ICES Journal of Marine Science, 58: 50–60. 2001
doi:10.1006/jmsc.2000.0989, available online at http://www.idealibrary.com on
Day–night and depth effects on catch rates during trawl surveys
in the North Sea
G. Petrakis, D. N. MacLennan, and A. W. Newton
Petrakis, G., MacLennan, D. N., and Newton, A. W. 2001. Day–night and depth
effects on catch rates during trawl surveys in the North Sea. – ICES Journal of Marine
Science, 58: 50–60.
Data from the Scottish participation in the International Young Fish Survey for the
period 1976–1993 were analysed to examine the effect of light level and depth on catch
rates. The species selected for this study were common dab, herring, haddock and
whiting. Differences between day and night were observed for the juvenile common
dab and herring and for the adult common dab and haddock. Differences between
shallow and deep water were observed for the juveniles of all the species and for the
adults of common dab, haddock and whiting. The mean lengths and the composition
of the common dab and whiting catches were affected by both light and depth,
indicating behaviour differences between the juveniles and adults of these species.
Whilst light seems to have no effect on the mean length and the catch composition of
herring, differences were observed between shallow and deep waters. In the case of
haddock, neither the light level nor the depth had any noticeable effect on the mean
length and catch composition of the catches. We conclude that the diel behaviour and
the geographic distribution are important factors in determining the quantity and
composition of trawl catches, but their effects are species dependent. When trawl
surveys are not exact replicates in terms of fishing times and areas, the estimation of
catch indices should allow for the possible bias introduced by these factors. There is a
need for models of the capture process that take account of such effects.
Key words: trawl survey, North Sea, demersal fish.
Received 26 June 1998; accepted 24 June 2000.
G. Petrakis: National Centre for Marine Research, Agios Kosmas Hellinikon, 16604,
Athens, Greece; D. N. MacLennan, and A. W. Newton: FRS Marine Laboratory,
PO Box 101, Victoria Road, Aberdeen, AB11 9DB, Scotland.
Introduction
Catch data from trawl surveys are used to derive
abundance indices and biomass estimates for commercially important fish stocks. Such data are routinely
collected in surveys conducted in the North Sea (see, for
example, Anon., 1992a) and in the Northwest Atlantic
(see, for example, Doubleday and Rivard, 1981). In the
North Sea, many research vessels from the surrounding
countries participate in a coordinated exercise known as
the International Young Fish Survey (IYFS). A brief
history of this work is given in Anon. (1990) and details
of the survey design can be found in Anon. (1986).
Standard indices of fish abundance are calculated for the
juveniles of herring (Clupea harengus), cod (Gadus
morhua), whiting (Merlangius merlangus), haddock
(Melanogrammus aeglifinus), Norway pout (Trisopterus
esmarki), sprat (Sprattus sprattus) and mackerel
(Scomber scombrus). The indices are calculated as follows. The surveyed region is divided into sub-areas 1 in
1054–3139/01/010050+11 $35.00/0
longitude by 0.5 in latitude known as ‘‘statistical
rectangles’’. The arithmetic mean of the catch per hour is
calculated for all hauls within each rectangle. The overall index for each species is then obtained as the average
for all rectangles within the standard area appropriate to
the species in question.
Many factors, such as the time of the day, influence
the capture efficiency of sampling gear and consequently
the derived abundance indices (Stewart and Galbraith,
1987; Stewart et al., 1991). However, no account is taken
of variation in the catch rates due to these factors, with
the exception of herring where the abundance index is
calculated excluding the night hauls. One of the assumptions of trawling surveys is that all fish in the survey area
are equally vulnerable to the sampling gear (Gunderson,
1993). The implication is that a constant proportion of
the target fish must always be near the bottom, below
the height of the trawl headrope, but this is not true
for species whose distribution changes with time due
to vertical diel migrations. In addition, the spatial
Day–night and depth effects on catch rates during trawl surveys in the North Sea
distribution of the fish in relation to the gear and the
behaviour of the fish in the vicinity of the gear are
sources of bias in biomass estimates (Parrish, 1963).
Previous studies have demonstrated that the daytime
catches for herring, cod, haddock, beaked redfish
(Sebastes spp) and American plaice (Hippoglossoides
platessoides) are higher than the night-time catches
(Woodhead, 1964; Beamish, 1966; Pitt et al., 1981;
Atkinson, 1989; Engas and Soldal, 1992; Marshall and
Frank, 1994). The daytime catches of sole and yellowtail
flounder (Limanda ferruginea) were reported lower than
those at night time (Woodhead, 1964; Walsh, 1988).
These differences have been attributed to changes in fish
behaviour which determine the interactions between
fish and gear. Additionally, differences in the size composition of the catch have also been observed between
daytime and night-time catches (Beamish, 1966; Walsh,
1988; Engas and Soldal, 1992), which suggest differences
in behaviour and catchability between young and adult
fish. For many demersal fish, a positive correlation
between size of fish and depth has been reported
(Macpherson and Duarte, 1991; Sinclair, 1992; Swain,
1993). Furthermore, the density of fish may vary within
the surveyed area since gradients can occur related to
oceanographic parameters like water depth, salinity,
temperature etc. (Daan et al., 1990; Gunderson, 1993).
In this paper we consider the data from the Scottish
IYFS in the period 1976–1993 covering four species:
herring, haddock, whiting and common dab (Limanda
limanda). These species were selected first, because they
are commercially important in the North Sea fisheries
and second, because they represent groups of species
with fundamentally different behaviour (herring:
pelagic; haddock and whiting: demersal; common dab:
benthic). The aim is to show how the catch rates and the
mean lengths are affected by time of fishing and whether
the water depth is an important determinant of the
distribution and the density of these species.
Materials and methods
Catch data from the Scottish research ships participating in the IYFS were examined, covering the 18 annual
surveys from 1976–1993. The surveys were always conducted during the first quarter of each year. Three
research vessels were involved as follows. From 1976–
1984 the side trawler ‘‘Explorer’’ was used, whereas
from 1985–1993 she was replaced by the stern trawler
‘‘Scotia’’. In addition, some hauls in 1978 and 1979 were
performed by the smaller vessel ‘‘Clupea’’. There were
912 valid hauls during the whole period. For 867 of the
hauls (95%) the duration of towing was recorded as 1 h;
for ten hauls (1%) the duration was longer and for
35 hauls (4%) it was shorter than 1 h. The ‘‘Grande
Overture Vertical’’ (GOV) bottom trawl was used in all
51
E6 E7 E8 E9 F0 F1 F2 F3 F4 F5 F6 F7
°N 52
51
50
2
7
8
4
11
49
1
2
8
9
48
1
6
9
11
47
1
15
12
13
46
6
15
14
14
10
11
3
45
16
15
15
16
10
9
3
44
18
17
11
12
9
8
9
9
9
13
15
7
6
9
10
10
43
North Sea
3
42
14
14
14
7
4
11
11
10
41
16
16
15
7
6
10
12
12
40
3
3
5
6
10
13
14
2
3
3
39
3
3
1
4
9
2
3
3
7
8
38
3
3
1
3
3
1
2
37
1
2
1
3
3
1
1
1
1
36
35
1
1
9
9
3
3
1
34
0°
°E
Figure 1. Sampling intensity in the North Sea showing the
number of years when each square was sampled by Scottish
research vessels participating in the IYFS.
surveys, rigged with a rubber disk or bobbin groundropes for fishing on clear and rough ground respectively
and a codend with mesh size of 20 mm (Anon., 1992b;
Anon., 1992c).
The Scottish vessels surveyed mainly in the northwest
North Sea, extending into the central North Sea in some
years (Figure 1). Stations fished between sunrise and
sunset were classified as daytime hauls (599) and those
between sunset and sunrise as night-time (225). Sunrise
and sunset were determined according to Anon. (1992c).
The hauls which started before sunrise or sunset (41 and
47 hauls, respectively) and finished after sunrise or
sunset, were classified as day or night depending on
when the main part of the haul occurred. The number of
day hauls was 637 (70%), the remaining 275 hauls (30%)
being night hauls. Two depth zones were chosen to give
a roughly equal division of hauls, namely shallow waters
(0–75 m) and deep waters (>75 m). Overall, 495 (54%)
and 417 hauls (46%) of the hauls were designated as
shallow and deep respectively. In 1976 only one haul was
executed at deep water; that year was ignored in the
statistical tests between shallow and deep water.
All catches were normalized to 1 h towing duration.
Where two or more hauls were executed in the same
statistical rectangle with the same timing and depth zone
52
G. Petrakis et al.
Table 1. Number of squares sampled each year according to the
time of day and the fishing depth (shallow <75 m; deep >75 m).
Results
Year
Day
Night
Shallow
Deep
Catch rates
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
15
25
33
26
35
18
27
37
41
33
32
36
27
32
28
35
35
31
4
14
15
17
15
6
5
13
11
19
15
18
16
19
15
24
22
19
18
23
34
22
21
16
17
25
26
28
24
29
20
25
17
23
25
20
1
16
13
15
24
8
15
25
27
22
22
23
23
25
27
36
32
29
(e.g. more than one daylight haul) the average number
of fish caught was estimated. Taking into account these
modifications, the number of rectangles used per year
and for each combination of factors is shown in Table 1.
The catch rate, as the average number of fish caught
per hour, was calculated separately for the juvenile and
the adult components. The distinction between juveniles
and adults was based on the length ranges used in the
IYFS for estimating the preliminary abundance indices.
Hence, for herring, haddock, and whiting, juveniles were
deemed to be fish of total length <20 cm. For common
dab there is no equivalent measure, so we decided to
classify as juveniles those fish with length <15 cm.
According to Leeuwen and Vethaak (1988), the backcalculated length at age 2 for common dab is 15.6 cm.
Thus we have designated the majority of common dabs
aged <2 as juveniles. Mean lengths were also calculated.
All estimators (mean lengths of total catch, catch rate
for juveniles and adults separately) were compared
according to time of fishing (day and night) and depth
strata.
Changes in the sampled area can lead to different
mean catch rates since the fish density varies from one
place to another. A year effect is expected because of
natural recruitment variation. Furthermore, a vessel
effect could be a source of bias since three different
vessels were used. In order to minimize these effects
the day:night and shallow:deep catch rate ratio was
calculated separately for each year.
Mann–Whitney’s non parametric U-test was performed to examine statistically the differences in the
catch rates. A one-way analysis of variance was used to
examine differences in mean length (Sokal and Rohlf,
1981).
The catch rates of common dab were generally higher
during the night tows [Figure 2(a) and (b)]. Exceptions
were observed in 1976 and 1992 for the juveniles and in
1976, 1983, 1984 and 1992 for the adults. The night
catches were on average 4.98 and 3.29 times the day
catches for the juveniles and the adults, respectively
(Table 2). The Mann–Whitney test showed significant
differences (5% level) between day and night for the
juveniles (p=0.002) but not for the adults (p=0.055,
Table 3). All the catch rates were higher in shallow
compared to deep water (Figure 2(c) and (d)]. On
average the shallow catch rates were 13.05 and 10.4
times the deep catch rates for the juveniles and the adults
respectively (Table 2). These differences are highly significant for both the juveniles and the adults (U-test
p<0.001, Table 3).
The catch rates of day hauls for both juvenile and
adult herring were higher than at night in all years
[except 1978 and 1990 for adults, Figure 3(a) and (b)].
The extremely high value of the day adult catches in
1988 was mainly due to great catches in two hauls off the
east coast of northern Scotland (80 600 and 84 600 fish
h 1, respectively). In the period 1976–1984, herring
catches were generally low. The day catch rates of
herring were on average 8.06 and 4.84 times the night
catch rates for the juveniles and the adults respectively
(Table 2). The Mann–Whitney test showed a significant
difference for the juvneiles but not for the adults
(p=0.000 and p=0.223, respectively, Table 3). In relation to depth the catch rates of the juveniles were higher
in shallow water in all years (except in 1977 and 1990)
whereas those of the adults were higher in deep water
(except 1976, 1979, 1980 and 1983) [Figure 3(c) and (d)].
The adult herring catches were very low both in shallow
and deep water until 1985. On average the shallow water
catch rates of juveniles was 7.91 times the deep catch
rate (Table 2). For this estimation, the years 1976, 1978
and 1993 were excluded because there were negligible
catches at the deep water stations. The opposite applies
to the catch rate of adults which in deep water was
7.32 times the catch rate in shallow water (Table 2).
The difference was found to be significant for the
juveniles but not for the adults (p=0.002 and p=0.178,
respectively, Table 3).
The day and night catch rates of haddock showed a
similar pattern to those of herring [Figure 4(a) and (b)].
The daytime catch rates were higher than at night both
for juveniles and for adults with exceptions in 1976, 1982
and 1985 for juveniles and in 1976, 1992 and 1993 for
the adults. The catch rates of juveniles showed large
fluctuations. On average the daily catch rates were 3.19
and 2.10 times the night catch rates for the juveniles and
Day–night and depth effects on catch rates during trawl surveys in the North Sea
1200
(a)
fishing
1000
800
Fish h
Fish h
–1
600
–1
fishing
1200
400
200
0
400
200
1000
fishing
800
–1
600
400
Fish h
fishing
–1
Fish h
600
0
(c)
200
(b)
800
76 78 80 82 84 86 88 90 92
Year
1000
0
1000
53
76 78 80 82 84 86 88 90 92
Year
(d)
800
600
400
200
0
76 78 80 82 84 86 88 90 92
Year
76 78 80 82 84 86 88 90 92
Year
Figure 2. Catch rates of common dab for the period 1976–1993. (a) Day (thin line) and night (thick line) catch rates of juveniles,
(b) day and night catch rates of adults, (c) shallow (thin line) and deep (thick line) water catch rates of juveniles and (d) shallow
and deep water catch rates of adults.
Table 2. Comparison of bottow trawl catch rates at different times of day and water depths. Data from
Scottish vessels participating in the IYFS during the period 1976–1993.
Species
Common dab
Herring
Haddock
Whiting
Maturity
Annual Ratio
Mean
s.d.
s.e.
J
J
A
A
J
J
A
A
J
J
A
A
J
J
A
A
N/D
SH/DE
N/D
SH/DE
D/N
sh/DE
D/N
DE/SH
D/N
DE/SH
D/N
DE/SH
D/N
SH/DE
D/N
DE/SH
4.98
13.05
3.29
10.40
8.06
7.91
4.84
7.32
3.19
8.35
2.10
4.21
2.75
4.11
1.25
6.67
5.86
14.59
4.63
12.01
6.54
8.81
4.38
6.90
3.10
11.11
1.56
3.73
2.46
2.87
0.93
10.67
1.38
3.54
1.09
2.91
1.59
2.27
1.03
1.67
0.73
2.69
0.37
0.91
0.58
0.70
0.22
2.59
The annual ratios are calculated from the average catch rates by condition in each year. J=juveniles,
A=adults, D=day, N=night, SH=shallow and DE=deep.
adults, respectively (Table 2). The difference was not
significant for the juveniles and significant for the adults
(p=0.090 and p=0.010, respectively). Haddock were
alwawys more abundant in deep water [Figure 4(c) and
(d)] except for the adults in 1979. On average the shallow
catch rates were 8.4 and 4.2 times the deep catch rates
for the juveniles and adults, respectively (Table 2).
Highly significant differences were found both for the
juveniles and adults (p=0.003 and p=0.000 respectively,
Table 3).
The daytime catch rates of juvenile whiting were
higher than at night in 14 out of the 18 years, whereas
negligible differences were observed for the adults
[Figure 5(a) and (b)]. The catch rates of the adults
showed an increasing trend after 1985. On average the
day catches of juveniles were 2.75 times the night catches
54
G. Petrakis et al.
Table 3. Results of the Mann–Whitney U-test comparing bottom trawl catch rates according to the
time (day or night, D–N) and water depth (shallow or deep, S–D).
Age group
Juveniles
Adults
Comparison
Dab
Herring
Haddock
Whiting
D–N
S–D
D–N
S–D
0.002
0.000
0.055
0.000
0.000
0.002
0.223
0.178
0.090
0.003
0.010
0.000
0.110
0.002
0.837
0.001
Numbers are p values, p<0.05 for 5% significance.
Fish h
1000
2000
76 78 80 82 84 86 88 90 92
Year
(d)
10 000
3000
2000
1000
0
4000
12 000
fishing
4000
6000
14 000
(c)
(b)
8000
0
76 78 80 82 84 86 88 90 92
Year
–1
fishing
–1
–1
2000
5000
Fish h
fishing
3000
0
10 000
(a)
Fish h
Fish h
–1
fishing
4000
76 78 80 82 84 86 88 90 92
Year
8000
6000
4000
2000
0
76 78 80 82 84 86 88 90 92
Year
Figure 3. Catch rates of herring for the period 1976–1993. (a) Day (thin line) and night (thick line) catch rates of juveniles, (b) day
and night catch rates of adults, (c) shallow (thin line) and deep (thick line) water catch rates of juveniles and (d) shallow and deep
water catch rates of adults.
whereas the catches of the adults were nearly the same.
The test showed no significant differences (p=0.110 and
p=0.837, respectively) between day and night catch rates
(Table 3). The catch rates for juvenile whiting were
higher in shallow water except for 1981 and 1985,
whereas those of the adult whiting were higher in deep
water except for 1977 and 1986 [Figure 5(c) and (d)]. The
shallow catch rates of juveniles were 4.11 times the
deep catch rates whereas the deep catch rates of adults
were 6.67 times the shallow catch rates (Table 3).
Highly significant differences were found both for
juveniles and adults (p=0.002 and p=0.001, respectively,
Table 3).
Mean length
For 13 of the 18 years, the mean length of the common
dab catch was larger by day than by night (Figure 6) and
also in deep compared to shallow water. The ANOVA
test showed significant differences in both cases
(p=0.023 and 0.014, respectively, Table 4).
In 15 of the 18 years, the estimated mean length
of herring caught at night was larger than that of
the day hauls (Figure 6) but not significantly (p=
0.328, Table 4). An increasing trend was observed
from 1982–1991 both for day and night herring mean
length. In all years (except for 1977) the mean lengths
of herring caught in deep water were significantly
larger than those from the shallow water stations
(p=0.043).
The mean lengths of haddock catches by day and by
night showed a similar fluctuation (Figure 6) but did not
show a clear pattern. No significant differences were
found (p=0.212, Table 4). No difference was found
between the mean length of haddock in shallow and
deep water (p=0.749).
Day–night and depth effects on catch rates during trawl surveys in the North Sea
1000
400
200
fishing
1500
1000
500
0
1000
500
2000
(c)
76 78 80 82 84 86 88 90 92
Year
(d)
1500
1000
500
0
76 78 80 82 84 86 88 90 92
Year
(b)
1500
0
76 78 80 82 84 86 88 90 92
Year
–1
fishing
–1
Fish h
600
2000
Fish h
–1
800
0
2000
(a)
fishing
1200
Fish h
Fish h
–1
fishing
1400
55
76 78 80 82 84 86 88 90 92
Year
Figure 4. Catch rates of haddock for the period 1976–1993. (a) Day (thin line) and night (thick line) catch rates of juveniles, (b)
day and night catch rates of adults, (c) shallow (thin line) and deep (thick line) water catch rates of juveniles and (d) shallow and
deep water catch rates of adults.
Fish h
1000
4000
3000
2000
1000
0
76 78 80 82 84 86 88 90 92
Year
(b)
2500
2000
1500
1000
500
4000
(c)
fishing
5000
3000
0
76 78 80 82 84 86 88 90 92
Year
–1
fishing
–1
–1
2000
6000
Fish h
fishing
3000
0
3500
(a)
Fish h
Fish h
–1
fishing
4000
76 78 80 82 84 86 88 90 92
Year
(d)
3000
2000
1000
0
76 78 80 82 84 86 88 90 92
Year
Figure 5. Catch rates of whiting for the period 1976–1993. (a) Day (thin line) and night (thick line) catch rates of juveniles, (b) day
and night catch rates of adults, (c) shallow (thin line) and deep (thick line) water catch rates of juveniles and (d) shallow and ddep
water catch rates of adults.
In 12 of the 18 years the mean lengths of the whiting
from the night hauls were larger than those of the
day hauls and the difference was significant (p=0.025,
Figure 6; Table 4). The mean lengths of whiting from
the shallow and the deep water showed a similar
pattern to the mean lengths of the herring. In all years
G. Petrakis et al.
18
17
16
15
22
20
18
16
14
12
30
Mean length (cm)
Herring
28
76 78 80 82 84 86 88 90 92
Year
Haddock
26
24
22
20
18
26
Whiting
24
22
20
18
16
76 78 80 82 84 86 88 90 92
Year
18
17
16
26
24
22
20
18
16
14
12
30
28
76 78 80 82 84 86 88 90 92
Year
Herring
76 78 80 82 84 86 88 90 92
Year
Haddock
26
24
22
20
18
16
76 78 80 82 84 86 88 90 92
Year
Common dab
19
15
76 78 80 82 84 86 88 90 92
Year
Mean length (cm)
Mean length (cm)
26
24
20
Mean length (cm)
14
Mean length (cm)
Common dab
26
Mean length (cm)
Mean length (cm)
19
Mean length (cm)
56
24
76 78 80 82 84 86 88 90 92
Year
Whiting
22
20
18
16
14
76 78 80 82 84 86 88 90 92
Year
Figure 6. Mean lengths of bottom trawl catches of four species from the Scottish IYFS data for the period 1976–1993. The graphs
compare the results by day (thin line) and night (thick line) (left column) and above (thin line) and below (thick line) 75 m depth
(right column).
(except 1984) this was larger at the deep than at the
shallow water stations and the difference was highly
significant (p<0.001).
Discussion
Our results show that the time of the day and the depth
are important determinants of the capture rate and the
size composition of trawl catches. These effects, how-
ever, are not the same for all the species. Thus, the catch
rates of juvenile common dab and herring and adult
haddock were affected by the time of fishing. With the
exception of adult herring, the catch rates of both age
groups of all the species were influenced by the depth.
The study of mean lengths showed time of day to be an
important determinant for common dab and whiting,
whilst depth had a notable effect on the mean lengths of
common dab, herring and whiting.
Day–night and depth effects on catch rates during trawl surveys in the North Sea
Table 4. Analysis of variance: Comparison of mean lengths and
juvenile:adult catch ratios (Log transformed) for each species
according to the time (day or night, D–N) and water depth
(shallow or deep, S–D).
Mean length
Log (J/A)
D–N
S–D
D–N
S–D
Dab
Herring
Haddock
Whiting
0.023
0.014
0.010
0.036
0.328
0.043
0.222
0.000
0.212
0.749
0.486
0.315
0.025
0.000
0.073
0.000
Numbers are p values, p<0.05 for 5% significance level.
In the context of trawl surveying to determine catch
indices relating to population density, the important
requirement is to produce unbiased time-series. This
may be achieved by ensuring that annual surveys are
conducted in identical circumstances with regard to, for
example, the time of day and area distribution of fishing,
and using the same fishing method, vessel and so forth.
Unfortunately, this ideal approach is impractical over a
long period whilst there are ongoing changes in fishing
methodology and also potentially substantial alteration
of stock distribution and/or behaviour affecting the
catchability.
Differences in trawl catch rates between day and
night, as evidenced in our study, may be attributed to
fish behaviour, notably vertical migration or gear
avoidance reactions. The behaviour of many fish
species is affected by the diel cycle (Helfman, 1993), at
least in waters penetrated by the light. Their activity
and position in relation to the bottom may change,
consequently affecting their availability to the bottom
trawl. Many species are known to remain near the
bottom by day, moving upwards at night (Beamish,
1966). The avoidance of the gear is related to the
distance at which the fish can see the oncoming trawl.
During daylight, when the visibility is better, the fish
can see the gear at a greater distance and a larger
proportion may escape by swimming over the headline,
swimming out of the trawl path, escaping under the
groundgear or through the meshes (Wardle, 1993;
Main and Sangster, 1981; Walsh, 1988, 1991, 1992;
Engas and Godo, 1989). Swimming speed and gear
avoidance ability increase with the body size (Wardle,
1977).
The variation of trawl catches between shallow and
deep water indicates differences in the horizontal
distribution of the species. The preferred environment of
a species is determined by many biotic and abiotic
parameters, such as prey abundance or predator avoidance, temperature, salinity, bottom type etc. (Fry, 1971;
Scott, 1982; Nakken and Raknes, 1987; Rose and
Leggett, 1989; Smith et al., 1991; Swain, 1993). The
geographical distribution of a species can depend on the
density of the population and so may change between
57
years (Sahrhage and Wagner, 1978; MacCall, 1990;
Swain and Wade, 1993); in some cases it is considered to
be age or size dependent (MacPherson and Duarte,
1991; Sinclair, 1992; Swain, 1993).
Differences in the mean lengths between day and night
hauls suggest that the diel behaviour and the ability to
avoid the gear are not the same for different ages/sizes.
Similarly, differences between the shallow and deep
water catches indicate some degree of horizontal
segregation between the different size classes.
Common dab occur mostly in shallow waters where
there is illumination. According to Ortega-Salas (1988),
they feed during the day mainly on benthos (Pihl, 1994).
The proportion of the population available to the
bottom trawl during the day cannot be lower than at
night, if, in the daytime, the fish are feeding on the
bottom. Therefore, the higher catch rate during the
night hauls might be attributed to avoidance of the gear
during the day hauls. Similar differences between day
and night catches were reported by Parrish et al. (1964)
for common dab and by Walsh (1988) for yellowtail
flounder. Walsh (1991) says that a large proportion
(95%) of juvenile yellowtail flounder and American
plaice manage to escape under the groundgear and
Parrish et al. (1964) suggest that juvenile common dab
can escape from the forward part of the trawl.
The different mean lengths of the common dab
catches between shallow and deeper waters indicates
that the bathymetrical distribution of the species is size
dependent. The bigger individuals prefer deeper waters.
The mean lengths during the day hauls are larger than
during the night hauls. This indicates that the small fish
are more able to avoid the gear during the day by
escaping through the meshes or under the groundgear.
Differences between the day and night catch rates of
herring are well known and only the daytime hauls are
taken into account to estimate the standard abundance
indices of this species in the North Sea. Herring is a
pelagic fish but it migrates vertically from the deeper
water by day to the upper water by night (Parrish et al.,
1964). Thus, during the day more fish are available to
the bottom trawl. Our results show this to be more
evident for the juveniles than the adults.
The mean lengths of the night catches of herring were
larger than those of the day catches, suggesting that a
greater proportion of the small fish undertakes vertical
migration. This finding is supported by Woodhead
(1964) who observed similar size differences in herring
catches according to the time of day. Differences in
feeding behaviour between juveniles and adults could be
a possible explanation. The main food of the juvenile
herring are zooplankton (mainly copepods) and the fish
follow the prey which move towards the surface at night.
The adult herring feed mainly on larger crustaceans
which remain close to the bottom (De Silva, 1973; Last,
1989; Arrhenius and Hansson, 1993).
58
G. Petrakis et al.
There are many contrasting references about the day
and night catch rates of haddock. Higher night catch
rates have been reported by Jones (1956) in the Faroe
islands and by Parrish et al. (1964) in the Moray Firth.
However, higher day catch rates were reported by
Parrish et al. (1964) in the Buchan deeps and the Orkney
and Shetland Islands, by Woodhead (1964) off the
Northern coast of Norway and on the Skolpen Bank, by
Shepherd and Forrester (1987) in the Northwest
Atlantic and by Engas and Soldal (1992) in the Barents
Sea. According to Beamish (1965), haddock were on the
bottom during the day and off the bottom during the
night in Nova Scotia and in the Gulf of St Lawrence.
Haddock generally swim in the direction of the tow
between the wings and bridles (Main and Sangster,
1981). When the fish tire they turn back and swim
upwards. Some fish rise far enough to escape over the
headline or through the netting panel in the top of the
gear. It is possible that during the day more fish can
avoid the gear because of better visual conditions.
Although more fish may therefore escape during the
day, according to our observations the day time catch
rates were higher. This leads to the conclusion that the
proportion of the haddock close to the bottom and
vulnerable to the trawl is higher during the day. This was
more evident in our results for the adult fish.
Less attention has been given to the diurnal behaviour
of whiting. Parrish et al. (1964) report that the catches of
whiting are higher during darkness and Woodhead
(1964) says that the catches just after sunrise are higher
than later in the day. Gordon (1977) reports that the
catch per hour by midwater trawl is greater during
darkness. From our results there are some indications
that the catch rates of the juvenile whiting are slightly
higher during the day whereas no differences have been
observed for the adult whiting.
There may have been a vessel effect in the catch rates
of adult whiting which were generally higher after 1985
when the RV ‘‘Scotia’’ was used for sampling. On the
other hand, there was no difference in the catch rates of
the juveniles around 1985. Generally, the juveniles
showed higher catch rates during the day when the
incoming year class was strong. When the population
density is high the competition for food increases. More
fish may then undertake vertical migration during the
night in search of prey that move upwards such as
Euphausiidae and Crangonidae (Hislop et al., 1991). The
adults feed mainly on fish living close to the bottom.
The bathymetrical separation of juvenile and adult
whiting was clear. The juveniles were more abundant in
the shallow areas and the adults in the deeper waters. An
important observation is that, in the period 1983–1988,
the mean length of whiting catches was larger during the
day whilst in the other years it was larger during the
night. During 1983–1988 the catch rates of juveniles in
daylight and darkness were similar and generally both
were low. This supports the view that the vertical
migration of young whiting depends on the population
density. When there are many young fish the food
availability is lower. A higher proportion then moves
upwards at night and so the mean length of the fish
remaining close to the bottom increases.
The behaviour of whiting in the bottom trawl
supports the hypothesis of density dependent vertical
migration of the juveniles. Whiting swim between the
wings, similar to the haddock but higher off the seabed
(Main and Sangster, 1981). The fish make repeated
attempts to escape from the top part of the net (Briggs
and Robertson, 1993). The small fish are more likely to
escape through meshes which can be seen. Thus during
daylight, the proportion of small fish that escape is
higher, the retained fish are larger and consequently the
mean length of fish caught during the day is expected to
be larger. However, in years when the abundance of
juveniles is high, their vertical migration is strong and
the mean length of the catches is larger during the
night.
Our analysis has shown that catch indices from
bottom trawl surveys depend on many factors in
addition to the target population density. These effects
are not the same for all species and age groups. Very
likely they are particular to the region concerned, the
North Sea in this case. We conclude that catch indices
based on simple averages of the catch rates without
regard to extraneous causes of variability are liable to be
biased. The longer the time-series, the more likely it is
that such problems will occur. Clearly there is a need for
more sophisticated models of the capture process to
remove unwanted variability from the catch indices
revealed by trawl surveys.
Acknowledgements
This work has been supported by the Human Capital
and Mobility Programme of the European Union under
contract No. FRV 4001 GT92 0604.
2001 Crown copyright
References
Anon. 1986. Manual for the International Young Fish Surveys
in the North Sea, Skagerrak and Kattegat, Revision III.
ICES CM 1986/H: 2.
Anon. 1990. Report of the International North Sea, Skagerrak
and Kattegat bottom trawl survey working group. ICES CM
1990/H: 3.
Anon. 1992a. Report of the workshop on the analysis of trawl
survey data. ICES CM 1992/D: 6.
Anon. 1992b. Evaluation of sources of variability in the fishing
power of the GOV trawl. ICES CM 1992/B: 39.
Anon. 1992c. Manual for the International Bottom Trawl
Surveys, Revision. IV. ICES CM 1992/H: 3.
Day–night and depth effects on catch rates during trawl surveys in the North Sea
Arrhenius, F., and Hansson, S. 1993. Food consumption of
larval, young and adult herring and sprat in the Baltic Sea.
Marine Ecology Progress Series, 96: 125–137.
Atkinson, D. B. 1989. Diel movements of beaked redfish and
the implications of these for stratified random bottom trawl
estimates of biomass and abundance. North American
Journal of Fisheries Management, 9: 163–170.
Beamish, F. W. H. 1966. Vertical migration by demersal fish in
the Northwest Atlantic. Journal of Fisheries Research Board
of Canada, 23: 109–139.
Bourdeau, P. R. 1990. Acoustic observations of patterns of
aggregation in haddock (Melanogrammus aeglifinus) and
their significance to production and catch. Canadian Journal
of Fisheries and Aquatic Sciences, 49: 23–31.
Briggs, R. P., and Robertson, J. H. B. 1993. Square mesh panel
studies in the Irish Sea nephrops fishery. ICES CM 1993/B:
20.
Cook, R. M., and Armstrong, D. W. 1986. Stock-related effects
in the recruitment of North Sea haddock and whiting.
Journal du Conseil International pour l’Exploration de la
Mer, 42: 272–280.
Daan, N. (ed.) 1989. Data base report of the stomach sampling
project 1981. ICES Cooperative Research Report No. 164.
Daan, N., Bromley, P. J., Hislop, J. R. G., and Nielsen, N. A.
1990. Ecology of North Sea fish. Netherlands Journal of Sea
Research, 26: 343–386.
De Silva, S. S. 1973. Food and feeding of the herring Clupea
harengus and the sprat S. sprattus in Inshore waters of the
West coast of Scotland. Marine Biology, 20: 282–290.
Doubleday, W. G., and Rivard, D. (eds) 1981. Bottom trawl
surveys. Canadian Special Publication on Fisheries and
Aquatic Sciences, 58: 273 pp.
Engas, A., and Godo, O. R. 1989. Escape of fish under the
fishing line of a Norwegian sampling trawl and its influence
on survey results. Journal du Conseil International pour
l’Exploration de la Mer, 45: 269–276.
Engas, A., and Soldal, A. V. 1992. Diurnal variations in bottom
trawl catch rates of cod and haddock and their influence on
abundance indices. ICES Journal of Marine Science, 49:
89–95.
Fry, F. E. J. 1971. The effect of environmental factors on the
physiology of fish. In Fish Physiology. Vol. VI, pp. 1–98. Ed.
by W. S. Hoar, and D. J. Randall. Academic Press, New
York.
Gordon, J. D. M. 1977. The fish populations in inshore waters
off the West Coast of Scotland: The distribution, abundance
and growth of the whiting (Merlangius merlangius L.).
Journal of Fish Biology, 10: 587–596.
Gunderson, D. R. 1993. Surveys of fisheries resources.
John Wiley & Sons, New York. 248 pp.
Helfman, G. S. 1993. Fish behaviour by day, night and twilight.
In Behaviour of teleost fishes, 2nd edn, pp. 479–512. Ed. by
T. J. Pitcher. Chapman and Hall, London.
Hislop, J. R. G. 1984. A comparison of the reproductive
strategies of cod, haddock, whiting and Norway pout in the
North Sea. In Fish Reproduction: strategies and tactics,
pp. 311–329. Ed. by G. W. Potts, and R. J. Wootton.
Academic Press, London.
Hislop, J. R. G., Robb, A. P., Bell, M. A., and Armstrong,
D. W. 1991. The diet and food consumption of whiting
(Merlangius merlangus) in the North Sea. Journal of Marine
Science, 48: 139–156.
Jones, R. 1956. A discussion of some limitations of the trawl as
a sampling instrument. Rapport et Process-Verbaux des
Reunions, Conseil Permanent Internationale pour
l’Exploration de la Mer, 140: 44–47.
Last, J. M. 1989. The food of herring, Clupea harengus, in the
North Sea, 1983–1986. Journal of Fish Biology, 34: 489–501.
59
Leeuwen, P. I., and Vethaak, D. 1988. Growth of European
flounder (Platichthys flesus) and common dab (Limanda
limanda) in Dutch coastal waters with reference to healthy
and diseased fish. ICES CM 1988/G: 54.
MacCall, A. D. 1990. Dynamic geography of marine fish
populations. University of Washington Press, Seattle, WA.
153 pp.
Macpherson, E., and Duarte, C. M. 1991. Bathymetric trends
in demersal fish size; is there a general relationship? Marine
Ecology Progress Series, 71: 103–112.
Main, J., and Sangster, G. I. 1981. A study of the fish capture
process in a bottom trawl by direct observation from a towed
underwater vehicle. Scottish Fisheries Research Report 23.
Marshall, C. T., and Frank, K. T. 1994. Geographic responses
of ground fish to variation in abundance: Methods of detection and their interpretation. Canadian Journal of Fisheries
and Aquatic Sciences, 51: 808–816.
Nakken, O., and Raknes, A. 1987. The distribution and the
growth of Northeast Arctic cod in relation to bottom temperatures in the Barent Sea, 1978–1994. Fisheries Research,
5: 243–252.
Newton, A. W. 1991. Monitoring the GOV trawl. Internal
report, DAFS Marine Laboratory.
Ortega-Salas, A. A. 1988. Diet, food consumption and growth
of the dab Limanda limanda (Linnaeus) in Isle of Man waters,
UK. Anales del Instituto de Ciencias del Mar Y Limnologia,
Universidad Nacional Autonoma De Mexico, 15: 31–40.
Parrish, B. B. 1963. Some remarks on the selection processes
in fishing operations. International Commission for the
Northwest Atlantic Fisheries, Special Publication, 5:
166–170.
Parrish, B. B., Blaxter, H. S., and Hall, W. B. 1964. Diurnal
variations in size and composition of trawl catches. Rapport
et Process-Verbaux des Reunions, Conseil Permanent
Internationale pour l’Exploration de la Mer, 155: 27–34.
Pihl, L. 1994. Changes in diet of demersal fish due to
eutrophication-induced hypoxia in the Kattegat, Sweden.
Canadian Journal of Fisheries and Aquatic Sciences, 51:
321–336.
Pitt, T. K., Wells, R., and McKone, W. D. 1981. A critique of
research vessel otter trawl surveys by the St John’s research
and resources services. In Bottom trawl surveys. Ed. by
W. G. Doubleday, and D. Rivard. Canadian Special
Publication on Fisheries and Aquatic Sciences, 58: 273 pp.
Rose, G. A., and Leggett, W. C. 1989. Interactive effects of
geophysically-forced sea temperatures and prey abundance
on mesoscale coastal distributions of a marine predator,
Atlantic cod (Gadus morhua). Canadian Journal of Fisheries
and Aquatic Sciences, 46: 1904–1913.
Sahghage, D., and Wagner, G. 1978. On fluctuation in the
haddock population of the North Sea. Rapports et ProcessVerbaux des Reunions, Conseil Permanent International
pour l’Exploration de la Mer, 172: 72–85.
Scott, J. S. 1982. Depth, temperature and salinity preference of
common fishes of the Scotian Shelf. Journal of Northwest
Atlantic Fisheries Science, 3: 29–40.
Shepherd, J. G., and Forrester, J. 1987. Diurnal variation in
catchability during bottom trawl surveys off the Northwest
United States. ICES CM 1987/b: 44.
Sinclair, A. 1992. Fish distribution and partial recruitment: the
case of Eastern Scotian Shelf cod. Journal of Northwest
Atlantic Fisheries Science, 13: 15–24.
Smith, S. J., Perry, R. I., and Fanning, L. P. 1991. Relationship
between water mass characteristics and estimates of fish
population abundance from trawl surveys. Environmental
Monitoring Assessment, 17: 227–245.
Sokal, R. R., and Rohlf, F. J. 1981. Biometry, 2nd edition.
Freeman, San Francisco.
60
G. Petrakis et al.
Sparholt, H. 1990. Using GLM analysis on the IYFS herring
data for the North Sea. ICES CM 1990/H: 6.
Stewart, P. A. M., and Galbraith, R. D. 1987. Investigating the
capture efficiency of survey gears. ICES CM 1987/B: 7.
Stewart, P. A. M., Galbraith, R. D., Hall, C. D., and
MacKenzie, I. G. 1991. Influence of gear and environmental
factors on catches in Scottish young fish surveys. ICES CM
1991/B: 59.
Swain, D. P. 1993. Age- and density-dependent bathymetric
pattern of Atlantic cod (Gadus morhua) in the southern Gulf
of St Lawrence. Canadian Journal of Fisheries and Aquatic
Sciences, 50: 1255–1264.
Swain, D. P., and Wade, E. J. 1993. Density-dependent
geographic distribution of Atlantic cod (Gadus morhua) in
the southern Gulf of St Lawrence. Canadian Journal of
Fisheries and Aquatic Sciences, 50: 725–733.
Walsh, S. J. 1988. Diel variability in trawl catches of juvenile
and adult yellowtail flounder on the Grand Banks and the
effects on resource assessment. North American Journal of
Fisheries Management, 8: 373–381.
Walsh, S. J. 1991. Diel variation in availability and vulnerability of fish to a survey trawl. Journal of Applied
Ichthyology, 7: 147–159.
Walsh, S. J. 1992. Size depended selection at the footgear of
a groundfish survey trawl. North American Journal of
Fisheries Management, 12: 625–633.
Wardle, C. S. 1977. Effects of size on the swimming speed of
fish. In Scale effects in animal locomotion, pp. 299–313. Ed.
by T. J. Pedley. Academic Press, New York.
Wardle, C. S. 1993. Fish behaviour and fishing gear. In
Behaviour of teleost fishes, 2nd edn, pp. 609–643. Ed. by
T. J. Pitcher. Chapman and Hall, London.
Woodhead, P. M. J. 1964. Diurnal changes in trawl catches of
fishes. Rapport et Proces-Verbaux des Reunions, Conseil
Permanent Internationale pour l’Exploration de la Mer, 155:
35–44.