Fisheries Research 72 (2005) 71–79

Catch fluctuations of the diamond squid Thysanoteuthis rhombus

in the Sea of Japan and models to forecast CPUE based on
analysis of environmental factors
Kazutaka Miyaharaa,∗ , Taro Otab , Nobuhisa Kohnoc , Yukio Uetad , John R. Bowere
a Hyogo Tajima Fisheries Technology Institute, 1126-5 Sakae, Kasumi, Hyogo 669-6541, Japan
b Tottori Prefectural Fisheries Research Center, Tomari, Tottori 689-0602, Japan
c Fukui Prefectural Fisheries Experimental Station, 23-1 Urasoko, Tsuruga, Fukui 914-0843, Japan
d Seikai National Fisheries Research Institute, Fisheries Research Agency, 1551-8 Taira-machi, Nagasaki 851-2213, Japan
e Hokkaido University, Northern Biosphere Field Science Center, 3-1-1 Minato-cho, Hakodate, Hokkaido 041-8611, Japan

Received 7 May 2004; received in revised form 12 August 2004; accepted 3 October 2004

Abstract

This paper describes recent catch fluctuations of the diamond squid Thysanoteuthis rhombus in the Sea of Japan and the
development of models used to forecast catch per unit effort (CPUE) off Hyogo Prefecture based on statistical analyses of
environmental indices measured 600 km upstream in the Tsushima Current. Annual catches during 1989–2002 fluctuated widely,
and prefectural annual catches were closely related to each other, especially among the western prefectures. Four indices in June
were closely related with CPUE during the fishing season (September–November) off Hyogo: (1) water temperature in the
Tsushima Strait, (2) salinity in the Tsushima Strait, (3) sea level at Izuhara (Tsushima Island) and (4) sea level difference
between Izuhara and Hakata (Kyushu Island). Using these indices as independent variables, simple and multiple regression
analyses were conducted, and CPUE was accurately estimated by both regression and extrapolation, indicating that the CPUE
off Hyogo can be forecasted 2 months before the fishery starts using the described models. The strong correlation among catches
in different prefectures suggests the indices discussed here affect the catches over a large area.
© 2004 Elsevier B.V. All rights reserved.

Keywords: Thysanoteuthis rhombus; Diamond squid; Forecasting; Sea of Japan; Tsushima Strait

1. Introduction

The diamond squid, Thysanoteuthis rhombus, is

∗ Corresponding author. Tel.: +81 796 36 0395; one of the most abundant and commercially impor-
fax: +81 796 36 3684. tant squids in the Sea of Japan, where a fishery began
E-mail address: kazutaka miyahara@pref.hyogo.jp
(K. Miyahara).
in the early 1960s (Nazumi, 1975a). Annual catches

0165-7836/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.fishres.2004.10.013
72 K. Miyahara et al. / Fisheries Research 72 (2005) 71–79
in the Sea of Japan ranged from 0 to 600 tons through
the late 1980s, but have reached more than 2500 tons
since the early 1990s (Bower and Miyahara, unpub-
lished data). This recent increase in catches has been
due in part to increased fishing effort, especially in
the western Sea of Japan, and as a result of the in-
creased catches, T. rhombus is now marketed through-
out Japan (Omoto et al., 1998). As the fishery con-
tinues to grow, it will become increasingly important
for managers of the fishery to accurately forecast catch
levels.
T. rhombus is epipelagic and occurs throughout trop-
ical and subtropical regions of the world’s oceans, but
reaches higher latitudes in some subtropical and tem-
perate areas associated with warm currents (Roper et
al., 1984; Nigmatullin and Arkhipkin, 1998). Around
Japan, spawning occurs over a wide area in south-
ern waters (Bower and Miyahara, unpublished data),
including upstream areas of the Tsushima Current
(Yamamoto and Okutani, 1975), which is thought to
transport part of the population through the Tsushima
Strait and into the Sea of Japan (Fig. 1) (Nishimura,
1966; Misaki and Okutani, 1976; Okiyama, 1993; Ki-
taura et al., 1998). Little is known about this migration,
however Nazumi (1975a) suggested that T. rhombus en-
ters the Sea of Japan from late May to June, and the
occurrence of small (<20.5 cm mantle length) squid off
Hyogo Prefecture (Fig. 1) in June and July (Nazumi,
1975b) also suggests the young stages pass through the
strait in late spring or early summer.
Populations of short-lived species such as squids
are unstable, and many species show a strong
environment–recruitment relationship (Bakun and
Csirke, 1998; Dawe et al., 2000; Rodhouse, 2001). As
a result, environmental indices are increasingly being
used to predict recruitment strength in squids (Pierce et
al., 1998; Yokota et al., 1998; Robin and Denis, 1999;
Ueta et al., 1999; Agnew et al., 2002; Denis et al., 2002;
Hayashi, 2003). Tamaki (1987) suggested that fishing
conditions for T. rhombus in the Sea of Japan can be
forecasted based on the flow pattern and strength of
the Tsushima Current, and water temperature on the
fishing grounds, but he did not clearly quantify any of
these relationships. Iizuka et al. (1986) also showed
that there is an exponential relationship between the
catch amount off Minato, Kyoto Prefecture, and water
temperature around the Tsushima Strait. However, both
of these studies examined data collected 20 years ago
when the fishery was still small, and since fishery data
are affected by changes that occur over time in fishing
gear and practices that increase efficiency, an updated
analysis of recent catches is needed.
The present study describes the catch fluctuations
of T. rhombus that occurred in the Sea of Japan in
1989–2002 and identifies environmental indices that
were closely related to fluctuations in the catch per
unit effort (CPUE) off Hyogo Prefecture. We then de-
velop forecasting methods based on regression analy-
ses and discuss their goodness of fit, parsimony and
importance.
2. Materials and methods
2.1. Annual catch
In the Sea of Japan, the T. rhombus fishery extends
along the coast of Japan as far east as Toyama Pre-
fecture (Takeda and Tanda, 1998). Here we used catch
data collected in 1989–2002 from Shimane, Tottori,
Hyogo, Kyoto, Fukui, Ishikawa and Toyama prefec-
tures (Fig. 1). The Japanese national government does
not publish official catch data for T. rhombus, and all
catch data analyzed here were gathered from local gov-
ernments or prefectural fishery research institutes. All
prefectures except Hyogo started gathering catch data
after 1989, and 9–14 years of data were available from
each prefecture. Annual catch variation was measured
as the coefficient of variation (CV). Catch fluctuations
were compared among the prefectures using the cor-
relation coefficient (r) to determine how synchronized
the fluctuations were.
2.2. CPUE
CPUE was used as an index to examine the rela-
tionship between catch fluctuations and environmen-
tal indices. We examined daily catch data landed in
1989–2003 at Kasumi fishing port (Kasumi-cho Fish-
eries Cooperative Association, Fig. 1), which is where
most of the catch in Hyogo Prefecture is landed. In
this region, the main fishing season occurs during
September–November (Takeda and Tanda, 1998), so
the mean daily catch value per boat (kg/(day boat))
during these 3 months was used as the CPUE for the
year.
 K. Miyahara et al. / Fisheries Research 72 (2005) 71–79 73

Fig. 1. Map of study area showing the Tsushima Strait, Tsushima Current, prefectures (administrative districts), fishing port, tide stations and
observation sites: (a) Shimane Prefecture, (b) Tottori Prefecture, (c) Hyogo Prefecture, (d) Kasumi fishing port, (e) Kyoto Prefecture, (f) Fukui
Prefecture, (g) Ishikawa Prefecture, (h) Toyama Prefecture, (i) Izuhara tide station, (j) Hakata tide station and (䊉) 18 observation sites in the
Tsushima Strait.

2.3. Environmental indices Genkai Fisheries Research and Development Center

The environmental data collected in 1989–2003 in-
cluded: (1) water temperature in the Tsushima Strait,
(2) salinity in the Tsushima Strait, (3) sea level at
Izuhara (Izuhara Tide Station of the Japan Coast Guard,
Tsushima Island, Nagasaki Prefecture; Fig. 1), and (4)
sea level difference (SLD) between Izuhara and Hakata
(Hakata Tide Station of the Japan Coast Guard, Kyushu
Island, Fukuoka Prefecture; Fig. 1). Water temperature
and salinity were measured by the Saga Prefectural
in early June at 18 stations (Fig. 1). Seafloor depths at
the stations ranged from ca. 30 to 140 m (mean = ca.
80 m). Mean temperature and salinity values from all
stations were determined for each of the following
depths: 0, 10, 20, 10 and 20 m combined, and at the
seafloor. The sea level at Izuhara was derived from the
Japan Oceanographic Data Center’s J-DOSS database
(http://www.jodc.go.jp/service j.htm). Monthly mean
values were determined from May to August during
1989–2003 for all months except August 1989 and May
74 K. Miyahara et al. / Fisheries Research 72 (2005) 71–79
1990, when no data were available. The SLD between
Hakata and Izuhara can be used as an indicator of the
flow strength of the Tsushima Current through the east-
ern channel of the Tsushima Strait (Kawabe, 1982),
and has been used to examine the catch fluctuations of
fishes in the Sea of Japan such as Japanese amberjack
Seriola quinqueradiata (Hara, 1990). These SLD data
also came from the J-DOSS database.
2.4. Environmental indices related to CPUE and
development of forecasting models
Environmental indices that were significantly corre-
lated with CPUE (P < 0.05) were selected and tested for
multicollinearity. Simple or multiple regression models
using the indices as independent variables and CPUE
as the dependent variable were applied using data col-
lected in 1989–2003 (n = 15). The goodness of fit and
parsimony of the models were examined based on three
values: (1) the significance level of the simple or mul-
tiple correlation coefficient r or R (F-tested for R), (2)
the sum of squares (SS) for the difference between
the observed and predicted CPUEs and (3) the Akaike
information criterion (AIC) value, which provides a
combined measure of statistical fit and model parsi-
mony (Akaike, 1974). We also tested the extrapolation
for forecasting using the method of Sakuramoto et al.
(1995). First, we estimated the parameters of regression
models using data collected during 1989–1998 (n = 10)
and forecasted the CPUE for 1999 using the indepen-
dent variables for 1999. Next, we estimated the new
parameters using data from 1989 to 1999 (n = 11) and
then forecasted the CPUE for 2000 using the variables
for 2000. Repeating this procedure, we then forecasted
the CPUE for 2003. The SS between the observed and
forecasted CPUEs along with the AIC value were then
compared with the regression models mentioned above.
3. Results
3.1. Fluctuation of catches in the Sea of Japan
Annual catches in the Sea of Japan fluctuated widely
from 52 to 3700 tons in 1989–2002 (overall CV = 82.2,
Hyogo CV = 78.2) and were highest every year in
Hyogo Prefecture, followed by Tottori, Shimane, Fukui
and Kyoto prefectures (Fig. 2). Correlations among
Fig. 2. Annual prefectural catches of the diamond squid in the Sea
of Japan in 1989–2003.
catch amounts in the five westernmost prefectures (Shi-
mane, Tottori, Hyogo, Kyoto and Fukui) were highly
significant (P < 0.001), as was the coefficient between
Ishikawa and Toyama (P < 0.01) (Table 1), indicat-
ing that these recent fluctuations in prefectural catches
were generally synchronized throughout the Sea of
Japan, especially in the western prefectures.
3.2. Indices related to CPUE and development of
forecasting models
CPUE ranged 14.5–179.7 kg/(day boat) and was
closely related to nine environmental indices (Table 2):
(1) the mean water temperature (MWT) in June at 0 m,
(2) the MWT in June at 10 m, (3) the MWT in June at
20 m, (4) the MWT in June at 10–20 m, (5) the MWT
in June at the seafloor, (6) the mean salinity in June
at the seafloor, (7) the sea level at Izuhara in June, (8)
the sea level at Izuhara in July, and (9) the SLD be-
tween Hakata and Izuhara in June. The MWTs were
strongly correlated at different depths and with the sea
levels in June (P < 0.001) and July (P < 0.05). In addi-
tion, the sea levels in June and July were highly cor-
related (P < 0.001), and the salinity at the seafloor was
correlated with MWTs at 0 and 10 m (P < 0.05).
Four indices with high absolute values of r were
then selected as independent variables: the MWT at
10–20 m depth (X1 ), the mean salinity at the seafloor
(X2 ), the sea level at Izuhara in June (X3 ) and the SLD
in June (X4 ). Of these four, X1 and X3 showed signif-
icant multicollinearity (P < 0.001) (Table 2), so they
were not used together in the multiple regression mod-
els described below. The relationship between CPUE
and each index was positive for all variables except X2 ,
and the coefficients of determination (r2 ) were highest
for X1 (r2 = 0.83) and X3 (r2 = 0.66) (Fig. 3).
 K. Miyahara et al. / Fisheries Research 72 (2005) 71–79 75

Table 1
Correlation coefficient r between annual prefectural catches in 1989–2002 (n)
Shimane Tottori Hyogo Kyoto Fukui Ishikawa Toyama
Shimane 0.789a 0.812a 0.857b 0.866b 0.563 0.382
Tottori 11 0.973b 0.938b 0.941b 0.362 0.187
Hyogo 12 11 0.966b 0.953b 0.458 0.243
Kyoto 12 11 11 0.973b 0.579 0.424
Fukui 11 11 11 11 0.510 0.294
Ishikawa 9 9 9 9 9 0.895a
Toyama 12 11 14 13 11 9

n: number of years used for correlation from 1989 to 2002.

a P < 0.01.
b P < 0.001

Table 2
Correlation matrix between CPUE and environmental indices
MWT MS, seafloor (X2 ) SL SLD, June (X4 )
0m 10 m 20 m 10–20 m (X1 ) Seafloor June (X3 ) July
CPUE 0.866c 0.900c 0.900c 0.911c 0.864c −0.592a 0.812c 0.634a 0.539a
MWT
0m 0.946c 0.880c 0.926c 0.864c −0.599a 0.752c 0.613a 0.396
10 m 0.952c 0.989c 0.897c −0.538a 0.872c 0.689b 0.465
20 m 0.987c 0.967c −0.442 0.902c 0.777c 0.400
10–20 m (X1 ) 0.942c −0.450 0.798c 0.733b 0.237
Seafloor −0.497 0.897c 0.740b 0.439
MS
Seafloor (X2 ) −0.417 −0.263 −0.319
SL
June (X3 ) 0.814c 0.492
July 0.404
SLD (X4 )

Data used for correlation were from 1989 to 2003 (n = 15). r: correlation coefficient, MWT: mean water temperature, MS: mean salinity, SL:
sea level at Izuhara, SLD: sea level difference between Hakata and Izuhara, Xi : indices selected as independent variables for regression models.
a P < 0.05.
b P < 0.01.
c P < 0.001.

Table 3
Variables used and values of the sum of squares (SS) and AIC
Model Group A Group B Group C, 9

1 2 3 4 5 6 7 8
Variables (Xi ) used for regression (i=) 1 1, 2 1, 4 1, 2, 4 3 2, 3 3, 4 2, 3, 4 2, 4
Regression for 1989–2003 (n = 15)
SS (×102 ) 61.9 52.6 53.3 46.0 123.7 95.5 114.3 90.1 186.6
AIC 136.9 136.5 136.7 136.5 147.3 145.4 148.1 146.5 155.5
Extrapolation for 1999–2003 (n = 10 + 1 × 5)
SS (×102 ) 64.8 55.4 70.5 55.3 134.1 108.1 153.5 116.2 248.7
AIC 137.6 137.2 140.9 139.2 148.5 147.3 152.5 150.4 159.8

The two lowest values in each row are italicised.

76 K. Miyahara et al. / Fisheries Research 72 (2005) 71–79
Fig. 3. Relationships between mean CPUE in September–November off Hyogo Prefecture and environmental indices in June around the Tsushima
Strait during 1989–2003. The value of r2 represents the coefficient of determination for each relationship. Broken lines show 95% confidence
limits.
To estimate CPUE as the dependent variable (Y),
we tested nine regression models (two simple regres-
sion models and seven multiple regression models) us-
ing the four indices. Since X1 and X3 showed signif-
icant multicollinearity, the models were divided into
three groups (Table 3): Group A, Models 1–4, all in-
cluded X1 as an index; Group B, Models 5–8, all in-
cluded X3 as an index; and Group C, Model 9, which
had X2 and X4 as indices. X2 and X4 were not used
in the simple regression models because of their low
coefficients of determination. Using the 1989–2003
catch data set, the model parameters were expressed as
follows:
• Group A:
Model 1: Y = 61.7X1 − 1098.4, r = 0.911
(P < 0.001);
Model 2: Y = 55.5X1 − 87.8X2 + 2052.1,
R = 0.925 (P < 0.001);
Model 3: Y = 56.5X1 + 4.1X4 − 1014.5, R = 0.924
(P < 0.001);
Model 4: Y = 51.7X1 − 78.2X2 + 3.6X4 + 1780.8,
R = 0.935 (P < 0.001).
• Group B:
Model 5: Y = 1043.9X3 − 1790.5, r = 0.812
(P < 0.001);
Model 6: Y = −145.9X2 + 879.8X3 + 3537.9,
R = 0.859 (P < 0.001);
Model 7: Y = 927.7X3 + 4.4X4 − 1596.8,
R = 0.828 (P < 0.001);
Model 8: Y = −136.6X2 + 801.3X3 + 3.3X4 +
3347.6, R = 0.867 (P < 0.001).
• Group C:
Model 9: Y = −222.4X2 + 9.2X4 + 7728.5,
R = 0.697 (P < 0.05).
The correlation coefficient of each model differed
significantly from zero (P < 0.05), and the SS and AIC
values were smallest in Group A (Table 3). In Group
A, the SS was smallest in Models 4 and 2, and the AIC
was similar in all four models. In Group B, the SS was
smallest in Model 8, and the AIC was smallest in Model
6. The results of extrapolation for forecasting with SS
and AIC values are also shown in Table 3. In Group A,
the SS was smallest in Models 4 and 2, and the AIC
was smallest in Models 2 and 1. In Group B, both the
SS and AIC were smallest in Model 6.
Based on these results, we selected Models 1, 2, 4
and 6 to compare the observed value of CPUE with
those outputted from both regression and extrapolation
(Fig. 4). In all models, the extrapolated values of the
SS and AIC were slightly higher than those of the re-
gressions (Table 3).
 K. Miyahara et al. / Fisheries Research 72 (2005) 71–79 77

Fig. 4. Comparison between observed value of CPUE and outputted ones from both regression and extrapolation: (
) observed CPUE, ()
outputted CPUE from regression and () outputted CPUE from extrapolation. See Table 3 as for the variables used for each model. Also refer
to the text regarding the process of extrapolation.

4. Discussion cruitment following periods of warmer temperatures

Catch amounts in the Sea of Japan showed similar
trends among prefectures and were significantly cor-
related among the western prefectures from Shimane
to Fukui, suggesting these fluctuations were regulated
by large-scale mechanisms. Dawe and Warren (1993)
reported that recruitment of the ommastrephid squid
Illex illecebrosus in the northwest Atlantic also is reg-
ulated by broad-scale mechanisms and hypothesized
that climatic variation is the key mechanism affecting
this recruitment. More detailed studies on differences
in the peak catch season (Takeda and Tanda, 1998), the
size of landed squid, growth, and migration among pre-
fectures are needed to determine the factors responsible
for T. rhombus stock-size fluctuations.
Strong relationships were found between CPUE off
Hyogo in September–November and several environ-
mental indices, including temperature at all depths
sampled in the Tsushima Strait in June. This suggests
temperature has an important effect on the recruitment
success of T. rhombus. Squid landings are closely re-
lated to seawater temperature in many squid species
(Robin and Denis, 1999; Dawe et al., 2000; Sakurai
et al., 2000), and similar patterns of increased re-
have been reported for I. illecebrosus (Dawe and
Warren, 1993) and Ommastrephes bartramii (Yatsu et
al., 2000). Bakun and Csirke (1998) have described
several possible mechanisms by which reduced tem-
peratures could adversely affect recruit success by de-
laying the development of prey for the young squid (i.e.,
the match/mismatch hypothesis of Cushing, 1975) and
slowing the growth of squid, thus increasing the pe-
riod during which they are highly vulnerable to pre-
dation. Zooplankton biomass also tends to decrease in
the southern Sea of Japan during years of decreased
temperature (Hirota and Hasegawa, 1999; Kang et al.,
2002), which could also adversely affect the year class.
The sea level difference between Hakata and Izuhara
was positively correlated with CPUE, which suggests
that catches increase when the flow strength of the
nearshore branch of the Tsushima Current through
the eastern channel of the Tsushima Strait (Fig. 1)
increases. Nishimura (1966) suggested that T. rhom-
bus is transported through the western channel of the
strait and then northward by the offshore branch of
the Tsushima Current, however, transport through the
eastern channel is also highly likely. The water temper-
ature differs between these branches (Watanabe et al.,
78 K. Miyahara et al. / Fisheries Research 72 (2005) 71–79
2003), and Kidokoro and Hiyama (1996) reported that
the growth of Japanese common squid Todarodes paci-
ficus varies between warm and cold-water areas in the
Sea of Japan. More information is needed on the mi-
gration routes of T. rhombus to better understand how
variation in the flow of the Tsushima Current affects its
stock size in the Sea of Japan.
The occurrence of young squid as small as 8.1 cm
in mantle length off Hyogo in June and July (Nazumi,
1975b) suggests that T. rhombus passes through the
Tsushima Strait in its post-paralarval and early juve-
nile stages. The young stages of squids are particularly
sensitive to environmental change (Waluda et al., 1999;
Sakurai et al., 2000; Rodhouse, 2001; Waluda et al.,
2001), which is presumably why such strong relation-
ships were found between CPUE and the environmen-
tal indices in the strait.
The other index besides temperature that showed a
highly significant correlation with CPUE was the sea
level at Izuhara in June. The models using temperature
(Group A) fit the observed data better than those us-
ing sea level (Group B), however, there are some cases
in which the Group B models might be preferable. Sea
level is calculated based on hourly tidal measurements,
so it is more stable than water temperature, especially
when observations are delayed, skipped, or made just
after atmospheric events such as the passage of a ty-
phoon or depression. Sea level data can also be used
to cross-check forecasts made using water temperature
data.
Salinity at the seafloor in June had a significant neg-
ative relationship with CPUE and statistically can be
used as an index in multiple regression models. The
young stages of T. rhombus generally occur in the up-
per 50 m (Yamamoto and Okutani, 1975), so there is
probably no direct effect of high salinity at the bottom
on T. rhombus. Rather, since salinity was also nega-
tively related with MWTs at 0 and 10 m in June, this
increased salinity might be a result of delayed seasonal
change of the Tsushima Current. The eastern part of the
Tsushima Strait is occupied by high salinity water from
March to May, after which low salinity water begins to
appear in the surface layer and deeper (Kawabe, 1982)
when coastal waters from the Chinese continent start to
influence southern upstream areas of the Tsushima Cur-
rent, where the SST is 4–6 ◦ C higher than that around
the Tsushima Strait in mid-May (Inoue, 1974). The va-
lidity of this suggestion needs to be further examined
through obtaining more extensive data on the relation-
ship between salinity and distribution.
The models described above have been used to accu-
rately forecast the CPUE off Hyogo up to 2 months be-
fore the peak catch season starts, and the Hyogo Tajima
Fisheries Technology Institute began making official
forecast announcements to the public using these mod-
els in 2003. The strong correlation between the catch
fluctuations in the prefectures discussed here suggests
that the environmental factors used in this study can be
used to forecast catches of T. rhombus in other prefec-
tures in the Sea of Japan as well.
Acknowledgments
We thank Drs. O. Kato and H. Yamada, Japan
Sea National Fisheries Research Institute, Fisheries
Research Agency, for their valuable advice on the
Tsushima Current. We also thank the researchers of
the prefectural fisheries research institutions of Saga,
Shimane, Kyoto, Ishikawa and Toyama for allowing
us to use their oceanographic and catch data. Thanks
are also due to R. Takeda, Hyogo Tajima Fisheries
Technology Institute, and M. Tanda, Hyogo Fisheries
Technology Institute, for their thoughtful comments
and encouragement.
References
Agnew, D.J., Beddington, J.R., Hill, S.L., 2002. The potential use of
environmental information to manage squid stocks. Can. J. Fish.
Aquat. Sci. 59, 1851–1857.
Akaike, H., 1974. A new look at the statistical model identification.
IEEE Trans. Automat. Contr. 19, 716–723.
Bakun, A., Csirke, J., 1998. Environmental processes and recruitment
variability. In: Rodhouse, P.G., Dawe, E.G., O’Dor, R.K. (Eds.),
Squid Recruitment Dynamics. The Genus Illex as a Model. The
Commercial Illex Species. Influences on Variability. FAO Fish-
eries Technical Paper No. 376. FAO, Rome, pp. 105–124.
Cushing, D.H., 1975. Marine ecology and fisheries. Cambridge Uni-
versity Press, London, 278 pp.
Dawe, E.G., Warren, W.G., 1993. Recruitment of short-finned squid
in the Northwest Atlantic Ocean and some environmental rela-
tionships. J. Cephalopod Biol. 2, 1–21.
Dawe, E.G., Colbourne, E.B., Drinkwater, K.F., 2000. Environmental
effects on recruitment of short-finned squid Illex illecebrosus.
ICES J. Mar. Sci. 57, 1002–1013.
Denis, V., Lejeune, J., Robin, J.P., 2002. Spatio-temporal analysis of
commercial trawler data using general additive models: patterns
of loliginid squid abundance in the northeast Atlantic. ICES J.
Mar. Sci. 59, 633–648.
 K. Miyahara et al. / Fisheries Research 72 (2005) 71–79
Hara, T., 1990. An abundance index of yellowtail immigrating
into the Sea of Japan and its yearly variation. Nippon Suisan
Gakkaishi 56, 19–24 (in Japanese with English abstract).
Hayashi, S., 2003. Recent trend in oval squid, Sepioteuthis lessoni-
ana, fishery in Toyama Bay, Japan Sea. Bull. Toyama Pref. Res.
Inst. 14, 11–28 (in Japanese with English abstract).
Hirota, Y., Hasegawa, S., 1999. The zooplankton biomass in the Sea
of Japan. Fish. Oceanogr. 8, 274–283.
Iizuka, S., Inoue, H., Munekiyo, M., 1986. Fishing conditions of
oceanic squid in the sea off Kyoto Prefecture. Bull. Kyoto Ocean.
Fish. Sci. 10, 13–18 (in Japanese with English abstract).
Inoue, N., 1974. Oceanographic features of the seas off western
Japan. In: Nippon Suisan Gakkai (Ed.), Tsushima Current. Ko-
seisha Koseikaku, Tokyo, pp. 27–41 (in Japanese).
Kang, Y.S., Kim, J.Y., Kim, H.G., Park, J.H., 2002. Long-term
changes in zooplankton and its relationship with squid, Todar-
odes pacificus, catch in Japan/East Sea. Fish. Oceanogr. 11,
337–346.
Kawabe, M., 1982. Branching of the Tsushima Current in the Japan
Sea. Part I. Data analysis. J. Oceanogr. Soc. Jpn. 38, 95–107.
Kidokoro, H., Hiyama, Y., 1996. Spatial variation in growth of
Japanese common squid, Todarodes pacificus STEENSTRUP in
the Sea of Japan. Bull. Jpn. Sea Natl. Fish. Res. Inst. 46, 77–86.
Kitaura, J., Yamamoto, G., Nishida, M., 1998. Genetic variation in
populations of the diamond-shaped squid Thysanoteuthis rhom-
bus as examined by mitochondrial DNA sequence analysis. Fish.
Sci. 64, 538–542.
Misaki, H., Okutani, T., 1976. Studies on early life history of de-
capodan Mollusca. VI. An evidence of spawning of an oceanic
squid, Thysanoteuthis rhombus TROCSHEL in the Japanese wa-
ters. Venus 35, 211–213.
Nazumi, T., 1975a. Notes on the fishery and the ecology of the squid,
Thysanoteuthis rhombus Trocshel, in the east San’in water. Bull.
Hyogo Pref. Fish. Stat. 15, 15–34 (in Japanese).
Nazumi, T., 1975b. Notes on the species of squid, collected in the
Tajima region of the Japan Sea in 1974. Bull. Hyogo Pref. Fish.
Stat. 15, 9–13 (in Japanese).
Nigmatullin, C.M., Arkhipkin, A.I., 1998. A review of the biol-
ogy of the diamondback squid, Thysanoteuthis rhombus (Oegop-
sida: Thysanoteuthidae). In: Okutani, T. (Ed.), Contributed Pa-
pers to International Symposium on Large Pelagic Squids. Japan
Marine Fishery Resources Research Center, Tokyo, pp. 155–
181.
Nishimura, S., 1966. Notes on the occurrence and biology of the
oceanic squid, Thysanoteuthis rhombus Trocshel, in Japan. Publ.
Seto Mar. Biol. Lab. 14, 327–349.
Okiyama, M., 1993. Kinds, abundance and distribution of the oceanic
squids in the Sea of Japan. In: Okutani, T., O’Dor, R.K., Ku-
bodera, T. (Eds.), Recent Advances in Cephalopod Fisheries Bi-
ology. Tokai University Press, Tokyo, pp. 403–415.
Omoto, S., Kajita, A., Shiraishi, H., 1998. The marketing and price
formation of diamond squid Thysanoteuthis rhombus in Japan.
In: Okutani, T. (Ed.), Contributed Papers to International Sympo-
sium on Large Pelagic Squids. Japan Marine Fishery Resources
Research Center, Tokyo, pp. 199–205.
Pierce, G.J., Bailey, N., Stratoudakis, Y., Newton, A., 1998. Distri-
bution and abundance of the fished population of Loligo forbesi
79
in Scottish waters: analysis of research cruise data. ICES J. Mar.
Sci. 55, 14–33.
Robin, J.P., Denis, V., 1999. Squid stock fluctuations and water tem-
perature: temporal analysis of English Channel Loliginidae. J.
Appl. Ecol. 36, 101–110.
Rodhouse, P.G., 2001. Managing and forecasting squid fisheries in
variable environments. Fish. Res. 54, 3–8.
Roper, C.F.E., Sweeney, M.J., Nauen, C.E., 1984. An annotated and
illustrated catalogue of species of interest to fisheries. In: FAO
Species Catalogue, vol. 3, Cephalopods of the World. FAO Fish.
Synop. 125, 187–188.
Sakurai, Y., Kiyofuji, H., Saitoh, S., Goto, T., Hiyama, Y., 2000.
Changes in inferred spawning areas of Todarodes pacificus
(Cephalopoda: Ommastrephidae) due to changing environmental
conditions. ICES J. Mar. Sci. 57, 24–30.
Sakuramoto, K., Nihira, A., Ohnishi, S., 1995. Factors of catch fluc-
tuations of skipjack tuna in the north-eastern waters of Japan and
its forecasting. Fish. Sci. 61, 921–925.
Takeda, R., Tanda, M., 1998. Fishing and migration of Thysan-
oteuthis rhombus Trocshel in the Japan Sea. In: Okutani, T.
(Ed.), Contributed Papers to International Symposium on Large
Pelagic Squids. Japan Marine Fishery Resources Research Cen-
ter, Tokyo, pp. 191–198.
Tamaki, T., 1987. Fishing condition of Thysanoteuthis rhombus in
the Tajima region in 1985. Contr. Fish. Res. Jpn. Sea Block 9,
99–104 (in Japanese).
Ueta, Y., Tokai, T., Segawa, S., 1999. Relationship between year-
class abundance of the oval squid Sepioteuthis lessoniana and
environmental factors off Tokushima Prefecture, Japan. Fish. Sci.
65, 424–431.
Waluda, C.M., Trathan, P.N., Rodhouse, P.G., 1999. Influence of
oceanographic variability on recruitment in the Illex argentinus
(Cephalopoda: Ommastrephidae) fishery in the South Atlantic.
Mar. Ecol. Prog. Ser. 183, 159–167.
Waluda, C.M., Rodhouse, P.G., Podesta, G.P., Trathan, P.N., Pierce,
G.P., 2001. Surface oceanography of the inferred hatching
grounds of Illex argentinus (Cephalopoda: Ommastrephidae)
and influences on recruitment variability. Mar. Biol. 139, 671–
679.
Watanabe, T., Ichihashi, M., Yamada, H., Katoh, O., 2003. Mean
water temperature in the Japan Sea (1971–2000). Contr. Fish.
Res. Jpn. Sea Block 41, 1–112 (in Japanese).
Yamamoto, K., Okutani, T., 1975. Studies on early life history of de-
capodan Mollusca. V. Systematics and distribution of epipelagic
larvae of decapod cephalopods in the southwestern waters of
Japan during summer in 1970. Bull. Tokai Reg. Fish. Res. Lab.
83, 45–96.
Yatsu, A., Mori, J., Tanaka, H., Watanabe, T., Nagasawa, K., Ishida,
Y., Meguro, T., Kamei, Y., Sakurai, Y., 2000. Interannual variabil-
ity in stock abundance of the neon flying squid, Ommastrephes
bartramii, in the North Pacific Ocean during 1979–1998: impact
of driftnet fishing and oceanographic conditions. Fish. Oceanogr.
9, 163–170.
Yokota, M., Kitada, S., Udono, K., Watanabe, S., 1998. Prediction
of the firefly squid catch by analysis of environmental factors in
Toyama Bay. Nippon Suisan Gakkaishi 64, 975–978 (in Japanese
with English abstract).