J.
Northw.
At!. Fish. SeL, Vol. 9:
37-43
Growth of Juvenile Pollock (Pollachius virens L.)
along the Atlantic Coast of Canada with Inferences of
Inshore-offshore Movements
D.
Clay
Marine
Fish Division,
Department
of Fisheries and Oceans,
Gulf
Fisheries Center,
P. O.
Box
5030,
Moncton,
New
Brunswick,
Canada E1C 9B6
and
W. T. Stobo, B. Beck, and P. C. F.
Hurley
Marine
Fish Division,
Department
of Fisheries and Oceans,
Bedford
Institute
of
Oceanography,
P. O.
Box
1006,
Dartmouth,
Nova Scotia, Canada B2Y 4A2
Abstract
Data collected
from
tagging
studies
along
the coast of Nova
Scotia
and offshore
ichthyo-
plankton
surveys were used to investigate
growth
and
inshore-offshore
migrations
of juvenile
pollock.
Age zero
pollock
from 0.3 to 4.2 cm in length were
found
in the
offshore
plankton
between
November
and
June
and small
pollock
at lengths of 7-11 cm were
caught
inshore in
July.
These
small
pollock
appear to have moved inshore after leaving the
offshore
pelagic
community
at
3-6
months
of age.
They
remain inshore
until
they reach
approximately
30+ cm
during
their
second
year. Tagged inshore
pollock
were observed to
grow
17-18 cm
during
their
first year of life, an
annual instantaneous
growth
rate (G) of4.1, and 27-30 cm by the end of
their
second year (G =1.8).
Seasonal variability in
growth
was also observed in
juvenile
pollock.
Mean absolute
growth
from
May
until
August
was 1.75 cm per month,
from
August
until
October
1.50 cm per month, and from
November
until April
about
0.5 cm per
month.
A
tagging
related
mortality
experiment
using
winter
tagged juveniles indicated a
11%
mortality
during
the first 3
months
after tagging.
Introduction
Pollock
(Pollachius
virens,
L.) have increased in
commercial
importance
during
the last decade,
how-
ever,
much
of the
biology
of this species remains
unclear.
Current
stock
assessments
(McGlade
et al.,
MS 1984) and
commercial
fisheries managers (Anon.
1976, 1984) have treated
pollock
in the
Northwest
Atlantic
as a
single
stock
encompassing
the Scotian
Shelf, Georges Bank, and the
Gulf
of Maine.
Until
recently, fisheries managers in the USA and Canada
accepted the
hypothesis
that
these fish move to Massa-
chusetts Bay and
Jeffrey's
Ledge in the
Gulf
of Maine
(Fig. 1) to spawn in late
autumn
and early winter. It was
thought
that
they then dispersed
throughout
the
Gulf
of
Maine and
Scotian
Shelf
during
summer. However, as
early as 1963, it had been speculated
that
spawning
also
occurred
on the Scotian
Shelf
(Steele, 1963).
More
recently the Scotian
Shelf
Ichthyoplankton
Program
(SSIP) of the Canadian
Department
of Fisheries and
Oceans, has
confirmed
the presence of
pollock
eggs
and larvae on the Scotian
Shelf
(O'Boyle
et al., 1984),
and
Scott
(1980)
found
larvae in the Bay of Fundy.
This
tagging
program
was established to investi-
gate seasonal movements and
distribution
of
pollock
on the Scotian
Shelf
and, where possible, to
study
growth
of tagged individuals. Initial attempts to ta.g
adult
pollock
from
offshore
areas using fish
collected
from
bottom
trawls
deployed
from
research vessels
proved unsuccessful.
Consequently,
large
numbers
of
O-group,
1-year
old, and
more
limited
numbers
of 2-
year-old
fish that were available
from
inshore
waters
were
marked.
In
addition,
to
investigate
possible
effects of tag
induced
mortality,
survival of
winter
tagged juveniles was evaluated by
holding
pollock
in
cages.
Although
tagging
was
conducted
in all seasons,
time
constraints
allowed
only
one
tagging
survival
study
and
winter
was chosen as a
possible
worst
case
scenario.
This
paper deals
with
the aspect of
individual
fish
growth
derived from the
inshore
tagging
program.
Modal lengths and length ranges of larval and
recently
metamorphosed fish
from
offshore
SSIP surveys are
presently
only
to establish
minimum
length
of O-group
fish at mid-summer.
http://journal.nafo.int
38
N. H
J.
Northw.
Atl. Fish. Sci., Vol.
9,1989
I
I
I
J
I
I
r
,
f
I
/
/
Maine
ATLANTIC
OCEAN
400
L.-
__
....L-
--l
~
____I
__'__
___L
__'____'
720 70° 68° 66° 64° 62° 60°
Fig. 1. Map
showing
the
Massachusett's
Bay and
Jeffrey's
Ledge
areas and
pollock
tagging
sites
along
the
Atlantic
coast
of
Nova
Scotia
and New
Brunswick
between
1978 and 1984. (See
Table
1
for
tagging
dates and
numbers
of fish
tagged.)
Materials and Methods
Between
1978 and 1984, a
large
scale
pollock
tag-
ging
study
was
conducted
(see Fig. 1
for
sites). All
tagging
was
done
using
a
dart
tag
with
nylon
T
anchor
tags
(made
by
Floy
Manufacturing,
Seattle,
Washing-
ton,
USA)
in the
dorsal
region,
posterior
to the
first
dorsal
fin.
During
1979,
tagged
fish
were
measured
for
fork
length
to the nearest 0.5 cm. In
other
years all fish or a
random
sub-sample
of fish
captured
were
measured
to
the
nearest cm.
Length
frequency
data
were
collected
from
all
tagged
fish and
measurements
from
a
limited
number
of
recaptures
were
made
during
tagging
oper-
ations
in 1979 and 1980. In
addition,
independent
of
tagging,
seasonal
length
frequencies
of
inshore
pol-
lock
were
collected.
Length
frequency
samples of
small
fish
(7-13
cm)
were
also
collected
from
four
areas by
scuba
divers
using
small seine nets between
mid-July
and
mid-August,
1982 (D.
Beanlands,
BID,
pers.
comm.).
Although
the
tagged
pollock
ranged
from
9 to 49
cm
(Table
1),
most
were
less
than
20 cm and
were
usually
not
present
until
mid-summer.
They
were
nor-
mally
found
in large
concentrations
near
fish
plant
out-
falls and
caught
by
angling,
dip
netting,
seining
or
trap
nets.
Two
major
sites of
tagging
for
juveniles
(ages 0
and
1)
were
Halifax
Harbour
and
Tiverton,
Nova
Scotia.
Absolute
growth
in
length
of
recaptured
fish was
calculated
as a
weighted
mean of
the
change
in
length,
divided
by
the
time-at-Iarge
in
months:
Absolute
growth
in
length
L
2
- L
1
where
L
1
and L
2
are
the
fork
lengths
(cm)
at
the
begin-
ning
and the end of
the
period
respectively, and t is the
time
period
(months)
at large.
Growth
in
length
is
des-
cribed
as the mean ± 1
standard
deviation.
Instantaneous
growth
rates (G)
for
either
monthly
or
annual
periods
have been
calculated
from
lengths
as
described
by
Ricker
(1975)
using
the
formula:
G=b(lnL
2-lnL
1)
where
L
1
and L
2
are the
lengths
at
the
beginning
and
end of the
period
respectively
and b is the
slope
of the
logarithmic
transformation
of
the
length
weight
regres-
sion.
The
'b' value was
obtained
from
the
length
weight
relationship
calculated
for
pollock
taken
from
the
same
areas and of a
similar
size range,
this
relationship
for
fish between 10 and 40 cm was:
W = 0.00885 X
FL3.11483,
n = 198, r = 0.99
where
W is the
weight
(g) and FL is
the
fork
length
(ern)
(Fig. 2).
Deriving
the
length
weight
relationship
from
these fish was necessary as the
tagged
fish
were
not
CLAY,
et
al.:
Growth
and
Inshore-Offshore
Movements of
Pollock
39
1000.-------y-------,------..--------,
OL.--
O:::::::""-_.....I-
--L
.L......-
-J
5 15 25 35 45
Fork
length
(cm)
Fig.2.
Length
weight
relationship
of
juvenile
pollock
(10 to 40 cm)
from
the east coast of
Canada
during
1979 and 1980.
from
the gear, preserved, and later identified,
counted
and the larvae measured
for
length to the nearest mm.
Length was measured as
notochord
length
until
the
posterior
edge of the
hypural
plate was observed to be
vertical, then standard length was used. No data were
available on the expected
shrinkage
after
the preserva-
tion, however, we feel
that
this
shrinkage
would
not
be
significant
for
the present use of the data.
The
length
data
from
these surveys were used solely to establish
that fish
caught
inshore
in
mid-summer
were of the
year-class
produced
in the
preceding
winter.
o
o
o
o
o
o
W = 0.00885 X
FL311463
n = 198 r
2
= 0.99
800
600
200
A
study
to assess possible large scale bias due to
differential
mortality
that
might
result
from
tagging
in
different
seasons was carried
out
using 150
pollock
captured in a research cod
trap
in
Halifax
Harbour
in
December 1979.
The
depth
at the
capture
sitewas8-10
m.
The
fish were
transported
in 85
liter
tanks
from
the
harbour
to a
holding
cage (30 m")
moored
in the
Bed-
ford
Basin at the
Bedford
Institute
of
Oceanography
(BIO).
They
were
acclimated
in the
holding
cage
for
16
days
prior
to tagging. Every 3 to 4 days
throughout
the
study
and
acclimation
period they were fed
chopped
cod and herring. In
January
1980,102
of the surviving
145 fish were tagged, and the rest left as
controls;
all
were measured to the nearest mm
fork
length
then and
in subsequent experiments.
The
cage was inspected
weekly
for
fish
mortality
and signs of
abnormal
swim-
ming
and feeding activity. In March when the cage was
replaced because of algal
fouling,
the fish were exam-
ined and measured. The fish were released at the
hold-
ing site at the end of May
after
counting
but
without
being measured.
1983
Deer
Island,
N.B.
30
May
800
12-23
Canso/Dover,
N.S.
27-31
May
5,131
20-29
1-5
Jun
20-47
Campobello,
N.B. 31
May
700
18-25
Campobello,
N.B. 1
Jun
1,497
17-26
Grand
Manan,
N.B. 3
Jun
1,927
17-25
----------------------.-.--
.----------------.------.
--------------
.. _------_ .. _-----------
1984
Grand
Manan,
N.B.
5-7
Jun
3,998
16-36
Total
1978-84
56,408
a
Tagging
along
the
South
Shore
was
conducted
at East
Pubnico,
Kelly
Cove,
Cape
Sable,
Lockeport,
and
Woods
Harbour.
weighed in
order
to keep
handling
to a
minimum.
Fish
sampled
from
the
ichthyoplankton
surveys had no
weight
data collected and thus the scaling
factor
(for
Ricker's 1975 equation) was set to 3.0.
TABLE
1.
Locations,
dates, and
associated
data
for
selected
pollock
tagging
trips
along
the
coasts
of
Nova
Scotia
(N.S.) and
New
Brunswick
(N.B.)
between
1978 and 1984.
Pollock
Length
range
Year
Location
Date
tagged
(cm)
1978
Canso/Dover,
N.S.
22-23
Jun
991
23-43
-
~
-- - --- -- --- . ---- -- -- -- ----
.---------------
1979
Canso/Dover,
N.S.
1-6
Jun
5,670
20-49
Liverpool,
N.S.
26
Jun
22
22-37
Sandy
Cove,
N.S.
8-14
Jun
39
35-42
Tiverton,
N.S.
9-13
Jul
2,313
22-31
Tiverton,
N.S.
21-23
Aug
657
22-36
Halifax,
N.S.
22-25
Aug
297
10-40
Tiverton,
N.S.
9
Oct
300
15-39
Gunning
Cove,
N.S. 29
Oct
151
12-24
Tiverton,
N.S.
30
Oct
48
15-39
Halifax,
N.S.
5-6
Nov
2,112
13-45
Lockeport,
N.S. 7
Dec
66
13-22
Gunning
Cove, N.S. 7
Dec
45
13-22
Tiverton,
N.S. 8
Dec
36
15-19
----------------------
---------------.-.
.-------------.-------.-._----------
1980 Hal ifax, N .S. 4 Jan
98
15-23
Tiverton,
N.S.
10 Jan
200
13-19
Halifax,
N.S.
4-6
Feb
97 15-21
Herring
Cove,
N.S.
Feb-Mar
202 14-21
Halifax,
N.S.
3
Apr
40
15-24
Halifax,
N.S.
13
May
1,632
15-29
Canso/Dover,
N.S.
14-16
May
3,000
20-37
Northwest
Cove, N.S.
7-9
Jul
53
20-24
Sambro,
N.S. 15
Jul
12
9-20
East
Ferry,
N .S. 17
Jul
91 9-11
Tiverton,
N.S.
17
Jul
600
9-45
Halifax,
N.S.
29-30
Jul
484
11-25
South
Shore,
N.S.
a
4~8
Aug
186
12-35
Tiverton,
N.S. 10 Sep 596
16-44
East Ferry, N.S. 10 Sep 901
12-18
Cape
Sable, N.S. 10 Nov
1,496
13-19
Cape
Sable, N.S. 28 Nov 1,000
14-20
Beaver
Harbour,
N.B.
18-19
Dec 1,996
12-17
----------._--------------------
-----_._------------------------------------------------------------
1981
Canso/Dover,
N.S.
21-26
May
3,013
20-35
Tiverton,
N.S. 22
Aug
300
23-26
Beaver
Harbour,
N.B.
7
Oct
800
13-22
Cape
Sable, N.S.
22-28
Oct
3,919
14-33
-----------------------------------.
-----------------
-._--------.-.-----
1982
Campobello,
N.B. 20-21
May
3,999
16-28
Canso/Dover,
N.S.
20-26
May
3,993
20-44
The SSIP surveys,
which
ran
concurrently
with
the
present work, were designed to provide
distribution
and
abundance
indices
(O'Boyle
et al., 1984).
They
were
not
designed to provide a larval fish age and
growth
data series,
consequently
the preservation
techniques
precluded the
study
of the larval
otoliths
for
daily
growth.
All fish
from
the samples were removed
Results and Discussion
Growth estimated from ichthyoplankton surveys
Fahay (1983) reported the size at
hatching
on
Georges Bank as 0.3 to 0.4 em in
length
with
the
young
40
J.
Northw.
Atl. Fish. Sci., Vol. 9, 1989
Nov Dec Jan Feb Mar
Apr
May
Fig.3.
Length
of
pollock
collected
from
three
plankton
gears
fished
over
the
Scotian
Shelf
during
1979 and 1980.
The
bars
repre-
sent
the range of
lengths,
the
numbers
over
the bars are the
sample
size, and the
gear
symbol
is
located
at the
median
length.
Although
we have no
direct
evidence at
this
time,
our
ichthyoplankton
data suggest
that
larval
pollock
concentrate
inshore as
they
grow
and
their
decline
in
abundance
in the SSIP samples
suggest
that
they
dis-
appear
from
the
offshore
plankton
by
July.
This
is
coin-
cident
with
the inshore appearance of small
pollock,
although
at a larger size than
that
reported
from
Europe
(Steele, 1963). The
question
whether
this inshore
movement is a passive
drift
or an active
migration
remains to be answered.
It is probable
that
larval
pollock
miqrate
inshore as
they
grow, entering bays and estuaries where small fish
are
commonly
found.
This
was suggested by Steele
(1963)
for
small
pollock
over 3-cm long in European
waters. Such a movement has also been noted by
Scott
(1980)
for
pollock
juveniles in the Bay of Fundy.
Steele's (1963) observation
that
they
remain
inshore
for
their
first 2 years of life is
supported
by the fact that no
small
pollock
have been
caught
over the Scotian Shelf
during
any of the juvenile surveys
that
have been
con-
ducted
annually
since 1978 using small mesh
bottom
trawl.
The fish were observed to remain in the
vicinity
of
the
outfalls
at
least
until
the
following
January.
Although
no temperature preference studies were
con-
ducted, the
pollock
were observed to leave the area
when the water
temperature
dropped
below
4° C,
dur-
ing the
January
to
April
period, when these
pollock
were 16-24 cm in length (Fig. 4). On calm,
sunny
days
they
often reappeared inshore, and resumed active
feeding. From
this
we
conclude
that
they
departed to
adjacent deep water, where
temperatures
were
more
stable. Sequential observations near the
effluent
out-
falls showed
they
were present at the
beginning
of May.
Around
mid-July
a new
cohort
(7-11 cm) arrived and
from
July
to November both
groups
utilized the area.
By November,
many
of the second year fish
which
were
in excess of 30 cm in length left the area,
presumably
moving
into
deeper
offshore
water.
The
new
cohort
Growth - inshore phase
Many
bays along the coast of Nova
Scotia
have at
least one fish processing plant, and all of these
dis-
charge
some
form
of fish
protein
back
into
the ocean.
We observed
that
during
the summer,
young
pollock,
frequently
in the
order
of several
thousands
of fish,
form
dense feeding
concentrations
near these
dis-
charge
pipes. Field observations of
gut
contents
of
some of these fish indicate flesh and skin
fragments
from larger fish
predominate
their
diets. No
quantita-
tive data were collected.
Size
dependent
gear avoidance by larval and post
larval fish is well
documented
(Clutter
and Anraku,
1968), therefore the
maximum
lengths
taken by these
gears
would
underestimate the
upper
limits
of
growth.
This
underestimate may also be
influenced
by the
change
in behaviour
from
a pelagic to demersal exist-
ence (Fahay, 1983). To provide a
first
estimate of G
for
the larval and post larval
planktonic
stages of
pollock,
the change in the median
lengths
of fish
caught
in the
neuston net are assumed to be representative of
their
growth;
in this case G is
approximately
0.4.
Due to
high
variability
in sample
numbers
and
length ranges of fish
caught
in the three gears, data
from
all gear types are presented to
provide
a
compo-
site view (Fig. 3). The data
indicate
larvae appear in the
offshore
water
column
in
November
at
lengths
of 0.3 to
1.1 cm and by May, the larvae are observed at
lengths
of
0.4 to 4.2 cm.
The
continued
presence of small
pollock
in the samples
throughout
the
Novemberto
May
period
indicates a
protracted
spawning
period; the increasing
trend in
maximum
and median
lengths
provide
a
min-
imum
estimate of
growth
of an earlier stage of the
year-class.
While the preservation
method
of samples pre-
cluded
the use of
otoliths
to
identify
young
of the year
pollock, the
authors
are also
not
aware of any
publica-
tion
on age validation
work
on
pollock
from
the east
coast of
North
America.
76
-
....
.
1
12
T87 4
11 x T
T
1
130
37
I
6 -
1 1
125
1 Il
J.
17
128
TI
11
x Bongo
• Isaac Kid
- Neuston
.5
4.0
1.5
1.0
3.5
4.5
_ 3.0
E
~
.!:
2.5
rn
c
Q)
.-J 2.0
settling to the
bottom
by the
time
they
had reached 5
cm. The SSIP survey data (Fig. 3)
confirm
that
pollock
spawn over the Scotian Shelf, as previously suggested
by Steele (1963). The data suggest
that
pollock
appear
to spawn on the Scotian Shelf
(oftshorej
between
November
and March
with
peak
spawning
occurring
between
December
and
February
as
reported
by
O'Boyle
at al., 1984. A
similar
temporal
pattern has
been observed on Georges Bank
(Colton
and
Byron,
1977;
Colton
at al., 1979). For
this
reason we have
arbitrarily
assigned 1
January
as the
birthdate
of
Sco-
tian Shelf
pollock
in
this
growth
study.
CLAY, et al.:
Growth
and
Inshore-Offshore
Movements of
Pollock
l"-
N
~-
50
L{)
0
co
1"-0'>
o~
N
I
C"')
40
~I
N
N
co
~
<D
C"')I"-
(;
0'>
OCO
E
N
~~
~
£ 30
I
II
0>
I
c
~
~
T'"""
~
co
0
11
~
~
0
en
0'>
I
LL
Ii
<D
Ii
0'>
I
I
N
~N
II
20
COT'"""
I
ON
C"')
~~
I
T'"""N
~I
C\i,....-
C"')
C"')
I"-
0
0
L{)
N
N
T'"""
0'> 0
I"-
co
<D L{)
10
!~
L{)
I~
N
~o
~
T'"""
~
C"')
I"-
~
T'"""
II
N
T'"""
T'"""
41
o
Jan Feb Mar Apr May Jun Jul Aug Sep
Oct
Nov Dec
Fig.4.
Median and range of
lengths
of
pollock
sampled at various
tagging
sites over several years (see
Table
1). All years
combined
to
produce
an
'annual
view'. The shaded area represents
ichthyoplankton
data (see Fig. 2), the
remainder
of the data
come
from
this
tagging
study. The bars represent the range of lengths, the
numbers
at one of the bars are the sample size, and the
dot
is the median length.
continues
to
grow
and replace the second
yearfish
and
carryon
in
the
localized
apparently
temperature
related movement.
The size range of the
inshore
pollock
studied
was
7-40+ cm (Fig. 4).
During
the
July
to
November
period,
there were
two
discrete length
groups
in this
popula-
tion
component
indicating
two
distinct
cohorts. A
third
cohort
of larger fish may be present
for
a
short
period in
spring,
mid-May
to mid-June.
The
smaller-sized
cohort
did
not
increase
greatly
in median length between
October
and April. The average length of the
cohort
of
smaller fish
during
this
winter
period, at the end of
their
first year, was 17-18 ern:
for
the
cohort
of larger fish, at
the end of
their
second year, the average length was
27-30
cm.
During
1978,97 previously tagged fish were recap-
tured at
their
release site. These were measured
for
length and released a second time. These data also
indicate
two
distinct
cohorts
inhabiting
the inshore
area (Fig. 5).
Pollock
of the larger and smaller
cohorts
from Halifax,
exhibited
a weighted mean
growth
of
1.38±0.48 (n=8) and 1.42±0.55 (n=10) cm per
month
respectively
from
August
to early November. The
values of G
forthis
time
period were approximatelyO.05
and 1.10 respectively. Between November and May the
cohort
of smaller fish, with an initial median size of
17-18 cm,
grew
about
0.52±0.24 (n=36) cm per
month
(G=0.04), then showed increased
growth
that
about
2.23±0.59 (n=6) cm per
month
(G = 0.15) between May
and July.
Growth
of a single
cohort
in
adjacent
years
(Fig. 5), suggests that juveniles
grow
approximately
1.75 cm per month
during
May to
November
and
about
0.5 cm per
month
during
the rest of the year. The
November to May length data
suggest
that
little
growth
occurs
during
the
winter
months.
The
cage data indicates an even
lower
growth
rate
of 0.12 cm per
month
in the
January
to
March
period.
This difference, however, is possibly
dueto
a
combina-
tion
of the experimental
conditions
and the
timing
of
the cage
experiment
in
mid-winter.
Pollock
from
Tiverton
(Fig. 5)
exhibited
a
weighted
mean
growth
of 1.75 (n=16) cm per
month
(G = 0.19)
between
July
and
October,
and a
lower
weighted
mean
growth
of 0.59 (n=9) cm per
month
(G = 0.04) between
October
and December.
The
generally
similar
growth
patterns at
Halifax
and
Tiverton
suggest a
summer
rate
(May
through
August) of
approximately
1.75 cm per
month, a subsequent
decline
during
autumn
(August
to
October/November)
to
about
1.5 cm per month, and a
winter
rate (November
through
April)
of
approximately
0.5 cm 'per month. With the
strong
seasonal
tempera-
ture
cycle
found
in
Atlantic
Canada and its
likely
effect
on
pollock
growth,
we
would
expect
that
the actual
pattern of annual
growth
should
be
curvilinear
rather
than linear as suggested in Fig. 5.
Although
the
nature
of the data makes it
difficult
to
consider
growth
other