAD (1971), FORBERG (1982),
BORISOV & DVININ (1986), OGANESYAN (1988). Further-
more, the question has been in focus in the ICES assess-
ment working group dealing with capelin assessment
(ICES 1991a).The main evidence for the capelin being
semelparous are that large quantities of dead capelin near
the spawning places in the spawning season is a common
phenomenon (RASS 1933; PROKHOROV 1960; MONSTAD
1971), that the spawning stock each year is totally
dominated by 3 or 4 years old capelin (PROKHOROV 1965,
1968; MONSTAD 1971; DOMMASNES 1985; ICES 1991a),
and that capelin older than 5 years are seldom found
(ICES 1991a). There seems to be a consensus that most
of the males die after spawning. The evidence for
regarding the female capelin as iteroparous is that not all
the oocytes mature when the capelin spawn for the first
time, rendering them physiologically capable of spawn-
ing for a second time (FORBERG 1982). Finally, the
physiological state of the females after spawning shows
that they are capable of surviving (OGANESYAN 1988),
and schools consisting of mostly spent females of which
a large proportion had begun feeding have been detected
several times just after spawning (PROKHOROV 1960;
GJØSÆTER 1989, 1990). However, a skewed proportion
of the sexes in favour of females among older fish, which
would be anticipated if males die while females survive
spawning, has not been detected. Thus, in the period
1969-1982 the average proportion of males among 4-
and 5-year-olds was 50.7 % and 52.6 % respectively
(ICES 1991a). Even if a certain proportion of fish sur-
vives the spawning process, the predation from cod and
other predators is intensive in the spawning areas, and
the chances of surviving a migration back to the feeding
areas are probably low (GJØSÆTER 1995). Disregarding
the fact that some females may survive spawning, it may
be concluded that only negligible amounts of fish in the
Barents Sea capelin stock will survive the first spawning
and live long enough to take part in a second spawning.
For stock assessment purposes the capelin may, there-
fore, be regarded as semelparous (ICES 1991a).
0.00
0.05
0.10
0.15
0.20
0.25
0.30
1973 1976 1979 1982 1985 1988 1991 1994
YEAR
NATURAL MORTALITY
COEFFICIENT M (MONTH
-1
)
2
3
Avr

480 Sarsia 83:453-496 – 1998
distribution area of juvenile capelin and could be the reason
for the disappearance of the capelin in the early 1960s
(HAMRE 1985). However, after 1962 the capelin have
visited the coast every year, although the abundance has
fluctuated, from almost zero to millions of tonnes.
7.2 ESTIMATES OF STOCK SIZE
Absolute estimates of stock size by acoustic methods
are available from 1972 onwards. A comprehensive
review of the time series of acoustic stock size estimates
is contained in articles by GJØSÆTER & al. 1998a & b.
That time series of stock size estimates are shown in Fig.
20.
Other kinds of stock size estimates include estimation
of spawning stock size from egg abundance (GJØSÆTER &
SÆTRE 1973b; DRAGESUND & al. 1973), from larval abun-
dance (DRAGESUND & al. 1973; SALVANES 1984), from
tagging experiments (DRAGESUND & al. 1973; DOMMASNES
1978a), from catch per unit of effort (GJØSÆTER 1972),
and from age material alone (OLSEN 1965). The abun-
dance of larvae in June has been estimated since 1981 and
an abundance index at the 0-group stage in August is
available from 1965 to present (cf. section 6.1).
The first attempt to estimate the relative size of the
spawning stock over a time period was made by OLSEN
(1965). He realised that the spawning stock size had
fluctuated widely, but because of the effects of changes
in availability, weather conditions, market demands and
other variables, catch per unit of effort estimates would
not give realistic indices of stock abundance. Conse-
quently, he developed a method, using the proportions
of spawners in different age groups in consecutive years
to directly estimate relative spawning stock size. For the
period 1959-1964, Olsen found that the relative strength
of age groups ranged from 0.1 to 2.2 (Table 12), with a
maximum in 1961 and a minimum in 1963-1964. He also
found a correlation between the relative spawning stock
size and the landings in the Norwegian fishery, and
concluded that the size of the catch was largely governed
by the size of the stock. In an attempt to analyse the
stock-recruitment relationship of capelin, GJØSÆTER
(1972) used catch per unit of effort in the Norwegian
fishery to assess the size of the spawning stock in the
period 1954 to 1964. He also provided a 3-scale classifi-
cation of spawning stock, i.e. poor, medium or strong in
the period 1954-1971 (Table 12). DRAGESUND & al. (1973)
estimated the spawning stock size on the basis of tagging
experiments, egg- and larval surveys, and acoustics in the
years 1970-1972 and DOMMASNES (1978a) estimated
spawning stock size on the basis of tagging experiments
in 1973, 1974 and 1975 (Table 12).
SALVANES (1984) analysed the abundance of capelin
larvae in the period 1972-1976, and estimated larval
7 ABUNDANCE
7.1 LONG TERM TRENDS
Estimates of capelin stock abundance before the 1970s
are mostly based on the success of the capelin fishery or
the fishery on young cod (called loddetorsk in Norwe-
gian, i.e. ‘capelin cod’) following the capelin to the coast
in spring, and are partly of anecdotal character. The
capelin spawning migration to the coast seems to have
been quite regular, but periods in which capelin failed to
approach the coast for spawning seem to have occurred
now and then. The earliest information describing such
periods dates back to 1799, when Sommerfelt (cited by
STERGIOU 1984) wrote: ‘They have the experience here
that it has abandoned the coast of Finnmark for many
years up to 16 or 20 in succession.’ The next official
record of such a period is a fishery report, according to
which there was no capelin fishery at all on the Finnmark
coast during the 1830-1840 period (STERGIOU 1984).
COLLETT (1903) mentions a longer period, 1819-1838,
but from 1840 until 1938 the capelin seem to have vis-
ited the Norwegian coast regularly. There was, however,
a period around the turn of this century when ‘the mass
approaches of capelin would seemingly not reach the
coast as before: The spawning these days is often taking
place so far from the coast that it is difficult for the
fishermen to find the capelin, and thereby to get enough
bait for the cod fishery’ (COLLETT 1903). From about
1868 there are annual records of the first date on which
the capelin appeared at the coast (MØLLER & OLSEN
1962a, 1962b). This record is continuos from 1868 to
1937, but from 1938 to 1942 no capelin appeared at the
coast of Finnmark. In 1962 this happened once again
(OLSEN 1965), and was caused by a serious decline in the
abundance of the year classes 1958-1960. This
temporary disappearance of capelin coincided with the
recruitment of two extraordinarily strong herring year
classes (1959 and 1960). It is likely, that the feeding area
of young herring in those years overlapped with the
Fig. 20. Acoustic estimates (billion individuals) of capelin from
1973 to 1996.

Gjøsæter – The population biology of capelin in the Barents Sea 481
mortality at the yolk sac stage and, finally, spawning
stock size. She used estimates of the age of the yolk sac
larvae from measurements of the length of the yolk sac
made by HELGESEN (1977). Salvanes’s estimates of
spawning stock size are somewhat lower than those
derived from the tagging data (Table 12). After this period,
acoustic methods represent the sole technique for stock
size estimation of adult Barents Sea capelin.
8 EXPLOITATION
The Norwegian capelin fishery has a long history. The
capelin were fished with beach seines at the coast of
Finnmark during the spawning season, and mainly used
as bait, fertiliser or animal food (SARS 1879; COLLETT 1903;
NITTER-EGENÆS 1967). From 1916 capelin were used for
meal and oil production in Finnmark, but it was not until
in the 1930s that a fishery for industrial purposes became
important (MØLLER & OLSEN 1962a). From the late 1950s
following the decline in abundance of the stock of
Table 12. Various estimates of spawning stock size during the period 1951-1976. Investigation
no. 1 and 2: GJØSÆTER (1972) from catch-per-unit-of-effort (million tonnes) and from various
subjective estimations respectively, no. 3: OLSEN (1965) from age-distribution (relative numbers),
no. 4: DRAGESUND & al. (1973) from tagging data (million tonnes), no. 5: DRAGESUND & al. (1973)
from egg and larval survey (million tonnes), no. 6: DOMMASNES (1978) from tagging data (million
tonnes), and no. 7: SALVANES (1984) from egg and larval surveys (million tonnes).
Investigation number
Year 1 2 3 4 5 6 7
1951 2.76 small
1952 1.24 small
1953 3.13 small
1954 6.34 medium
1955 4.75 medium
1956 9.74 strong
1957 7.08 medium
1958 5.96 medium
1959 8.48 strong 0.48
1960 6.64 medium 1.725
1961 13.17 strong 2.217
1962 small 0.46
1963 small 0.115
1964 small 0.09
1965 small
1966 strong
1967 strong
1968 strong
1969 strong
1970 strong
1971 very strong 5.8 3.2
1972 4.8 1.38
1973 2.2, 4.1 1.13
1974 1.1, 2.0 0.78
1975 1.1, 1.1 0.56
1976 1.11
Norwegian spring spawning herring, the fleet of purse
seiners increasingly focused their effort on the capelin,
and by 1957 purse-seiners had totally replaced the beach-
seines (STERGIOU 1984). From 1961 pelagic trawls were
also employed in the fishery, which at that time took
place in the spawning season only. Beginning in 1968, a
summer fishery rapidly developed in the open sea
(MONSTAD & GJØSÆTER 1972; ICES 1997).
The Russian (former Soviet) capelin fishery also has
long traditions, and was carried out with beach seines
and nets along the Kola coast during the spawning sea-
son (PROKHOROV 1965). From the early 1960s, purse
seines and pelagic trawls replaced the beach seines and
the fishing was expanded into the open areas of the
Barents Sea in the 1970s (STERGIOU 1984; ICES 1997).
In the 1970s, the fishery for capelin became of prime
importance to Norwegian and Russian fleets, with nearly
3 million tonnes landed in 1977. Since then, two stock
collapses have resulted in the fishing for capelin to be-
come more variable. Furthermore, the capelin fishery was

482 Sarsia 83:453-496 – 1998
0
50
100
150
200
250
300
350
400
450
1914 1919 1924 1929 1934 1939 1944 1949 1954 1959 1964
YEAR
LANDINGS (THOUSAND
TONNES)
Norway
Russia
Sum
A
0
500
1000
1500
2000
2500
3000
1965 1970 1975 1980 1985 1990 1995
YEAR
LANDINGS (THOUSAND
TONNES)
Norway
Russia
Others
Sum
B
Fig. 21. Landings of Barents Sea capelin during the period 1914-
1967 (A, see text for data sources), and during the period 1965-
1997 (B, ICES (1997)).
closed from 1987-1990 and again from 1994 to the present
time (GJØSÆTER 1995; ICES 1997). In Fig. 21A and B,
catch statistics are presented for the period 1914 to 1967
and for the period 1965 to 1997, respectively. The
landings increased sharply in the 1950s, but declined
almost to zero in 1962-1964. From 1965 the increase in
catch continued until the early 1970s. From 1972 to 1983
Norwegian landings fluctuated around 1.5 million tonnes,
while in this period the Russian landings increased sharply
and brought the total annual landings up to 2-3 million
tonnes. From 1984 catches decreased, partly because of
quota restrictions, but primarily because the stock
collapsed. The fishery was closed from autumn 1986
until autumn 1990, and the catches taken from the
recovered stock in 1991-1993 were relatively small com-
pared to the period 1970-1985. The fishery was closed
again in spring 1994 when a new stock collapse was
evident.
9 STOCK ASSESSMENT AND MANAGEMENT
Contrary to most fish stocks which are handled by as-
sessment working groups within the ICES system, the
capelin stock is not assessed by sequential population
analysis (SPA) (MOHN & COOK 1993). The main reason
is that most capelin spawn only once, and then die. The
VPA and other SPA techniques rely on records of catch-
at-age and measures or assumptions of natural mortality
to reconstruct the initial number of fish in each year
class. A useful estimate of the number of fish in a cohort
at the time when it was recruited can only be obtained if
the terminal size of the cohort is known (either if a long
time has passed since it was recruited to the fishery and
it can be assumed to be removed from the stock, or if the
terminal cohort size can be assessed by other means).
Another problem is that the capelin, being a forage fish,
is subjected to a very high natural mortality, as compared
to the fishing mortality, and SPA would, if only for that
reason, not give very accurate results. On the other hand,
the Barents Sea capelin stock is one of the few fish stocks
where direct absolute estimates of its size can be ob-
tained annually (GJØSÆTER & al. 1998a, b), and the need
for a survey-independent method (like SPA) for assess-
ment purposes is less pressing.
Various ad hoc methods have replaced SPA in the stock
assessment of capelin (TJELMELAND 1985, 1996; BOGSTAD
1997). These methods project the stock from the stock
estimate, obtained during the annual acoustic survey in
September-October forward in time until spawning in
March-April, taking into consideration natural mortality,
maturation and growth. It is then possible to analyse the
effect of a fishery on the spawning stock.
A catch quota regulation of the capelin fishery was in-
troduced in the winter season of 1974, when the Norwe-
gian quota was set at 720 000 tonnes. In spring 1978, a
Norwegian catch quota was set at 1 150 000 tonnes, and
in the autumn of that year at 350 000 tonnes. These catch
quotas were not based on an evaluation of the stock situ-
ation under different management regimes, but were rather
acts of precaution because the acoustic stock size esti-
mates showed that the maturing part of the population
was small and could be endangered by a free fishery.
In 1978, the USSR/Norwegian Fisheries Commission
agreed to regulate the Barents Sea capelin fishery bilater-
ally (HAMRE 1985). It was agreed that a TAC assessment
should be based on acoustic stock measurements carried
out jointly in the autumn, and that the assessment should
aim at a minimum remaining spawning stock of 500 000
tonnes (ANON 1978a, 1978b). A rule of a closed season,
lasting from 1 May to 14 August was introduced, and
the catch of juvenile capelin below 11 cm was limited to
15 % by weight. In 1981, the proportion of allowed
catch of fish below 11 cm was reduced to 10 %, and in
1984 the opening date of the autumn fishery was
postponed to 1 September. In 1981, a minimum mesh
size of 16 mm in capelin nets (both trawls and purse
seines) was introduced.
HAMRE & TJELMELAND (1982) estimated sustainable
yield of the Barents Sea capelin stock, using the model
‘Capelin’ (TJELMELAND 1985). Due to the mass mortality

Gjøsæter – The population biology of capelin in the Barents Sea 483
of post-spawners, the stock-recruitment relationship of
capelin is a matter of great importance in determining the
maximum sustainable yield (MSY), see section 6.1.1 for
a discussion of stock-recruitment relationships. HAMRE
& TJELMELAND (1982) calculated equilibrium sustainable
yield estimates under different fishing strategies, and
found that with no fishing the stock would stabilise at
about 5.5 million tonnes, of which about 1.8 million tonnes
would spawn. When fished only during the winter, the
maximum yield was 1.6 million tonnes for a parent stock
of 0.33 million tonnes, and when fished only during
autumn MSY was 1.7 million tonnes, achieved at an
equilibrium state of spawners of 0.45 million tonnes.
When fished both during autumn and winter, which usu-
ally has been done, a MSY of 1.65 million tonnes is
obtained for a spawning stock of 0.4 million tonnes.
In the paper by HAMRE & TJELMELAND (1982), ecologi-
cal considerations were also discussed in connection with
the yield estimates, a rather novel approach at that time.
Together with the yield, they calculated the ‘M-output
biomass’, the annual biomass output removed through
natural mortality. This is a rough estimate of that
proportion of the total production of capelin which is
available to capelin predators. This quantity increased
steadily for increasing spawning stock sizes, and is con-
siderably reduced when the yield is at its maximum (Fig.
22). The authors concluded that, irrespective of fishing
strategy, the yield curve was fairly flat near its maxi-
mum, and consequently there was not much difference in
yield for spawning stock sizes in the range 0.3 to 0.5
million tonnes. The M-output biomass, however, in-
creased by 0.5 million tonnes for this increase in spawn-
ing stock size. The authors also addressed the question
of when capelin should be harvested. They stated that
by managing an exclusive autumn fishery with an MSY-
strategy, about 2 million tonnes of capelin remain as
food for other stocks, excluding the biomass of
postspawners. However, when the catch is taken during
the winter, the importance of the capelin as a forage fish
increases and the M-output biomass becomes some 2.5
million tonnes, based on the same criteria. Consequently,
taking into consideration the ecological importance of
the capelin stock, this fact justified, in the authors’
opinion, an exploitation strategy which aims at an
equilibrium state of spawning stock of 0.5 million tonnes,
and a harvesting strategy where all, or at least the main
part, of the catch is taken during winter.
The strategy of leaving 0.5 million tonnes to spawn
was established in 1978 (ANON 1978a, 1978b) and has
been followed in the management of the capelin stock
since. The ICES working group dealing with capelin
adopted this strategy, as well as the recommendation,
originally given by a Soviet-Norwegian assessment work-
ing group (ANON 1983), that most of the quota should be
taken during winter (ICES 1984b). Little is lost in yield
by that strategy, while much is gained by a larger M-
output biomass and also through reduced uncertainty in
the stock assessment. A catch quota for an autumn fish-
ery has to be recommended on the basis of a stock prog-
nosis, made one year earlier and projected one and a half
year ahead of time. This implies a lot more uncertainty
than the procedure involved when recommending a quota
for the winter fishery, based on an autumn assessment
in the year before, i.e. 3-4 months before the fishery beg-
ins.
When the capelin stock declined severely during the
period 1983 to 1986, it was realised that the rationale
and the model underlying the management of the capelin
in the years prior to that period was inadequate (ICES
1995b; TJELMELAND & BOGSTAD 1993). It worked well
during the 1970s, when there were no herring in the Barents
Sea, and the recruitment to the cod stock was moderate.
The management in the years 1991 and 1993 introduced
a new element, i.e. the role of the cod as predator on
capelin was taken into consideration when modelling the
spawning stock size (BOGSTAD 1997). However, the
management objective from the pre-decline period, to
aim at a spawning stock level of capelin of about 0.5
million tonnes, remained unchanged. The method
developed by BOGSTAD & GJØSÆTER (1994) and used
during this period, was fairly simple and can only be
applied when the abundance of capelin is high. During
the last few years, more sophisticated models taking into
consideration both the mortality induced by cod on
immature and maturing capelin and the changes in capelin
recruitment when herring is present in the Barents Sea,
have been developed (TJELMELAND 1996; BOGSTAD 1997).
Because of the new collapse of the capelin stock in 1994
and subsequent years, new tools for the assessment of
the Barents Sea capelin stock will be introduced, where
both multispecies and environmental effects will be taken
0
1
2
3
4
5
6
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8
SPAWNING STOCK (MILL. TONNES)
BIOMASS (MILL. TONNES)
Yield
MOB
Stock
Fig. 22. Relationship between yield and spawning stock size
and between M output biomass and spawning stock size.

484 Sarsia 83:453-496 – 1998
based model of capelin distribution and migrations (GISKE
& al. 1992; FIKSEN & al. 1995). The former models the
distribution and migration of capelin on the basis of a
‘comfort function’ depending on temperature, food
density, light, ocean currents and capelin density, as-
suming that the capelin will migrate in the direction where
the comfort function has the steepest gradient. The latter
predicts growth and distribution, assuming that the
capelin will always behave in a way that maximises the
ecological fitness. Although there currently are problems
in fitting spatial distributions, obtained from these mod-
els, to observed distributions, this approach seems inter-
esting (BOGSTAD 1997).
10.2 MULTISPECIES MODELS
In the early 1980s an initiative was taken at the Institute
of Marine Research, Bergen to model stock interactions
in order to improve the scientific basis for management.
This initiative has resulted in the model MULTSPEC
and a family of related models (MEHL & al. 1986;
TJELMELAND 1987a, 1987b, 1992; BOGSTAD & TJELMELAND
1992; TJELMELAND & BOGSTAD 1993; ULLTANG 1995;
TJELMELAND 1996; BOGSTAD 1997; BOGSTAD & al. 1997;
TJELMELAND & BOGSTAD 1998), which are extensions to a
Beverton & Holt type of model (BEVERTON & HOLT 1957).
Output from these models is currently used by ICES in
the management of the Barents Sea capelin stock (cf.
section 9).
While these models are built on fundamental biological
and oceanographic explanatory variables using a ‘bot-
tom-up’ approach, a ‘top-down’ approach is used in a
model called ‘Systmod’, which considers the system on
a superior level, and models the most essential compo-
nents of the system from the top of the food-chain and
downwards (HAMRE & HATLEBAKK 1998). This model
only takes into consideration the most vital dynamics of
the key stocks, the interaction between them and the
influence of the environment, and may prove suitable in
fishing strategy analysis and for forecasts of catch and
stock developments.
Other types of multispecies models, involving capelin,
also exist. The model presented by REED & BALCHEN
(1992) is a component of a larger system of physical and
biological dynamic models describing the Barents Sea
ecosystem (BALCHEN 1976). Multispecies VPA or
MSVPA (POPE 1979; HELGASON & GISLASON 1979) has
also been tried for ecosystems of the Barents Sea (ICES
1996d, USHAKOV & al. 1992).
into account. However, the basic rationale behind the
capelin fishery management; a constant escapement strat-
egy for spawners, will not be abandoned. The size of the
target spawning stock will probably not change much
either, but it may change as a consequence of changes in
recruitment perspectives. It remains to be seen whether
these new models will enable managers to reach the ulti-
mate goal, i.e. to keep the capelin stock (and the other
exploited fish stocks in the area) at a level where a sus-
tainable production to the benefit of the whole ecosys-
tem, including man, is achieved.
10 MODELLING
Because of the complicated nature of an ecosystem, vari-
ous kinds of mathematical simplifications of the reality
are needed, both for the purpose of understanding the
processes taking place in the system, and to enable con-
struction of workable models for use in management of
human exploitation of the stocks. Many different kinds
of models exist, but the aim of the models mentioned
here is to aid in the assessment and management of stocks.
Some models are directly built as a tool to calculate fish
quotas, while others are built with the aim to increase the
general understanding of the ecosystem which, in the
long run, will aid in the management of the system. Since
capelin is one of the main components of the Barents Sea
ecosystem, it plays an important role, even in
multispecies and system models. I will not go into
modelling detail here, but point the way for further read-
ing.
10.1 SINGLE-SPECIES MODELS
At least two different types of single-species models
have been built, i.e. analytical population dynamics
models empirically tuned to fit the observed situation
and theoretical models involving dynamic optimisation
of behaviour. Examples of the former type is the model
‘capelin’, formerly used in conjunction with management
of the capelin stock (TJELMELAND 1985) and in part
‘Capstock’, a model resembling a VPA type of model
but based on annual stock size estimates (ICES 1991a).
These models are based on an annual stock size estimate,
the former predicts the future state of the stock as a
function of future catches, while the latter keeps track of
the stock history, giving estimates of stock abundance
and production through the year as well as of fishing and
natural mortality. Both of these models include a length
based maturation model and an empirical growth model.
Examples of theoretical models are a multi-dimensional
continuum model (REED & BALCHEN 1982), and a fitness-

Gjøsæter – The population biology of capelin in the Barents Sea 485
The explanation of the collapses should rather be
sought in recruitment failure, not because of shortage of
spawners but due to low survival of the larvae. This
hypothesis, now frequently called ‘Hamre’s hypothesis’
because it was first suggested and later on elaborated by
him (HAMRE 1985, 1988, 1991, 1994), is that the young
herring, when they appear in the Barents Sea, will graze
down the capelin larvae and thereby cause failure of re-
cruitment to the capelin stock. This hypothesis has gained
general acceptance after the mechanism behind this hy-
pothesis was verified in the field (HUSE 1994; HUSE &
TORESEN 1995), and it was shown that even though the
production of capelin larvae was ‘normal’ in years when
herring were abundant, these capelin year classes never
recruited to the stock, as measured by acoustic methods,
as one year and older fish (GJØSÆTER 1995). Based on
tank experiments MOKSNESS & ØIESTAD (1987) found that
0-group herring could also be responsible for mass mor-
tality of capelin larvae. However, the survival of the
capelin year classes 1983 and 1989, born in the same
years as strong herring year classes, suggests that the
predation on capelin larvae by 0-group herring is not as
important as that by age one to age four herring.
The following scenario for the major perturbations of
the capelin stock in the Barents Sea may then be out-
lined: A long period of cold ocean climate from 1977 to
1981, during which the capelin stock was large and sta-
ble, ended with a major inflow event of Atlantic water
during 1982 and early 1983 (SKJOLDAL & al. 1992). De-
spite the fact that this inflow, ending in winter 1983, left
the south-western Barents Sea filled with water contain-
ing low abundance of zooplankton, it probably enhanced
the survival conditions for herring and cod larvae,
spawned along the west coast of Norway, but growing
up in the Barents Sea. For the first time in 20 years, a
strong year class of herring grew up in the Barents Sea,
but left the area in May-June 1986. In 1984 and 1985,
this herring caused failure of recruitment to the capelin
stock. Concurrently, the mortality of older capelin was
increasing caused by enlarged food demand from a grow-
ing cod stock (MEHL 1991). TJELMELAND & BOGSTAD (1993)
estimated that the natural mortality of capelin had
increased from 0.04-0.09 month–1 in the period 1979-
1983, to 0.18-0.26 month–1 in the period 1985-1987. An
instantaneous mortality coefficient of 0.26 month–1
means that 96 % of the stock will die per year. At the
same time, the fishing pressure lead to increased fishing
mortality when the stock decreased in size (GJØSÆTER
1995). Since the plankton biomass was low in 1984, the
individual growth of capelin decreased (SKJOLDAL & al.
1992). These mechanisms, working together, caused the
11 A BRIEF HISTORY OF STOCK AND FISHERY
GJØSÆTER (1995) gave a short summary of the develop-
ment of the capelin stock, with emphasis on the question
of whether the fishery or other causes were the main
reasons for the large stock fluctuations observed in the
1980s and 1990s. As mentioned previously, little is
known about the size of the stock prior to 1973. But
from then on the stock size, distribution and composition
have been monitored during annual surveys in the autumn,
in the beginning by the IMR in Bergen, but later through
the joint effort of IMR and PINRO, Murmansk. The
stock size increased from 1973 to 1975 (Fig. 20), due to
the recruitment of the three large but slow-growing year
classes of 1971-1973. Since individual growth was slow,
these year classes matured at a relatively old age and,
consequently, a large number of immature fish was accu-
mulated in the stock (HAMRE 1991). When the bulk of
this accumulated stock spawned and died in 1975-1977,
the stock size decreased sharply. The next peak in the
stock size in 1980 was caused by high individual growth
rate in year classes of intermediate size. But once again
the stock declined when a large portion of it spawned
during 1980-1981. From 1984 to 1986, the stock was
reduced from about four million tonnes to 100 000 tonnes
(Fig. 20), and the possible causes of this will be dis-
cussed below. From 1988, the stock size increased sharply
to reach six to seven million tonnes in 1990 and 1991.
This was due to a very large year class in 1989, high
individual growth and low natural (and no fishing)
mortality. In 1993 a new stock collapse was evident, and
at present (spring 1997) the stock is still at a low level,
although there are signs of a new stock recovery.
Various explanations of the stock collapse(s) have been
suggested. HOPKINS & NILSSEN (1991) argued that the first
decline began as early as 1975 and that fishing was an
important contributory factor during the whole period.
HAMRE (1991), TJELMELAND & BOGSTAD (1993) and
GJØSÆTER (1995) all agree that a too high exploitation
pressure played a role in the two last fishing seasons
(autumn 1995 and winter 1996), but maintain that the
decreasing stock size prior to 1982-1983 can be explained
by a natural response to changes in recruitment, growth,
and maturation. GJØSÆTER (1995) argues that since the
impact of fishing will mainly be on reduced recruitment
caused by a too small spawning stock, and since the
production of larvae was at the same high level up to
1985, when the dwindling of the adult stock was already
evident, fishing could not be the main cause of the stock
collapse. This argument is supported by the stock devel-
opment in 1991-1994, when the catches were low and
the fishery was banned when the first signs of stock
collapse was evident.

486 Sarsia 83:453-496 – 1998
first stock collapse. When the herring left the Barents
Sea in spring 1986, the survival rate of capelin larvae
increased dramatically, but a small spawning stock size
probably impeded the production of a large year class
before 1989. Coinciding with the decline in the capelin
stock in 1985-1986, there was an increase in zooplankton
biomass to a maximum around 1987 (SKJOLDAL & al. 1992;
DALPADADO & SKJOLDAL 1996). Reduced predation
pressure by capelin was probably a major cause of this
increase in zooplankton biomass, particularly in the case
of larger species such as krill and amphipods. This high
abundance of food for the capelin was probably the main
reason for an observed increase in individual capelin
growth, even when the capelin stock was rapidly in-
creasing. Another reason for the rapid recovery of the
capelin stock was that the natural mortality of capelin
was drastically reduced (ICES 1997). This was probably
a result of a reduction in some major stocks of predators,
i.e. cod (MEHL 1991) and harp seals (HAUG & NILSSEN
1995; WIIG 1988).
A new major inflow event occurred in 1989 and 1990,
the cod and herring stocks recruited large numbers of
young fish to the Barents Sea, and a repetition of the
events described above was evident from 1992.
The decisive role, given to the herring in this scenario,
implies that the capelin stock should also have fluctu-
ated widely prior to the period of heavy exploitation of
the stock. Unfortunately, proofs of such events are lack-
ing, since the capelin was only observed when it ap-
proached the coast for spawning, and an absence of capelin
reported in one area could be caused by variations in
migration routes and spawning areas. However, there is
at least one example of a capelin stock collapse, prior to
the ones discussed above, i.e. that in 1962-1964. At that
time, the yield of the capelin fishery, which had been
steadily increasing from the early 1950s (Fig. 21A)
suddenly decreased and a rising trend was not re-estab-
lished until 1966. In 1962 the capelin did not approach
the Norwegian coast at all during the spring season (OLSEN
1965, 1968) and the spawning stock was small in the
period 1962-1965 (Table 11). The 1959 year class of
herring, which probably stayed in the Barents Sea in the
period 1960-1963, is known to be one of the most
numerous year classes in this century (HAMRE 1991).
Whether the absence of capelin from the coast, reported
to have occurred intermittently in previous times (SARS
1879; HJORT 1914; STERGIOU 1984; MØLLER & OLSEN
1962a) occurred simultaneously with the presence of
herring in the Barents Sea is unknown. However, it is
known that good recruitment to the herring stock is a
cyclic event. It is also known from Russian investigations
of cod stomachs, dating back to the turn of this century,
that herring and capelin have replaced each other in a
cyclic manner as the main prey for cod (PONOMARENKO &
PONOMARENKO 1975). This may be taken as supporting
evidence for the stock of capelin being small when the
herring were abundant and vice versa.
The antagonistic roles of herring and capelin in the
Barents Sea also suggest that the main reason why the
capelin stock was large and relatively stable and was able
to withstand a high fishing pressure from about 1965 to
1983, was the absence of herring during that period.
When the herring stock is at a ‘normal’ level, the chances
that no recruiting year classes will be rich and enter the
Barents Sea are probably low. It is possible and even
likely, that a situation with large perturbations of the
capelin stock, as witnessed during the last 15 years, is
the rule and that the large and stable capelin stock, seen
in previous years, is the exception rather than the other
way round.
Only continued effort of investigations of the Barents
Sea, to obtain direct evidence of the interactions between
fish stocks and between those fish stocks and their bio-
logical and physical environment, together with further
effort in modelling these processes, will throw more light
on the population biology of capelin and its place in the
ecosystem. We have learned much from previous investi-
gations, modelling activity and the experience gained by
observing the development of the fishery during this cen-
tury. But a lot more remains to be learned in the future.
12 ACKNOWLEDGEMENTS
I want to thank the editorial board of Sarsia and in particular
Jarl Giske for inviting me to prepare this review. I owe special
thanks to Johannes Hamre for his reading of and constructive
comments on the manuscript. I am also indebted to my good
friend and colleague Nikolay Ushakov of PINRO, Murmansk
for his help in pointing out to me Russian literature not readily
available outside Russia, for supplying me with unpublished
material and results of capelin research at PINRO, and for ren-
dering the abstract into Russian. Finally, two anonymous ref-
erees are thanked for constructive criticism of the manuscript.

Gjøsæter – The population biology of capelin in the Barents Sea 487
13 REFERENCES
Ajiad, A.M. & H. Gjøsæter 1990. Diet of polar cod,
Boreogadus saida, in the Barents Sea related to fish
size and geographical distribution. – ICES Council
Meeting 1990/G:48. 9 pp.
Ajiad, A.M. & T.Ya. Pushchaeva 1992. The daily feeding
dynamics in various length groups of the Barents Sea
capelin. – Pp. 181-192 in: Bogstad, B. & S. Tjelmeland
(eds). Interrelations between fish populations in the
Barents Sea. Proceedings of the fifth PINRO-IMR
Symposium, Murmansk, 12-16 August 1991. Institute
of Marine Research, Bergen, Norway.
Alvheim, O. 1985. Investigations on capelin larvae off
northern Norway and in the Barents Sea in 1981-
1984. – Pp. 171-184 in: Gjøsæter, H. (ed.): The
proceedings of the Soviet-Norwegian symposium on
the Barents Sea capelin. Institute of Marine Research,
Bergen, Norway, 1985
Alvheim, O., A. Dommasnes & O. Martinsen 1983a. Lodde-
undersøkelser vinteren 1982. (Capelin investigations
during the winter 1982). – Fisken og Havet 1983
(2):11-24. (In Norwegian with English abstract)
Alvheim, O., A. Dommasnes, O. Martinsen & S. Tjelmeland
1983b. Loddeundersøkelser i Barentshavet vinteren
1981. (Capelin investigations in the Barents Sea
during the winter 1981). – Fisken og Havet 1983
(2):1-10. (In Norwegian with English abstract)
Anon 1978a. Report of the meeting of scientists from
PINRO and the Norwegian Institute of Marine
Research, Bergen, April 1978. 26 pp.
— 1978b. Report of the 2nd USSR/Norwegian Working
Group Meeting on Barents Sea Capelin. Hammerfest,
October 1978. 19 pp.
— 1983. Report of the 7th USSR/Norwegian Working
Group Meeting on Barents Sea Capelin. Hammerfest,
October 1983.18 pp. + 4 appendices.
Anthonypillai, V., A. Dommasnes, H. Gjøsæter & J. Hamre
1993. Undersøkelser av sild og lodde i Barentshavet
vinteren 1993. (Investigations on herring and capelin
in the Barents Sea during winter 1993). – Rapport
fra Havforskningsinstituttet, Senter for Marine
Ressurser nr. 17 1993. 12 pp. (In Norwegian, with
English abstract)
Antipova, T.V. & T.B. Nikiforova 1990. Osobennosti pitaniya
kolyuchego skata v Barentsevom more. (The food
habits of the Barents Sea thorny skate). – Pp. 167-172
in: Kormovye resursy i pishchevye vzaimootnosheniya
ryb severnoj atlantiki. Murmansk. (In Russian)
Antipova, T.V., I.Ya. Ponomarenko & N.A. Yaragina 1980.
Seasonal and year-to-year dynamics of occurrence fre-
quency of the main food objects for haddock and
their stomach fullness degree in the Barents Sea. –
ICES Council Meeting 1980/H:21. 28 pp.
Bailey, R.F.J., K.W. Able & W.C. Leggett 1977. Evidence
for the presence of a metamorphic check in capelin
(Mallotus villosus) otoliths and implications for age
determination. – Journal of the Fishery Research
Board of Canada 34:2008-2014.
Bakke, S. & H. Bjørke 1973. Diving observations on Barents
Sea capelin at the spawning grounds off Northern
Norway. – Fiskeridirektoratets Skrifter Serie Hav-
undersøkelser 16:140-147.
Bakkeplass, K. & K. Lauvås 1992. Intern toktrapport F/F
“G.O. Sars” 25. mai til 21. juni 1992. (Unpublished
cruise report R/V “G.O. Sars” 25. May to 21 June 1992).
Institute of Marine Research, Bergen. (In Norwegian)
Balchen, J.G. 1976. Principles of migration in fishes. –
SINTEF technical reports no. 81.
Beltestad, A.K., O. Nakken & O.M. Smedstad 1975. Inves-
tigations on diel vertical migration of 0-group fish
in the Barents Sea. – Fiskeridirektoratets Skrifter Serie
Havundersøkelser 16 (7):229-244.
Berestovskij, E.G. 1989. Pitanie kambaly-ersha Hippoglossoides
platessoides limandoides (Bloch) v nekotorykh
rajonakh Barentseva i Norvezhskogo morej. (Feeding
of long rough dab Hippoglossoides platessoides
limandoides (Bloch) at some areas of the Barents and
Norwegian Seas). – Pp. 109-123 in: Podrazhanskaya,
S.G. (ed.). Sutochnye ritmy i ratsiony pitaniya
promyslovykh ryb Morovogo okeana. Moskva. (In
Russian)
Beverton, R.J.H. & S. Holt 1957. On the Dynamics of
Exploited Fish Populations. –Fishery Investigations
Series II, Vol. XIX. 533 pp.
Bjørke, H. 1976. Some preliminary results on food and
feeding of young capelin larvae. – ICES Council
Meeting 1976/H:37. 12 pp.
Bjørke, H., H. Gjøsæter & H. Loeng 1988. Drift of capelin
larvae - a population sink? Poster no 6 in the poster
series “The killing of the North-East Atlantic
plankton feeders: Man the murderer or accessory to
the crime?” – ICES Council Meeting 1988/R.
Bjørke, H., J. Gjøsæter & R. Sætre 1972. Undersøkelser på
loddas gytefelt i 1972. (Investigations at the spawning
grounds of capelin in 1972). – Fiskets Gang 58:710-
716. (In Norwegian with English abstract)
Bogstad, B. 1997. Multispecies models for the Barents Sea.
– Dr.scient. thesis, Department of Mathematics, Uni-
versity of Bergen/Institute of Marine Research,
Bergen.158 pp.
Bogstad, B. & H. Gjøsæter 1994. A method for estimation
of the consumption of capelin by cod in the Barents
Sea. – ICES Journal of Marine Science 51:273-280.
Bogstad, B., K. Hiis Hauge & Ø. Ulltang 1997. MULTSPEC
- A multispecies model for fish and marine mammals
in the Barents Sea. – Journal of Northwest Atlantic
Fishery Science 22:317-341.
Bogstad, B. & S. Mehl 1997. Interactions between Cod (Gadus
morhua) and its prey species in the Barents Sea. –
Proceedings. Forage Fishes in Marine Ecosystems.
Alaska Sea Grant College Program. AK-SG-97-01,
1997:591-615.

488 Sarsia 83:453-496 – 1998
Bogstad, B. & S. Tjelmeland 1992. A method for estimation of
predation mortalities on capelin using a cod-capelin
model for the Barents Sea. – Pp. 111-138 in: Bogstad, B.
& S. Tjelmeland (eds). Interrelations between fish
populations in the Barents Sea. Proceedings of the fifth
PINRO-IMR Symposium, Murmansk, 12-16 August
1991. Institute of Marine Research, Bergen, Norway.
Borisov, V.M. & Yu.F. Dvinin 1986. Some biochemical
indices and attempt of using them to estimate the
Barents Sea capelin natural mortality. – Dynamics
of commercial fish abundance. NAUKA Press,
Moscow: 131-141. (In Russian)
Buzeta, R., J. Hamre, I. Røttingen & M. Aksland 1975.
Loddeundersøkelser i Barentshavet i Juni 1975. (Capelin
investigations in the Barents Sea in June 1975). – Fiskets
Gang 62:49-53. (In Norwegian with English abstract)
Collett, R. 1903. Meddelelser om Norges fiske i aarene
1884-1901. Mallotus villosus (Müller) 1776. – Pp.
147-162 in: Christiana Vitenskabers Selskabs
forhandlinger 1403 Nr. 9. (In Norwegian)
Dalpadado, P. & H.R. Skjoldal 1996. Abundance, maturity
and growth of the krill species Thysanoessa inermis
and T. longicaudata in the Barents Sea. – Marine
Ecology Progress Series 144:175-183.
Dolgov, A.V. & K.V. Drevetnyak 1993. Some peculiarities of
feeding and feeding interrelations of deepwater redfish
(Sebastes mentella) in the Barents and Norwegian Seas.
– ICES Council Meeting 1993/G:60. 14 pp.
Dommasnes, A. 1978a. Survival of tagged capelin (Mallotus
villosus Müller) and estimates of the 1973, 1974 and
1975 spawning stocks from tag returns. –
Fiskeridirektoratets Skrifter serie Havundersøkelser
16:339-358.
— 1978b. Loddeundersøkelser i Barentshavet i juni-juli
1978. (Capelin investigations in the Barents Sea in
June-July 1978). – Fisken og Havet 79 (4):49-64.
(In Norwegian with English abstract)
— 1981. Stock size and mortality estimates for Barents
Sea capelin based on acoustic methods. – ICES
Council Meeting 1981/H:45. 11 pp.
— 1984. Intern toktrapport fra tokt med F/F “Eldjarn”
23. mars til 2. april 1984. (Unpublished cruise report
R/V “Eldjarn” 23 March to 2 April 1984). Institute
of Marine Research, Bergen 8 pp. (In Norwegian)
— 1985. Age distribution for spawning capelin 1954-
1984. A review of some selected literature. – ICES
Council Meeting 1985/H:36. 8 pp.
Dommasnes, A., B. Ellertsen, H. Loeng, F. Rey & S.
Tjelmeland 1979b. Intern toktrapport “G.O. Sars”
14. juni-24.juli 1979. (Unpublished cruise report
“G.O. Sars” 14. June-24. July 1979.) (In Norwegian)
Dommasnes, A. & J. Hamre 1977. Gyteinnsiget av lodde
vinteren 1977. (The spawning migration of capelin
during the winter of 1977). – Fisken og Havet 1977
(3):1-9. (In Norwegian with English abstract)
Dommasnes, A., L. Midttun & T. Monstad 1979a. Lodde-
undersøkelser i Barentshavet vinteren 1978. (Capelin
investigations in the Barents Sea during the winter
1978). – Fisken og Havet, 79 (1):1-16. (In Norwe-
gian with English abstract)
Dommasnes, A., T. Monstad & O. Nakken 1978. Lodde-
undersøkelser i Barentshavet i juni og juli 1977.
(Capelin investigations in the Barents Sea in June
and July 1977). – Fisken og Havet 78 (2):23-40. (In
Norwegian with English abstract)
Dragesund, O., J. Gjøsæter & T. Monstad 1971. Preliminary
results of the Norwegian capelin investigations during
winter and spring 1971. – ICES Council Meeting
1971/H:24. 19 pp.
— 1973. Estimates of Stock Size and Reproduction of
the Barents Sea Capelin in 1970-1972. – Fiskeridirek-
toratets Skrifter Serie Havundersøkelser 16:105-139.
Dushchenko, V.V. 1985. Soviet investigations of the intra-
specific structure of the Barents Sea capelin. – Pp. 223-
229 in: Gjøsæter, H. (ed.). The proceedings of the
Soviet-Norwegian symposium on the Barents Sea capelin.
Institute of Marine Research, Bergen, Norway, 1985.
Ellertsen, B., H. Loeng, F. Rey & S. Tjelmeland 1980.
Intern toktrapport “G.O. Sars” 18. juni-19.juli 1980.
(Unpublished cruise report “G.O. Sars” 18. June-19.
July 1980.) (In Norwegian)
Fiksen, Ø., J. Giske & D. Slagstad 1995. A spatially explicit
fitness-based model of capelin migrations (in) the
Barents Sea. – Fisheries Oceanography 4(3):193-208.
Forberg, K. 1982. A histological study of development of
oocytes in capelin, Mallotus villosus (Müller). – Jour-
nal of Fish Biology 20:143-154.
— 1983. Maturity classification and growth of capelin,
Mallotus villosus villosus (M), oocytes. – Journal of
Fish Biology 22:485-496.
Forberg, K. & S. Tjelmeland 1985. Maturity studies of
Barents Sea capelin. Variations in length at maturity
for female capelin. – Pp. 213-221 in: Gjøsæter, H.
(ed.). The proceedings of the Soviet-Norwegian
symposium on the Barents Sea capelin. Institute of
Marine Research, Bergen, Norway, 1985.
Fossum, P. 1992. The recovery of the Barents Sea capelin
(Mallotus villosus) from a larval point of view. –
ICES Journal of Marine Science 49:237-243.
Fossum, P. & K. Bakkeplass 1989. Loddelarveundersøkelsene
1989. Havforskningsinstituttets egg- og larveprogram
(HELP). (Investigations on capelin larvae during
1989. Institute of Marine Research’s egg and larval
program (HELP). Havforskningsinstituttet, Bergen,
Norway. ISBN 82-7461-014-8. 10 pp. (In Norwegian)
Friðgeirsson, E. 1976. Observations on spawning behaviour
and embryonic development of the Icelandic capelin.
– Rit Fiskideildar, Journal of the Marine Research
Institute Reykjavik 5(4):1-35.
Galkin, A.S. & S.M. Kovalev 1975. Plodovitost moivi
Mallotus villosus villosus Müller Barentseva Morja.
(Fecundity of capelin in the Barents Sea). – Voprosy
Ikhtiologii 15(4):646-651. (In Russian).
Giske, J., H.R. Skjoldal & D.L. Aksnes 1992. A conceptual
model of distribution of capelin in the Barents Sea. –
Sarsia 77:147-156.
Gjøsæter, H. 1983. Intern toktrapport fra tokt med F/F “Johan
Ruud” 17. april til 6. mai 1983. (Unpublished cruise re-
port R/V “Johan Ruud” 17 April to 6 May 1993). Insti-
tute of Marine Research, Bergen 7 pp. (In Norwegian)

Gjøsæter – The population biology of capelin in the Barents Sea 489
— 1985a. Report on the otolith workshop held at the
Soviet/Norwegian symposium on the Barents Sea
capelin. – Pp. 233-236 in: Gjøsæter, H. (ed.). The
proceedings of the Soviet-Norwegian symposium on
the Barents Sea capelin. Institute of Marine Research,
Bergen, Norway, 1985.
— 1985b. Standard sampling procedure for Barents Sea
capelin. A description of standard sampling technique
and methods applied to improve the
representativeness of the samples. – Pp. 119-133
in: Gjøsæter, H. (ed.). The proceedings of the Soviet-
Norwegian symposium on the Barents Sea capelin.
Institute of Marine Research, Bergen, Norway, 1985.
— 1985c. Growth of the Barents Sea Capelin of the
year classes 1975-1981. – Pp. 193-212 in: Gjøsæter,
H. (ed.). The proceedings of the Soviet-Norwegian
symposium on the Barents Sea capelin. Institute of
Marine Research, Bergen, Norway, 1985.
— 1985d. Toktrapport F/F “Michael Sars”, tokt nr. 7
1985. (Unpublished cruise report R/V “Michael Sars”,
cruise no. 7 1985), Institute of Marine Research,
Bergen. 16 pp. (In Norwegian)
— 1986. Growth of the Barents Sea Capelin compared
to stock size and geographical distribution. – ICES
Council Meeting 1986/H:38. 16 pp.
— 1988. Toktrapport F/F “Michael Sars”, 5. April-17.
April 1988. Unpublished cruise report R/V “Michael
Sars”, 5 April-17 April 1988), Institute of Marine
Research, Bergen. 8 pp. (In Norwegian)
— 1989. Toktrapport F/F “Eldjarn”, tokt nr. 5 1989.
(Unpublished cruise report R/V “Eldjarn”, cruise no.
5 1989), Institute of Marine Research, Bergen. 13
pp. (In Norwegian)
— 1990. Toktrapport F/F “Michael Sars” , tokt nr. 5
1990. (Unpublished cruise report R/V “Michael Sars”,
cruise no. 5 1990), Institute of Marine Research,
Bergen. 12 pp. (In Norwegian)
— 1991. Intern toktrapport fra tokt med F/F “G.O.
Sars” tokt nr. 3 1991. (Unpublished cruise report R/
V “G.O. Sars” cruise no. 3 1991). IT XII-91, Institute
of Marine Research, Bergen 15 pp. (In Norwegian)
— 1994. Lodda i Barentshavet vinteren 1994. (The
capelin in the Barents Sea during winter 1994). –
Fisken og Havet nr. 2-1994. 10 pp. (In Norwegian
with English abstract)
— 1995. Pelagic Fish and the Ecological Impact of the
Modern Fishing Industry in the Barents Sea. – Arctic
48:267-278.
— 1997a. The Barents Sea Capelin Stock (Mallotus
villosus): A Brief Review. –Proceedings. Forage Fishes
in Marine Ecosystems. Alaska Sea Grant College Pro-
gram. AK-SG-97-01, 1997: 469-484.
— 1997b. The Barents Sea capelin. A working document
to the Northern Pelagic and Blue Whiting Fisheries
Working Group Meeting 29.04-07.05 1997. 11 pp.
Unpublished manuscript.
Gjøsæter, H. & B. Bogstad 1998. Effects of the presence of
herring (Clupea harengus) on the stock-recruitment
relationship of Barents Sea capelin (Mallotus
villosus). – Fisheries Research 38:57-71.
Gjøsæter, H., A. Dommasnes & B. Røttingen 1998a. The
Barents Sea capelin stock 1972-1997. A synthesis of
results from acoustic surveys. – Sarsia 83:497-510.
— 1998b. Acoustic investigations of size and distribution
of the Barents Sea capelin stock 1972-1997. – Fisken
og Havet no. 4 1998:1-71.
Gjøsæter, H. & J. Gjøsæter 1986. Observations on the
embryonic development of capelin (Mallotus villosus
Müller) from the Barents Sea. – Fiskeridirektoratets
Skrifter Serie Havundersøkelser 18:59-68.
Gjøsæter, H., A. Hassel, H. Loeng, F. Rey & H.R. Skjoldal
1983. Preliminære resultater fra tokt i Barentshavet
med M/S “Lance” og F/F “G.O. Sars” i Mai/Juni 1983.
(Preliminary results from cruises with M/S “Lance”
and R/V “G.O. Sars” in May/June 1983. – Institute
of Marine Research, Bergen, Report FO8310. 21pp.
Gjøsæter, H. & K. Korsbrekke 1990. Schooling-by-size in
the Barents Sea Capelin Stock. – ICES Council
Meeting 1990/D:28. 9 pp.
Gjøsæter, H. & H. Loeng 1987. Growth of the Barents Sea
capelin, Mallotus villosus, in relation to climate. –
Environmental Biology of Fishes 20(4):293-300.
Gjøsæter, H. & N.G. Ushakov 1997. Acoustic estimates of
the Barents Sea stock of polar cod (Boreogadus saida).
– Proceedings. Forage Fishes in Marine Ecosystems.
Alaska Sea Grant College Program. AK-SG-97-01,
1997:485-504.
Gjøsæter, J. 1972. Recruitment of the Barents Sea capelin
1951-1961. – ICES Council Meeting 1972/H:24. 9pp.
Gjøsæter, J., K. Hansen, R. Sætre & T. Westergaard 1974.
Undersøkelser på loddas gytefelt i 1974. (Investigations
at the spawning grounds of capelin in 1974). –Fiskets
Gang 60:740-744. (In Norwegian with English abstract)
Gjøsæter, J. & O. Martinsen 1976. Undersøkelser på loddas
gytefelt i 1975. (Investigations at the spawning
grounds of capelin in 1975). – Fiskets Gang 62:30-
32. (In Norwegian, with English abstract)
Gjøsæter, J. & T. Monstad 1973. Fecundity and egg size of
spring spawning Barents Sea capelin. – Fiskeridirek-
toratets Skrifter Serie Havundersøkelser 16:98-104.
— 1982. Primary growth rings in otoliths of the Barents
Sea capelin. – ICES Council Meeting 1982/H:9. 13 pp.
Gjøsæter, J. & R. Sætre 1973a. Undersøkelser på loddas
gytefelt i 1973. (Investigations at the spawning
grounds of capelin in 1973). – Fisken og Havet
1973(3):4-7. (In Norwegian with English abstract)
— 1973b. The Use of Data on Eggs and Larvae for Esti-
mating Spawning Stock of Fish Populations with
Demersal Eggs. – Pp. 139-149 in: Blaxter, J.H.S. (ed.).
The Early Life History of Fish. Springer-Verlag, Berlin.
— 1975. Predation on eggs of capelin, Mallotus villosus
by diving ducks. – Astarte 7:35-41.
Gjøsæter, J., R. Sætre & H. Bjørke 1972. Dykkender beiter
på loddeegg. (Diving ducks feed on capelin eggs). –
Sterna 11:173-176. (In Norwegian)

490 Sarsia 83:453-496 – 1998
— 1980. Loddeundersøkelser i Barentshavet vinteren
1980. (Capelin investigations in the Barents Sea
during the winter 1980). – Fisken og Havet 1980(5):3-
22. (In Norwegian with English abstract)
Hamre, J. & I. Røttingen 1977. Loddeundersøkelser i Barents-
havet i juni-juli 1976. (Capelin investigations in the
Barents Sea in June-July 1976). – Fisken og Havet
1977(2):35-46. (In Norwegian with English abstract)
Hamre, J. & R. Sætre 1976. Gyteinnsiget av lodde vinteren
1976. (The spawning migration of capelin during
the winter of 1976). – Fisken og Havet 1976(2):43-
51. (In Norwegian with English abstract)
Hamre, J., & S. Tjelmeland 1982. Sustainable yield estimates
of the Barents Sea Capelin stock. – ICES Council
Meeting 1982/H:45. 24 pp.
Hansen, J.R., R. Hansson & S. Norris (eds) 1996. The state
of the European Arctic Environment. – EEA Envi-
ronment Monograph No. 3. European Environment
Agency, Copenhagen. 136 pp.
Harden Jones, F.R. 1968. Fish Migration. – Edward Arnold
Ltd., London. 325 pp.
Hassel, A. 1984. Quantitative and qualitative aspects of capelin
feeding in relation to zooplankton sources in the Barents
Sea in May and August 1981. –Contribution to the
joint soviet/Norwegian symposium on the Barents Sea
capelin, Bergen, Norway, August 1984. 14 pp. Unpub-
lished manuscript.
Hassel, A., H.R. Skjoldal, H. Gjøsæter, H. Loeng & L. Omli
1991. Impact of grazing from capelin (Mallotus villosus)
on zooplankton: a case study in the northern Barents Sea
in August 1985. – Pp. 371-388 in: Sakshaug, E., C.C.E.
Hopkins & N.A. Øritsland (eds). Proceedings of the Pro
Mare Symposium on Polar Marine Ecology, Trondheim,
Norway 12-16 May 1990. Polar Research 10(2).
Haug, T., H. Gjøsæter, U. Lindstrøm, & K.T. Nilssen 1995a.
Diet and food availability for north-east Atlantic minke
whales (Balaenoptera acutorostrata), during summer
of 1992. – ICES Journal of Marine Science 52:77-86.
Haug, T., H. Gjøsæter, U. Lindstrøm, K.T. Nilssen & I.
Røttingen, 1995b. Spatial and temporal variations
in north-east Atlantic minke whale Balaenoptera
acutorostrata feeding habits. – Pp. 225-239 in: Blix,
A.S., L. Walløe & Ø. Ulltang (eds). Whales, seals,
fish and man. Developments in Marine Biology 4.
Elsevier, Amsterdam.
Haug, T. & K.T. Nilssen 1995. Ecological implications of harp
seal Phoca groenlandica invasions in northern Norway.
– Pp. 545-556 in: Blix, A.S., L. Walløe & Ø. Ulltang
(eds). Whales, seals, fish and man. Developments in
Marine Biology 4. Elsevier, Amsterdam.
Helgason, Th. & H. Gislason 1979. VPA-analysis with
species interactions due to predation. – ICES Council
Meeting 1979/G:52. 10 pp.
Helgesen, I. 1977. Aldersbestemmelse og utvikling av
loddelarver (Mallotus villosus Müller) i plommesekk-
stadiet. (Age determination and development of capelin
larvae (Mallotus villosus Müller) at the yolk sac stage).
– Cand.real. thesis, Department of Fisheries Biology,
University of Bergen. 93 pp. (In Norwegian)
Gundersen, A.C. 1993a. Utbredelse av lodde (Mallotus villosus)
i Barentshavet i det første leveåret, 1981-1991. (Dis-
tribution of capelin in the Barents Sea during its first
year of life, 1981-1991). Hovedfagsoppgave i Fiskeri-
biologi. Institutt for fiskeri- og marinbiologi, Uni-
versitetet i Bergen 199. 106 pp. (Cand. scient. thesis,
University of Bergen, Norway). (In Norwegian)
— 1993b. Distribution of larval and 0-group capelin
(Mallotus villosus) in the Barents Sea in relation to
environmental factors 1981-1991. – ICES Council
Meeting 1993/H:15. 20 pp.
— 1995. Distribution of capelin during the first year of
life in the Barents Sea, in the period 1981-1991. – Pp.
23-34 in: Hylen, A. (ed.). Precision and relevance of
the pre-recruit studies for fishery management related
to fish stocks in the Barents Sea and adjacent waters.
Proceedings of the sixth IMR-PINRO Symposium,
Bergen, 14-17 June 1995. Institute of Marine Re-
search, Bergen, Norway, 1995.
Gundersen, A.C. & H. Gjøsæter 1998. A comparison between
abundance estimates of the Barents Sea capelin
(Mallotus villosus, Müller) at the larval, 0-group,
and 1-group stage, for the year classes 1981-1994. –
ICES Journal of Marine Science 55:95-101.
Hamre, J. 1977. On the interpretation of the central zone
in otoliths of capelin from the Barents Sea. – ICES
Council Meeting 1977/H:37. 23 pp.
— 1985. Assessment and management of Barents Sea
capelin. – Pp. 5-24 in: Gjøsæter, H. (ed.). The proceed-
ings of the Soviet-Norwegian symposium on the Barents
Sea capelin. Institute of Marine Research, Bergen, Nor-
way, 1985
— 1988. Some aspects of the interrelation between the
herring in the Norwegian Sea and the stocks of capelin
and cod in the Barents Sea. –ICES Council Meeting
1988/H:42. 15 pp.
— 1991. Interrelation between environmental changes
and fluctuating fish populations in the Barents Sea. –
Pp. 259-270 in: Kawasaki, T., S. Tanaka, Y. Toba &
A. Tanaguchi (eds). Long-term variability of pelagic
fish populations and their environment. Proceedings
of the International Symposium, Sendai, Japan, 14-
18 November 1989. Pergamon Press.
— 1994. Biodiversity and exploitation of the main fish
stocks in the Norwegian - Barents Sea ecosystem. –
Biodiversity and Conservation 3:473-492.
Hamre, J. & E. Hatlebakk 1998. System Model (Systmod)
for the Norwegian Sea and the Barents Sea. – Pp. 93-
115 in: Rødseth, T. (ed.). Models for multispecies
management. Physica-Verlag, Heidelberg.
Hamre, J. & A. Krysov 1994. Survey report R/V “Johan
Hjort” 6-30 June 1994. Unpublished cruise report,
Institute of Marine Research, Bergen.
Hamre, J. & T. Monstad 1979. Loddeundersøkelser i
Barentshavet vinteren 1979. (Capelin investigations
in the Barents Sea during the winter 1979). – Fisken og
Havet 1979(4):3-17. (In Norwegian with English ab-
stract)

Gjøsæter – The population biology of capelin in the Barents Sea 491
Hjort, J. 1914. Fluctuations in the great fisheries of Northern
Europe, viewed in the light of biological research. –
Conseil Permanent International pour L’Exploration
de la Mer, Rapports et Procès-Verbaux, vol XX, 228 pp.
Hognestad, P. 1969a. Forekomst av egg og yngel av fisk i
Vest- og Nord-Norske kyst- og bankfarvann i mai og
juni 1967. (Occurrence of fish eggs and fry in coastal
areas of northern Norway in May-June 1967). –
Fiskets Gang 55:10-12. (In Norwegian)
— 1969b. Forekomst av fiskelarver i Nord-Norske kyst-
og bankfarvann våren 1968. (Occurrence of fish
larvae in coastal areas of northern Norway in spring
1968). –Fiskets Gang 55:58-61. (In Norwegian)
— 1969c. Forekomst av fiskelarver i Nord-Norske kyst-
og bankfarvann våren 1969. (Occurrence of fish
larvae in coastal areas of northern Norway in spring
1969). –Fiskets Gang 55:739-743. (In Norwegian)
— 1971. Registrering av egg og larver av fisk i Nord-
Norske kyst- og bankfarvann med F/F “Asterias” våren
1970. (Registrations of fish eggs and larvae in coastal
areas of northern Norway with R/V “Asterias” in spring
1970). – Fiskets Gang 57:149-153. (In Norwegian)
Hopkins, C.C.E. & E.M. Nilssen. 1991. The rise and fall of
the Barents Sea capelin (Mallotus villosus): a multivariate
scenario. – Pp. 525-534 in Sakshaug, E., C.C.E. Hopkins
& N.A. Øritsland (eds). Proceedings of the Pro Mare
Symposium on Polar Marine Ecology, Trondheim, Nor-
way 12-16 May 1990. Polar Research 10(2).
Huse, G. 1994. Interactions between herring (Clupea harengus
L.) and capelin (Mallotus villosus Müller) in the Barents
Sea. – Cand. scient. thesis, Department of Fisheries and
Marine Biology, University of Bergen, 1994. 104 pp.
Huse, G. & H. Gjøsæter 1997. Fecundity of the Barents Sea
capelin (Mallotus villosus). – Marine Biology
130:309-313.
Huse, G. & R. Toresen 1995. Predation by juvenile herring
(Clupea harengus L.) on Barents Sea capelin (Mallotus
villosus Müller) larvae. – Pp. 59-73 in: Hylen, A. (ed.).
Precision and relevance of pre-recruit studies for fishery
management related to fish stocks in the Barents Sea
and adjacent waters. Proceedings of the sixth IMR-
PINRO symposium, Bergen, 14-17 June 1994. Institute
of Marine Research, Bergen, Norway. ISBN 82-7461-
039-3.
ICES 1965. Preliminary Report of the joint Soviet-
Norwegian investigations in the Barents Sea and
adjacent waters September 1965. – ICES Council
Meeting 1965. Herring Committee No 161. 15 pp.
— 1966. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1966. – ICES Council
Meeting 1966/H:23. 19 pp.
— 1967. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1967. – ICES Council
Meeting 1967/H:31. 18 pp.
— 1968. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1968. – ICES Council
Meeting 1968/F:33. 12 pp.
— 1969. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1969. – ICES Council
Meeting 1969/F:34. 14 pp.
— 1970. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1970. – ICES Council
Meeting 1970/H:34. 14 pp.
— 1971. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1971. – ICES Council
Meeting 1971/H:32. 13 pp.
— 1973a. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1972. – ICES Council
Meeting 1973/H:15. 15 pp.
— 1973b. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1973. – ICES Council
Meeting 1973/H:25. 27 pp.
— 1974. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1974. – ICES Council
Meeting 1974/H:33. 23 pp.
— 1975. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1975. – ICES Council
Meeting 1975/H:48. 23 pp.
— 1976. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1976. – ICES Council
Meeting 1976/H:43. 25 pp.
— 1977. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1977. – ICES Council
Meeting 1977/H:45. 25 pp.
— 1978. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1978. – ICES Council
Meeting 1978/H:33. 25 pp.
— 1979. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1979. – ICES Council
Meeting 1979/H:65. 26 pp.
— 1980. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1980. – ICES Council
Meeting 1980/G:53. 25 pp.
— 1981. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1981. – ICES Council
Meeting 1981/G:78. 27 pp.
— 1982a. Report of the Arctic Fisheries Working Group.
– ICES Council Meeting 1982/Assess:1. 53 pp.
— 1982b. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1982. – ICES Council
Meeting 1982/G:44. 27 pp.

492 Sarsia 83:453-496 – 1998
— 1983. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1983. – ICES Council
Meeting 1983/G:35. 28 pp.
— 1984a. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1984. – ICES Council
Meeting 1984/H:36. 28 pp.
— 1984b. Atlanto-Scandian Herring and Capelin Work-
ing Group Report. ICES Headquarters, 25-28 October
1983. – ICES Council Meeting 1984/Assess:4. 31 pp.
— 1985a. Preliminary Report of the joint International
0-group fish survey in the Barents Sea and adjacent
waters August/September 1985. – ICES Council
Meeting 1985/G:75. 27 pp.
— 1985b. Report of the Atlanto-Scandian Herring and
Capelin Working Group. ICES Headquarters 29 Oc-
tober to 1 November 1984. – ICES Council Meeting
1985/Assess:4. 25 pp.
— 1986. Preliminary Report of the International 0-
group fish survey in the Barents Sea and adjacent
waters August-September 1986. – ICES Council
Meeting 1986/G:78. 27 pp.
— 1987. Preliminary Report of the International 0-
group fish survey in the Barents Sea and adjacent
waters August-September 1987. – ICES Council
Meeting 1987/G:38. 31 pp.
— 1988. Preliminary Report of the International 0-
group fish survey in the Barents Sea and adjacent
waters August-September 1988. – ICES Council
Meeting 1988/G:45. 37 pp.
— 1989. Preliminary Report of the International 0-
group fish survey in the Barents Sea and adjacent
waters August-September 1989. – ICES Council
Meeting 1989/G:40. 40 pp.
— 1990. Preliminary Report of the International 0-
group fish survey in the Barents Sea and adjacent
waters August-September 1990. – ICES Council
Meeting 1990/G:46. 36 pp.
— 1991a. Report of the Atlanto-Scandian Herring and
Capelin Working Group, Copenhagen 8-12 April 1991.
– ICES Council Meeting 1991/Assess:17. 62 pp.
— 1991b. Preliminary Report of the International 0-
group fish survey in the Barents Sea and adjacent
waters August-September 1991. – ICES Council
Meeting 1991/G:50. 33 pp.
— 1992. Preliminary Report of the International 0-
group fish survey in the Barents Sea and adjacent
waters August-September 1992. – ICES Council
Meeting 1992/G:82. 33 pp.
— 1994. Preliminary Report of the International 0-
group fish survey in the Barents Sea and adjacent
waters August-September 1993. – ICES Council
Meeting 1994/G:3. 38 pp.
— 1995. Preliminary Report of the International 0-
group fish survey in the Barents Sea and adjacent
waters August-September 1994. – ICES Council
Meeting 1995/G:31. 36 pp.
— 1996a. Report of the Atlanto-Scandian Herring,
Capelin and Blue Whiting Assessment Working
Group 12-18 October 1995. – ICES Council Meeting
1996/Assess:9. 150 pp.
— 1996b. Preliminary Report of the International 0-
group fish survey in the Barents Sea and adjacent
waters August-September 1995. – ICES Council
Meeting 1996/G:30. 37 pp.
— 1996c. Preliminary Report of the International 0-
group fish survey in the Barents Sea and adjacent
waters August-September 1996. – ICES Council
Meeting 1996/G:31. 37 pp.
— 1996d. Report of the Multispecies Assessment Work-
ing Group. – ICES Council Meeting 1996/Assess:9.
150 pp.
— 1997. Report of the Northern Pelagic and Blue
Whiting Fisheries Working Group. ICES Headquarters
29 April-7 May 1997. – ICES Council Meeting 1997/
Assess:14. 192 pp.
Kennedy, C.R. 1979. The distribution and biology of the
cestode Eubothrium parvum in capelin, Mallotus
villosus, (Pallas) in the Barents Sea, and its use as a
biological tag. – Journal of Fish Biology 15:223-236.
Kovtsova, M.V. 1988. Fatness dynamics and feeding pecu-
liarities of haddock (Melanogrammus aeglefinus)
(Linne) in the Barents Sea. – ICES Council Meeting
1988/G:32. 15pp.
Krysov, A. & R. Toresen 1993. Joint survey report R/V
“Professor Marty” and R/V “G.O. Sars” 1.-28. June
1993. Unpublished cruise report, Institute of Marine
Research, Bergen. 12 pp + 7 pp appendix.
Lahn-Johannesen, J., D. Møller & S. Olsen 1966. Rapport
om tokt med “G.O. Sars” til Lofoten og Barentshavet
fra 17. februar til 1. april 1966. (Report from survey
with “G.O. Sars” to Lofoten and the Barents Sea
from 17 February to 1 April 1966). –Fiskets Gang
48:446-451. (In Norwegian)
Lahn-Johannesen, J. & T. Monstad 1970. Loddeinnsiget
1970. (The spawning migration of capelin 1970). –
Fiskets Gang 56:871-879. (In Norwegian)
Loeng, H. 1981. Nordlig utbredelse av lodde og temperatur-
forholdene i Barentshavet om høsten. (On the northerly
distribution of capelin and temperature conditions in
the Barents Sea in autumn). – Fisken og Havet 1981
(1):1-10. (In Norwegian with English abstract)
— 1989a. Ecological features of the Barents Sea. – Pp.
327-365 in: Rey, L. & V. Alexander (eds). Proceedings
of the sixth conference of the Comité Arctique Inter-
national 13-15 May 1985. E.J. Brill, Leiden.
— 1989b. The Influence of Temperature on Some Fish
Population Parameters in the Barents Sea. – Journal
of Northwest Atlantic Fishery Science 9:103-113.
— 1991. Features of the physical oceanographic conditions
of the Barents Sea. – Pp. 5-18 in Sakshaug, E., C.C.E.
Hopkins & N.A. Øritsland (eds). Proceedings of the Pro
Mare Symposium on Polar Marine Ecology, Trondheim,
Norway 12-16 May 1990. Polar Research 10(1).

Gjøsæter – The population biology of capelin in the Barents Sea 493
Loeng, H. & H. Gjøsæter 1990. Growth of 0-group fish in
relation to temperature conditions in the Barents
Sea during the period 1965-1989. – ICES Council
Meeting 1990/G:49. 9 pp.
Lund, A. 1981. Ernæring hos lodde, Mallotus villosus villosus
Müller, i Barentshavet. (Food and feeding of capelin
in the Barents Sea). – Cand. real. thesis, Department
of Fisheries Biology, University of Bergen, Norway.
128pp. (In Norwegian)
Luka, G.I. 1978. On systematics and biological groups of capelin
in the Barents Sea. –Trudy PINRO 4(1):36-49. (In
Russian)
— 1984. Diurnal vertical capelin migrations in the Barents
Sea. – Contribution to the joint soviet/Norwegian sym-
posium on the Barents Sea capelin, Bergen, Norway,
August 1984. 11 pp. Unpublished manuscript.
Luka, G.I. & V.P. Ponomarenko 1983. Seasonal and daily vertical
migrations and structure of capelin concentrations in the
Barents Sea. – ICES Council Meeting 1983/H:27. 16 pp.
Martinsen, G.V. 1933. K rasproctraneiio moyvi (Mallotus
villosus) v Barentsovom More. (On the distribution
of capelin in the Barents Sea). – Trudy Gasudarstvennogo
Okeanografitsjeskogo Instituta, Moskva IV(1):36-54.
(In Russian, with English summary)
Mehl, S. 1991. The northeast Arctic cod stock’s place in
the Barents Sea ecosystem in the 1980s: an overview.
– Pp. 525-534 in Sakshaug, E., C.C.E. Hopkins &
N.A. Øritsland (eds). Proceedings of the Pro Mare
Symposium on Polar Marine Ecology, Trondheim,
Norway 12-16 May 1990. Polar Research 10(2).
Mehl, S., O. Nakken, S. Tjelmeland & Ø. Ulltang 1986.
The construction of a multi-species model for the
Barents Sea with special reference to the cod-capelin
interactions. – Pp. 941-963 in: Alton, M. (ed.).
Proceedings of a workshop on comparative biology,
assessment and management of gadoids from the
North Pacific and Atlantic Oceans. Northwest and
Alaska Fisheries Center, Seattle.
Mehl, S. & N.A. Yaragina 1992. Methods and results in the
joint PINRO-IMR stomach sampling program. – Pp.
5-16 in: Bogstad, B. & S. Tjelmeland (eds). Interre-
lations between fish populations in the Barents Sea.
Proceedings of the fifth PINRO-IMR Symposium,
Murmansk, 12-16 August 1991. Institute of Marine
Research, Bergen, Norway.
Mehlum, F. & G.W. Gabrielsen 1995. Energy expenditure
and food consumption by seabird populations in the
Barents Sea region. – Pp. 457-470 in: Skjoldal, H.R.,
C. Hopkins, K.E. Erikstad & H.P. Leinaas (eds). Ecology
of Fjords and Coastal Waters. Elsevier, Amsterdam.
Migalowskiy, J.P. 1967. Analysis of the sexual cycle of the
females of capelin. – Trudy polarny nauchno-
issledovanie Insituta. morskogo rybnoi Khoz
Okeanografitsjeskogo 10:179-185. (In Russian, Eng-
lish summary)
Møller, D. & S. Olsen 1962a. Lodda og loddefisket. (The capelin
and the capelin fishery). – Fiskets Gang nr. 3 1962:27-36.
— 1962b. Norwegian capelin investigations. – ICES
Council Meeting 1962/Distant Northern Seas Committee
No. 34. 18 pp.
Møller, D., S. Olsen., K. Palmork & G. Vestnes 1961.
Loddeundersøkelser med F/F “Johan Hjort” 22.
februar-18. mars 1961. (Capelin investigations with
R/V “Johan Hjort” 22 February to 18 March 1961.)
– Fiskets Gang nr. 47:347-350. (In Norwegian)
Mohn, R.K. & R. Cook 1993. Introduction to Sequential
Population Analysis. Workbook. – NAFO Scientific
Council Studies, 17. 109 pp.
Moksness, E. 1982. Food uptake, growth and survival of
capelin larvae (Mallotus villosus Müller) in an outdoor
constructed basin. – Fiskeridirektoratets Skrifter Serie
Havundersøkelser 17:267-285.
Moksness, E. & V. Øiestad 1987. Interactions of Norwegian
spring-spawning herring larvae (Clupea harengus)
and Barents Sea capelin larvae (Mallotus villosus) in
a mesocosm study. – Journal du Conseil International
pour l’Exploration de la Mer 44:32-42.
Monstad, T. 1971. Alder, vekst og utbredelse av lodde
(Mallotus villosus) i Barentshavet og ved kysten av
Nord-Norge 1968-1970. (Age, growth and
geographical distribution of capelin (Mallotus villosus)
in the Barents Sea and at the coast of northern Norway
1968-70). – Cand. real. thesis, Department of Marine
Biology, University of Bergen. 80 pp. (In Norwegian)
Monstad, T. & J. Gjøsæter 1972. Norwegian capelin inves-
tigations. – Annales biologique 29:162-163.
— 1977. Growth of the Barents Sea capelin of the year
classes 1967-70. – ICES Council Meeting 1977/H:39.
15 pp.
Monstad, T. & H. Gjøsæter 1987. Observations on polar
cod (Boreogadus saida) in the Barents Sea 1973-
1986. – ICES Council Meeting 1987/G:13. 24 pp.
Mork, J. & E. Friis-Sørensen 1983. Genetic variation in
capelin (Mallotus villosus) from Norwegian waters.
– Marine Ecology Progress Series 12:199-205.
Nikolsky, G.V. 1963. The ecology of fishes. – Academic
Press, London. 352 pp.
Nitter-Egenæs, W. 1967. The capelin fishery. – Norwegian
Fishery and Maritime News 14(1):11-15.
Nordøy, E.S., P-.E. Mårtensson, A.R. Lager, L.P. Folkow &
A.S. Blix, 1995a. Food consumption of the Northeast
Atlantic stock of harp seals. – Pp 255-260 in: Blix,
A.S., L. Walløe & Ø. Ulltang (eds). Whales, seals, fish
and man. Developments in Marine Biology 4. Elsevier,
Amsterdam.
Nordøy, E.S., L.P. Folkow, P.-E. Mårtensen & A.S. Blix,
1995b. Food requirements of Northeast Atlantic minke
whales. – Pp. 307-318 in: Blix, A.S., L. Walløe & Ø.
Ulltang (eds). Whales, seals, fish and man. Developments
in Marine Biology 4. Elsevier, Amsterdam.
Novikova, N.S. 1962. Some data on the food rations of the
Barents Sea capelin by cod and haddock. – Doklady
Akademii Nauk SSSR,146(4):960-962.
Oganesyan, S.A. 1988. Relationship between abundance of
first- and repetitive-maturing specimens in the spawn-
ing portion from the Barents Sea capelin Mallotus
villosus (Müller) population. – Biology and farming
of Salmonidae, Leningrad:136-145. (In Russian)

494 Sarsia 83:453-496 – 1998
Olsen, S. 1965. Abundance estimates of Barents Sea capelin.
– Fiskeridirektoratets Skrifter Serie Havundersøkelser
13(8):76-82.
— 1968. Some results of the Norwegian capelin inves-
tigations 1960-1965. – Pp. 18-23 in: Blacker, R.W.
(ed.). Symposium on the ecology of pelagic fish
species in Arctic waters and adjacent seas. Conseil
Permanent International pour l’Exploration de la
Mer. Rapports et Procès-Verbaux des Réunions 158.
Ozhigin V.K. & G.I. Luka 1985. Some peculiarities of capelin
migrations depending on thermal conditions in the
Barents Sea. – Pp. 135-147 in: Gjøsæter, H. (ed.).
The proceedings of the Soviet-Norwegian symposium
on the Barents Sea capelin. Institute of Marine
Research, Bergen, Norway, 1985.
Ozhigin V.K. & N.G. Ushakov 1985. The effect of the
thermal conditions of the sea and atmospheric
circulation on the distribution of the Barents Sea
capelin feeding areas. – Pp. 149-156 in: Gjøsæter,
H. (ed.). The proceedings of the Soviet-Norwegian
symposium on the Barents Sea capelin. Institute of
Marine Research, Bergen, Norway, 1985.
Panasenko, L.D. 1981. Diurnal rhythms and rations of
capelin feeding in the Barents Sea. – ICES Council
Meeting 1981/H:26. 14 pp.
— 1984. Feeding of the Barents Sea capelin. – ICES
Council Meeting 1984/H:6. 16 pp.
Panasenko L.D. & V.N. Nesterova 1983. Stomach fullness
and condition factor of capelin under different food
supply. – ICES Council Meeting 1983/H:24. 12 pp.
Panasenko, L.D. & M.S. Sobolova 1980. Food interrelations
between the Barents Sea capelin and polar cod. –
ICES Council Meeting 1980:G:23. 15 pp.
Penin, V.V. 1971. On the Influence of Hydrometeorological
Factors on the Distribution of Capelin in the Spawning
Period and the Time of their Approach to the
Southern Barents Sea Coast. – ICES Council Meeting
1971/H:8. 7 pp.
Pitt, T.K. 1958. Distribution, spawning and racial studies
of the capelin Mallotus villosus (Müller) in the
offshore Newfoundland area. – Journal of Fishery
Research Board of Canada 15(3):275-293.
Ponomarenko, V.P. & I.Ya. Ponomarenko 1975.
Consumption of the Barents Sea capelin by cod and
haddock. – ICES Council Meeting 1975/F:10. 8 pp.
Ponomarenko, V.P., I.Ya. Ponomarenko & N.A. Yaragina
1978. Consumption of the Barents Sea capelin by
cod and haddock in 1974-1976. – ICES Council
Meeting 1978/G:23. 22 pp.
Ponomarenko, I.Ya. & N.A. Yaragina 1978. Year-to-year
and seasonal fluctuations in the feeding of the Barents
Sea cod on capelin in 1947-1976. – ICES Council
Meeting 1978/G:24 19 pp.
— 1984. Predation by cod on the Barents Sea capelin in
1975-1981. – Contribution to the joint soviet/Nor-
wegian symposium on the Barents Sea capelin, Bergen,
Norway, August 1984. 13 pp. Unpublished manuscript.
Pope, J. 1979. A modified cohort analysis in which constant
natural mortality is replaced by estimates of predation
levels. – ICES Council Meeting 1979/H:16. 8 pp.
Pozdnjakov, J.F. 1957. O plodovitosti moivy Barentseva
morya (On the fecundity of capelin in the Barents
Sea. –Doklady Akademia Nauk SSSR 112:777-778
(In Russian)
— 1960. Materiali o razvitii moivi Barentseva Morya.
(Materials on the development of capelin in the
Barents Sea). – Trudy Murmanskogo Morskogo
Biologischesko Instituta, 2(6):211-225. (In Russian)
Prokhorov, V.S. 1960. Post-spawning Survival of the
Barents Sea Capelin (Mallotus villosus villosus
Müller). – ICES Council Meeting 1969 165. 5 pp.
— 1963. Some features of the ecology of capelin
(Mallotus villosus (Müller)) in the Barents Sea. –
Trudy PINRO vol. 15:163-176. (In Russian)
— 1965. Ecology of the Barents Sea Capelin (Mallotus
villosus villosus) and prospects for its commercial
utilization. – Fishery Research Board of Canada
Translation Series No. 813. 131 pp.
— 1968. Materials on the ecology of capelin in the Barents
Sea. – Pp. 23-31 in: Blacker, R.W. (ed.). Symposium
on the ecology of pelagic fish species in Arctic waters
and adjacent seas. Conseil Permanent International pour
l’Exploration de la Mer. Rapports et Procès-Verbaux
des Réunions 158.
Rass, T.S. 1933. Nerest moivy (Mallotus villosus Müller)
Barentseva Morja. (The spawning of capelin in the Barents
Sea). – Trudy Gasudarstvennogo Okeanografitsjeskogo
Instituta, Moskva IV(1):3-35. (In Russian, with English
summary)
Reed, M. & J.G. Balchen 1982. A multidimensional con-
tinuum model of the fish dynamics and behaviour:
Application to the Barents Sea capelin (Mallotus
villosus). – Modelling, Identification and Control
3(2):65-109.
Sakshaug, E., A. Bjørge, B. Gulliksen, H. Loeng & F. Mehlum
(eds) 1992. Økosystem Barentshavet. (The Barents
Sea ecosystem). Pro Mare. Norwegian Scientific
Council, Norwegian Ministry of Environment. (In
Norwegian). 304 pp.
Salvanes, A.-G.V. 1984. Fordelingsmønster og dødelighet
av loddelarver, samt berekning av gytebestand, basert
på larveundersøkelsene langs Troms- og
Finnmarkskysten 1972-1976. (Distribution pattern
and mortality of capelin larvae, and estimation of
spawning stock, based on the larval investigation
along the coast of Troms and Finnmark 1972-1976).
– Thesis in Fisheries Biology, Department of Fisheries
Biology, University of Bergen. 111 pp.
Sætre, R. & J. Gjøsæter 1975. Ecological Investigations on
the Spawning Grounds of the Barents Sea Capelin. –
Fiskeridirektoratets Skrifter Serie Havundersøkelser
16:203-227.
Sangolt, G. 1989. Intern toktrapport fra tokt med F/F
“Michael Sars” 6. mars til 22. mars 1989. (Unpublished
cruise report R/V “Michael Sars” 6 March to 22 March
1989). Directorate of Fisheries, Bergen. 3 pp.
Sangolt, G. 1992. Intern toktrapport fra tokt med F/F “Michael
Sars” 9. mars til 22. mars 1992. (Unpublished cruise
report R/V “Michael Sars” 9 March to 22 March
1992). Directorate of Fisheries, Bergen. 3 pp.

Gjøsæter – The population biology of capelin in the Barents Sea 495
Sars, G.O. 1879. Report submitted to the Department of
the Interior on the practical and scientific
investigations of the Finnmark capelan-fisheries,
made during the spring of the year 1879. – Report of
US Commercial Fish and Fisheries 8:167-187.
Serebrov, L.I. 1985. On density of distribution and
orientation of capelin in schools. – Pp. 135-147 in:
Gjøsæter, H. (ed.). The proceedings of the Soviet-
Norwegian symposium on the Barents Sea capelin.
Institute of Marine Research, Bergen, Norway, 1985.
Serebryakov, V.P., V.K. Aldonov, A.V. Astafeva & N.V.
Mukhina 1984. The distribution of capelin larvae in
the Norwegian and Barents Seas in April-May 1968-
1983. Paper presented to the Soviet-Norwegian
Symposium on the Barents Sea Capelin, Bergen,
Norway, 14-17 August 1984. 50 pp. Unpublished
manuscript. (A shortened version without most of
the charts published as pp 185-192 in: Gjøsæter, H.
(ed.). The proceedings of the Soviet-Norwegian
symposium on the Barents Sea capelin. Institute of
Marine Research, Bergen, Norway, 1985.)
Serebryakov, V.P., N.G. Ushakov, V.K. Aldonov & A.V.
Astafeva 1985. Population fecundity and abundance
of year-classes of the Barents Sea capelin 1972-1984.
– Pp 25-30 in: Gjøsæter, H. (ed.). The proceedings
of the Soviet-Norwegian symposium on the Barents
Sea capelin. Institute of Marine Research, Bergen,
Norway, 1985.
Shevchenko, A.V. & A.S. Galkin 1983.O prognozirovanii
srokov podkhoda mojvy Barentseva morya na nerest
k beregam Murmana. (On the prediction of spawning
migrations of the Barents Sea capelin to the
Murmansk Coast). Biologiya i promysel
pelagicheskikh ryb severnogo bassejna. (Biology of
and fishery for pelagic fishes in the northern seas). –
Sbornik Nautsnikh Trudov pp. 31-36. PINRO
Murmansk. (In Russian)
Shulga, A.D. & V.V. Belusov 1976. Dependence of certain
biological characteristics of the Barents Sea capelin
on the temperature regime. – ICES Council Meeting
1976/H:5. 12 pp.
Skjoldal, H.R., H. Gjøsæter & H. Loeng 1992. The Barents
Sea capelin in the 1980s: ocean climate, plankton,
and capelin research. – ICES Marine Science
Symposia 195:278-290.
Skjoldal, H.R. & F. Rey 1989. Pelagic production and vari-
ability of the Barents Sea Ecosystem. – Pp. 241-286
in: Sherman, K. & L.M. Alexander (eds). Biomass
Yields and Geography of Large Marine Ecosystems.
American Association for the Advancement of
Science. Selected Symposium 111.
Solemdal, P. & P. Bratland 1986. Klekkeforløp for lodde i
Varangerfjorden 1986. (Hatching progress of capelin
in the Varanger fjord 1986). Unpublished cruise
report, Institute of Marine Research, Bergen. 9 pp.
Stergiou, K.I. 1984. Capelin (Mallotus villosus) and
climatic change in the Barents Sea. – Master of
Sciences thesis, Institute of Oceanography, McGill
University, Montreal, Canada. 243 pp.
Strøm, A. & T. Monstad 1969. Rapport om loddeunder-
søkelsene vinteren 1969. (Report on the capelin
investigations in winter 1969. – Fiskets Gang nr.
55:334-336. (In Norwegian)
Strøm, A. & G. Vestnes 1967. Rapport om loddetokt med
F/F “G.O. Sars” og M/S “Steinhauk” til Finnmark og
Barentshavet januar- mars 1967. (Report on capelin
investigations with R/V “G.O. Sars” and M/S
“Steinhauk” January to March 1967. – Fiskets Gang
nr. 53:344-346. (In Norwegian)
Strøm, A., G. Vestnes & S. Olsen 1968. Rapport om lodde-
undersøkelsene vinteren 1968. (Report on the capelin
investigations in winter 1968.) – Fiskets Gang
54:371-372. (In Norwegian)
Tangen, Ø. 1995. Intern toktrapport F/F “G.O. Sars” 31.
mai-27 juni 1995. (Unpublished cruise report R/V
“G.O. Sars” 31. May - 27 June 1995.) Institute of
Marine Research, Bergen. (In Norwegian)
Tangen, Ø. & K. Bakkeplass 1996. Intern toktrapport F/F
“M. Sars” 7. juni-7 juli 1996. (Unpublished cruise
report R/V “M. Sars” 7. June - 7 July 1996.) Institute
of Marine Research, Bergen. (In Norwegian)
Tereshchenko, E.S. 1996. Results from studies on capelin
otolith structure. – ICES Council Meeting 1996/
H:20. 14pp.
Timofeev, S.F. 1988. Vliyanie mojvy Mallotus villosus
villosus na produktivnost planktona Barentseva
morya. (Effect of capelin Mallotus villosus villosus
on plankton productivity in the Barents Sea.). –
Voprosy Ichtiologii 28(5) 858-860. (In Russian)
Tjelmeland, S. 1985. The capelin assessment model - a
documentation. – Pp. 31-43 in: Gjøsæter, H. (ed.).
The proceedings of the Soviet-Norwegian symposium
on the Barents Sea capelin. Institute of Marine
Research, Bergen, Norway, 1985.
— 1987a. The effect of ambient temperature on the spawn-
ing migration of capelin. – Pp. 225-236 in: Loeng, H.
(ed.). The effect of oceanographic conditions on distri-
bution and populaiton dynamics of commercial fish
stocks in the Barents Sea. Proceedings of the third
Soviet-Norwegian symposium, Murmansk, 1986. In-
stitute of Marine Research, Bergen.
— 1987b. Mortality on Barents Sea capelin calculated
by the IMR multispecies model for the Barents Sea.
I: Mortality on mature capelin induces by cod. –
ICES Council Meeting 1987/H:55. 22 pp.
— 1992. A stochastic model for the Barents Sea capelin
stock with predation from an exogenous cod stock.
– Pp. 139-160 in: Bogstad, B. & S. Tjelmeland (eds).
Interrelations between fish populations in the Barents
Sea. Proceedings of the fifth PINRO-IMR
Symposium, Murmansk, 12-16 August 1991. Institute
of Marine Research, Bergen, Norway.
— 1996. Can the Management of the Barents Sea
Capelin be Improved by Multispecies Modelling? –
Proceedings. Forage Fishes in Marine Ecosystems.
Alaska Sea Grant College Program. AK-SG-97-01,
1997: 617-632.

496 Sarsia 83:453-496 – 1998
Tjelmeland, S. & B. Bogstad 1993. The Barents Sea Capelin
Stock Collapse: A Lesson to Learn. – Pp. 127-139
in: Smith, S.J., J.J. Hunt & D. Rivard (eds). Risk
Evaluation and Biological Reference Points for
Fisheries Management. Canadian Special
Publications on Fisheries and Aquatic Sciences. 120.
— 1998. Biological modelling. – Pp. 69-81 in: Rødseth,
T. (ed.). Models for multispecies management.
Physica-Verlag, Heidelberg.
Tjelmeland, S. & K. Forberg 1984. A method for assessing
early and late spawning capelin. – Contribution to
the joint Soviet/Norwegian symposium on the
Barents Sea capelin, Bergen, Norway, August 1984.
22 pp. Unpublished manuscript.
Toresen, R. (ed.) 1997. Ressursoversikt 1997. (Overview
of the resources), – Fisken og Havet, Særnummer 1
1997. (In Norwegian, with English summary)
Ulltang, Ø. 1975. The present capelin situation in the Barents
Sea. – ICNAF Research documents 75/40. 2 pp.
— 1995. Multispecies models and management with
reference to the Institute of Marine Research’s
multispecies model for the Barents Sea. – Pp. 659-
670 in: Blix, A.S., L. Walløe & Ø. Ulltang (eds).
Whales, seals, fish and man. Proceedings of the
International Symposium on the Biology of Marine
Mammals in the Northeast Atlantic, Tromsø, 29
November - 1 December 1994. Developments in
Marine Biology 4. Elsevier, Amsterdam.
Ushakov, N.G. V.A. Korzhev & V.L. Tretyak 1992. North-
east Artcic cod importance in capelin stock dynamics.
– Pp. 99-109 in: Bogstad, B. & S. Tjelmeland (eds).
Interrelations between fish populations in the Barents
Sea. Proceedings of the fifth PINRO-IMR
Symposium, Murmansk, 12-16 August 1991. Institute
of Marine Research, Bergen, Norway.
Ushakov, N.G. & V.K. Ozhigin 1987. The abundance of
year classes of the Barents Sea Capelin and
peculiarities of the distribution of the young in
relation to hydrographic conditions. – Pp. 159-167
in: Loeng, H. (ed.). The effect of oceanographic
conditions on distribtion and population dynamics
of commercial fish stocks in the Barents Sea.
Proceedings of the third Soviet-Norwegian
Synposium, Murmansk, 26-28 May 1986. Institute
of Marine Research, Bergen 1987.
Vesin, J.-P., W.C. Legett & K.W. Able 1981. Feeding
Ecology of Capelin (Mallotus villosus ) in the Estuary
and Western Gulf of St. Lawrence and its
Multispecies Implications. – Canadian Journal of
Fisheries and Aquatic Sciences 38:257-267.
Vilhjálmsson, H. 1994. The Icelandic Capelin Stock. Capelin,
Mallotus villosus (Müller) in the Iceland - Greenland -
Jan Mayen area. – Rit Fiskideildar. Journal of the
Marine Research Institute, Reykjavik vol. XIII no. 1.
281 pp.
Wiig, Ø. 1988. Grønlandssel og selinvasjon: Hva vet vi og
hva tror vi? (Harp seals and seal invasions: knowledge
and beliefs) – NATUREN 1988(2):35-41. (In Norwegian)
Øiestad, V. & E. Moksness 1979. Interactions between
Atlanto-scandian herring larvae (Clupea harengus L.)
and capelin larvae (Mallotus villosus) in a concrete
enclosure. – ICES Council Meeting 1979/F:52. 6 pp.
Zenkevitch, L. 1963. Biology of the Seas of the USSR. –
George Allen and Unwin Ltd. London. 955 pp.
Zilstra, J.J. 1971. Herring larvae in the central North Sea. –
Berichte der Deutschen Wissenschaftlichen Kommision
für Meeresforschung 21:92-115.
Accepted 24 April 1998
Editorial responsibility: Ulf Båmstedt