Abstract–
Age and growth of the swordfish (Xiphias gladius) in Taiwan waters was studied from counts of growth bands on cross sections of the second ray of the first anal fin. Data on lower jaw fork length and weight, and samples of the anal fin of male and female swordfish were collected from three offshore and coastal tuna longline fishing ports on a monthly basis between September 1997 and March 1999. In total, 685 anal fins were collected and 627 of them (293 males and 334 females) were aged suc cessfully. The lower jaw fork lengths of the aged individuals ranged from 83.4 to 246.6 cm for the females and from 83.3 to 206 cm for the males.The radii of the fin rays and growth bands on the cross sections were mea sured under a dissecting microscope equipped with an image analysis system. Trends in the monthly mar ginal increment ratio indicated that growth bands formed once a year. Thus, the age of each fish was deter-mined from the number of visible growth bands. Two methods were used to estimate and compare the standard and the generalized von Bertalanffy growth parameters for both males and females. The nonlinear least square estimates of the generalized von Berta lanffy growth parameters in method II, in which a power function was used to describe the relationship between ray radius and LJFL, were recommended as most acceptable. There were sig nificant differences in growth param eters between males and females. The growth parameters estimated for females were the following: asymptotic length (L∞) = 300.66 cm, growth coef ficient (K) = 0.040/yr, age at zero length (t0) = –0.75 yr, and the fitted fourth parameter (m) = –0.785. The growth parameters estimated for males were the following: asymptotic length (L∞) = 213.05 cm, growth coefficient (K) = 0.086/yr, age at zero length (t0) = –0.626 yr, and the fitted fourth parameter (m) = –0.768.
Manuscript accepted 21 February 2002. Fish. Bull. 100:822–835 (2002).
Age and growth of the swordfish
(Xiphias gladius L.) in the waters
around Taiwan determined from anal-fin rays
Chi-Lu Sun
Sheng-Ping Wang
Su-Zan Yeh
Institute of Oceanography National Taiwan University No. 1, Sec. 4, Roosevelt Road Taipei, Taiwan 106E-mail address (for C. L. Sun): chilu@ccms.ntu.edu.tw
The swordfish (Xiphias gladius L.) is a et al., 1995), and vertebrae (Esteves et cosmopolitan species found in the trop- al., 1995). In contrast, only a few at ical, subtropical, and temperate waters tempts have been made to determine of the world’s oceans and adjacent seas the age of swordfish in the Pacific (Sakagawa, 1989). In the Pacific Ocean, Ocean. Yabe et al. (1959) estimated the swordfish is generally distributed from growth of swordfish caught in the Asia to the Americas between 50°N western North Pacific (140°–160°E) by and 50°S (Bartoo and Coan, 1989). In longline during the period from 1948 the waters of Taiwan, the swordfish is to 1956 using the modal analysis of an incidental bycatch of the offshore length frequencies. Castro-Longoria tuna longline and harpoon fisheries. and Sosa-Nishizaki (1998) compared Both fisheries contributed an esti- the age estimates of swordfish caught mated 1528 metric tons (99%) to the by drift gillnet vessels off Baja Califor total swordfish landings from Taiwan nia from 1992 to 1993 based on otolith waters in 1999. microstructure and cross sections of
Information on age and growth of the second ray from the first anal fin, fishes is a central element in fishery and highly recommended the use of management (Brothers, 1983). Mea- cross sections of the second ray to de surement of the age of the fish provides termine the ages of swordfish in the the key variable of time needed to es- Pacific Ocean. Uchiyama et al. (1998) timate life history and biology factors, evaluated various hard parts (includ such as mortality and growth. Mortal- ing rays of the first dorsal and first ity and growth-rate models provide anal fins, vertebrae, and sagittae) for quantitative information on the status aging swordfish in the central North of fish stocks and at the same time may Pacific by Hawaii longline fishery from be used in more sophisticated models, 1991 to 1993, and provided preliminary such as yield-per-recruit analyses and estimates of length-at-age.
cohort analyses (Powers, 1983), which The objectives of our study were to will directly contribute to the rational estimate the age and growth of sword-exploitation of fish resources, as well as fish by counting the growth rings on to the development of proper manage- the cross sections of the second ray of
ment plans. the first anal fin and to compare the
Most age determination studies of generalized growth function proposed swordfish have dealt with Atlantic pop- by Richards (1959) with the standard ulations and have used different hard von Bertalanffy model for represent-parts, such as anal-fin rays (Berkeley ing the best growth model of swordfish and Houde, 1983; Wilson and Dean around Taiwan waters. The informa-1983; Prince et al., 1988; Ehrhardt, tion is crucial because it will allow 1992; Esteves et al., 1995; Ehrhardt et the age composition of the catch to be al., 1996), otolith (Radtke and Hurley, determined, which in turn will allow 1983; Wilson and Dean, 1983; Esteves the status of the swordfish stock in the
T
A
IW
A
N
Nanfangao Shinkang Tungkang 119 120 121 22 123 20.0 20.5 21.0 21.5 22.0 22.5 23.0 23.5 24.0 24.5 25.0 25.5 E 120 140 160 180 160 140 120 100 80 W 40 20 0 20 40 N S Taiwan Pacific Ocean N E Figure 1Three fi shing ports in Taiwan where the swordfi sh anal-fi n ray samples were collected in this study.
Figure 2
The fi shing grounds (oblique dot lines) of the long-line and harpoon fi shing boats based in Tungkang (A), Nanfangao (B), and Shinkang (C) fi shing ports.
waters around Taiwan to be assessed by using yield-per-recruit and sequential population analyses.
Material and methods
Data on lower jaw fork length (LJFL) and weight, and samples of the fi rst anal fi n of male and female sword-fi sh were collected from three offshore tuna longline and harpoon fi shing ports (Fig. 1) on a monthly basis between September 1997 and March 1999. The fi shing grounds for those vessels are shown in Figure 2. In total, 769 LJFLs and 685 anal fi ns were collected. The anal fi ns were frozen for approximately one month before being thawed and boiled to remove the tissue and to separate the second rays. Three cross sections ranging in thickness from 0.5 to 0.75 to 1.0 mm were taken successively along the length of each ray with a low-speed “ISOMET” saw (model no. 11-1280) and diamond wafering blades, at a location equivalent to 1/2 of the maximum width of the condyle base mea-sured above the line of maximum condyle width (Fig. 3)
(Ehrhardt, 1992; Ehrhardt et al., 1996). The sections were immersed in 95% ethanol for several minutes, placed in a labeled (with sampling date and a number) small plastic case to air dry, and then stored for later reading. Anal-fi n ray sections were then taken from the small cases randomly and examined under a dissecting microscope (model: Leica MZ 6) with transmitted light at various
A
B
C
0.5 ~ 1 mm d CONDYLE BASE d/2 SPINE SHAFT TRANSVERSE SECTION SECTION
A
B
Figure 3The second ray of the first anal fin showing the location of the cross section at a distance equal to a half of the width of the condyle base above the base (A), and a section of the second ray of the first anal fin (B).
magnifications from 8× to 16× depending on the size of the section. The clearest one of the three sections from each fin ray was read three times by one reader about one to three months apart with no knowledge of fish length. The pre cision of readings was evaluated as the average percent error (APE, Beamish and Fournier, 1981) and coefficient of variation (CV, Campana et al., 1995, 2001). For those sections resulting in three different readings, each section was reread by two to three readers simultaneously. Speci mens whose age estimates still disagreed were omitted from further analyses.
The images of the anal-fin ray sections were captured by using an Image Analysis Software package (Media Cybernetics, 1997) in combination with a dissecting mi croscope equipped with a charged coupled device (CCD) camera (model: Toshiba IK-630) and a Pentium II com puter equipped with a 640 × 480 pixel frame grab card and a 800 × 600 pixel monitor. The images were measured in microns after distance calibrations were incorporated. The distances from the focus to the distal edge of the sec tion (ray radius) and from the focus to the distal edge of each growth band (annulus) were measured and recorded
(Fig. 4). The focus, the growth band, and the false growth band (multiple bands) were defined according to Berkeley and Houde (1983), Tserpes and Tsimenides (1995), and Ehrhardt et al. (1996).
The marginal increment ratio (MIR), which was used to validate the reading of annuli, was estimated for each specimen by the following formula (Prince et al., 1988; Esteves et al., 1995):
MIR = (S – Sn)/(Sn – Sn–1), where S = ray radius; and
Sn and Sn–1= the distance from ray focus to bands n and n−1, respectively.
The mean MIR and the standard deviation were computed for each month by sex for all ages combined and also for each age separately.
Growth was analyzed by using the back-calculation of length-at-age for each sex. For this purpose, a relation-ship was determined between the ray radius and the LJFL. This relationship and the distance from the focus
FOCUS 1 S FOCUS 1 2 3 4 5 S FOCUS S 1 2 3 4 5 6 7 8 9 10 (1 mm)
A
B
C
Figure 4The section of three typical second anal fin rays of sword-fish. Ray radius (S) measured from focus to edge; annuli for estimated age 1+ (A), age 5+ (B) and age 11+ (C).
to successive rings were used to back-calculate lengths at presumed previous ages (Ehrhardt, et al., 1996). For the relationship between ray radius and LJFL and the back-calculation of lengths-at-age, the following two methods were used.
Method I
The relationship between ray radius (S) and LJFL (L) was determined by using the standard linear regression procedure, L = a + bS (Berkeley and Houde, 1983). This relationship and the distance from focus to successive growth bands, which we assumed to be based on annual growth events, were used to back-calculate the lengths at presumed ages by the following formula (Fraser, 1916):
S
a n −
Ln − = (L a),
S where L = LJFL at time of capture;
Ln = LJFL when band n was formed;
a = the intercept on the length axis from the regression line of length (L) on ray radius (S), e.g. L = a + bS; and
Sn = the distance from ray focus to band n. Method II
The relationship between ray radius and LJFL was determined by using a power function procedure, L = aSb
(Ehrhardt, 1992; Ehrhardt et. al., 1996). Parameters of this function were estimated by nonlinear least square fits to the observed data. This relationship and the distance from focus to successive growth bands were used to back-calculate the lengths at presumed ages by the following formula (Tserpes and Tsimenides, 1995; Ehrhardt et al., 1996):
b
Sn
Ln =
S L,
where b = the exponent of the regression of length (L) on ray radius (S) which is assumed to be a power function of the form L = a Sb.
The data of the back-calculated length-at-age from method I and method II were then applied to the following stan dard von Bertalanffy growth equation (standard VB) and to the generalized growth function (generalized VB) (Rich ards, 1959): Standard VB: ( Lt = L∞
(
1 − e− k t−t0))
; Generalized VB: 1 Lt = L∞(
1 − e− K (1−m)(t−t0))
1− m ,where Lt = the mean LJFL at age t; L∞ = the asymptotic length;
t0 = the hypothetical age at length zero; k and K = the growth coefficients; and
m = the fitted fourth growth-function parameter. Parameters of the above two equations for male and fe male were estimated, respectively, by fitting a curve to the observed back-calculated LJFL-at-age by using a nonlinear least square procedure (Gauss-Newton method, NLIN of SAS Institute, 1990). The measure of goodness-of-fit chosen was r2. A multivariate statistical procedure (Hotelling’s T 2)
was used to test for differences in growth between males and females (Bernard, 1981) for the two growth models and two methods. The r2 values were ranked between the two
different growth functions with the smaller as 1. A non-parametric test (Friedman 1937, 1940) was then employed
Figure 5
The size-frequency distribution by 5-cm intervals (A) and the proportions (B) of female and male swordfish collected from Tungkang, Nanfangao, and Shinkang fish markets, September 1997 to March 1999.
B
A
F requency Propor tionLower jaw fork length (cm)
on the ranked results of r2 to test the significance of the
goodness-of-fit comparisons between the two growth func tions (Chen et al., 1992). Friedman’s test statistic χ2
r was
calculated with a correction for tied ranks (Zar, 1999). The association of rank ordering of r2 values between the two
growth functions was measured nonparametrically by us ing the Kendall coefficient of concordance (Kendall, 1962).
Results
Of the 685 anal fin rays sampled, 627 (334 females and 293 males) were aged successfully. The average percent error (APE) was 5.18% (5.09% for females and 5.29% for males) and the coefficient of variation (CV) was 8.50% (8.28% for female and 8.76% for males). The LJFL of the aged fish ranged from 83.4 to 246.6 cm for the females and from 83.3 to 206 cm for the males. The remaining 58 fin rays
were considered unreadable mostly because the opaque translution zonation was so unclear that annuli could not be defined (32 specimens); 7 specimens were unreadable because of the existence of multiple bands, which made the identification of annuli difficult or resulted in aging discrepancies between readings and readers; and 19 speci mens were unreadable because of both the above factors.
For all the 769 swordfish with LJFL measured, indi viduals ranged from 83.4 to 290 cm for females and 78 to 206.6 cm for males (Fig. 5A). The proportion of the females (Fig. 5B) varied at sizes less than 195 cm, then increased to 100% at 210 cm and thereafter.
The relationship between LJFL and round weight for 227 specimens (sexes pooled) is shown in Figure 6; AN COVA revealed no differences between males and females (P>0.05). The LJFL-EFL (eye fork length) conversion equation is LJFL = 1.0647EFL + 7.7911, with df = 563 and r2=0.99.
Figure 6
Relationship between round weight and lower jaw fork length for the swordfish collected from Shinkang fish market regardless of sex.
RW = 1.3528x10-6 LJFL3.4297 r 2 = 0.9664 n = 227 0 20 40 60 80 100 120 140 160 180 0 0 100 150 200 250
Lower jaw fork length (cm)
Round weight (kg)
5
The monthly means of marginal increment ratio, MIR, for females with all ages combined, dropped drastically from the maximum of 0.86 in June to the minimum of 0.32 in July and August (Fig. 7). Similarly, the monthly means of MIR value for males declined sharply from the maximum of 0.65 in June to 0.37 and 0.35 in July and August. For both females and males, the monthly means of MIR dur ing the period from September to March were not differ ent (ANOVA, PU=0.95, P-=0.48), but the monthly means of
MIR in April, May, and June were significantly higher than that in July or August, respectively (two sample t-tests, P<0.001). Also, the mean MIR in August was significantly lower than that in September (t-tests, PU<0.001, P-<0.01).
The trends exhibited by monthly means of MIR for females and males for ages 2 to 5, respectively, were the same as those just described for all ages combined (Fig. 8). These patterns indicated the formation of one ring per year dur ing the period from July to August. The MIR analysis by age was not performed for age 1 and ages greater than 5 because the formula used to determine MIR does not apply to samples less than or equal to age 1 and there was a lack of a sufficient number of samples for ages greater than 5.
The mean band radii, by band group for female and male, are shown in Table 1. The observed LJFLs of female and male swordfish were plotted against their corresponding ray radii for method I and method II, respectively (Fig. 9). The relationships between LJFL and ray radius are de-scribed as follows: Method I Female: LJFL = 21.137S + 65.091 [r2=0.8894, n=334]; Male: LJFL = 19.966S + 68.160 [r2=0.8737, n=293]. Figure 7
Monthly means of marginal increment ratio of female and male swordfish in the waters around Taiwan for all ages combined. Vertical bars are 1 SE, numbers on the top of vertical bars are sample sizes.
0 0.2 0.4 0.6 0.8 1 1.2 1.4 J M A M J J A S O N D Female 0 0.2 0.4 0.6 0.8 1 1.2 J M A M J J A S O N D Month Male 33 9 12 36 32 19 19 21 14 14 14 22 27 8 5 40 16 20 21 23 13 7 14 14
Mean marginal increment ratio
F F Method II Female: LJFL = 73.754S0.5193 [r2 = 0.8753, n=334]; Male: LJFL = 77.075S0.4772 [r2=0.8742, n=293].
ANCOVA revealed significant differences in the rela tionship between the males and females for both methods (for method I, F333, 292=8.36, P<0.001; for method II, F333,
292 =15.59, P=0.004). The average back-calculated
lengths-at-age obtained using method I and method II are shown in Table 2. Growth rates were higher during the first year of age (mean 95.2 cm and 96.1 cm LJFL for males and females, respectively, for method I, compared to 88.6 cm and 90.4 cm LJFL for males and females, respectively, for method II). After the first year of age, the growth rates of both sexes slowed appreciably. Growth rates of females were always higher than those of males, especially after the age of three. Also, the growth rates were always higher for method II than for method I except for the first year of age. Fitted standard VB and generalized VB growth
Figure 8
Monthly means of marginal increment ratio of female and male swordfish in the waters around Taiwan for ages 2 to 5 respectively. Vertical bars are 1 SE, numbers on the top of vertical bars are sample sizes.
Female Male Age 2 Age 3 Age 4 Age 5 0 0.2 0.4 0.6 0.8 1 J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D 0 0.2 0.4 0.6 0.8 1 1.2 0 0.5 1 1.5 2 0 0.5 1 1.5 2 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 0 0.5 1 1.5 0 0.5 1 1.5 Month 8 1 9 7 9 2 5 12 8 5 7 7 7 8 4 4 1 6 5 7 5 5 6 3 4 1 1 2 2 2 2 1 2 2 1 7 5 3 5 1 10 4 4 1 1 1 8 6 11 4 1 1 7 9 1 2 3 3 2 2 7 1 1 2 2 11 4 4 3 6 1 1 5 5 4 4 4 2 2
Mean marginal increment ratio
curves for males and females with method I and method II back-calculation are shown in Figure 9, and the esti mated parameters corresponding to each curve are shown in Table 3.
Hotelling’s T2 test results showed significant difference
in growth parameters between female and male swordfish for either standard or generalized VB with either method I or method II back-calculation (Table 4). The calculated T2
is considerably higher than the tabulated value in Table 4 for each case, and all the parameters, except for m of the generalized VB, significantly affect the differences in growth between male and female swordfish (The Roy-Bose simultaneous confidence intervals around differences be-tween parameter values fail to include zero). The results of goodness-of-fit comparison showed that the generalized VB had larger r2 ranks for method II, but had equal ranks
with the standard VB for method I. Considering the tie rank groups in each sex-by-method, Friedman’s χ2
r sta
tistics is 2 (n=4). This was not significant at the 5% level (i.e. χ2=3.841, df=1), which indicated no significant differ
ence in the r2 rank ordering of the values between the two
growth functions. Kendall’s coefficient of concordance was 0.5 (n=4), which did not indicate a good agreement in the r2 ranks for all sexes-by-method.
Discussion
Just as reported by Berkeley and Houde (1983), Radtke and Hurley (1983), Wilson and Dean (1983), Tsimenides and Tserpes (1989), Ehrhardt (1992), Tserpes and Tsi menides (1995), and Stone and Porter (1997), females in our study were typically larger than males although the length-weight relationship between the sexes did not differ significantly. The overall sex ratio for the sampling period in our study did not deviate significantly from 1: 1 (P<0.01) but differed substantially from the ratios of 2.3 females to 1 male and 2.7 females to 1 male reported respectively by Stone and Porter (1997) and Caton et al. (1998). This discrepancy may have been caused by the
dif-Table 1
Mean radius from focus to distal edge of each band of female (A) and male (B) swordfish in the waters around Taiwan. Roman numerals indicate the number of bands.
A
Mean radius (mm) from focus to each bandBand
class size I II III IV V VI VII VIII IX XII
0 1 1.44 2 1.44 2.38 3 42 1.44 4 39 1.51 3.52 5 26 1.43 3.84 4.23 6 20 1.51 3.77 4.24 4.75 7 13 1.43 3.90 4.40 4.80 8 13 1.58 4.10 4.52 4.95 5.84 9 7 1.60 3.81 4.35 4.86 5.80 10 2 1.49 3.96 4.46 4.84 5.64 6.66 11 3 1.55 3.81 4.18 4.67 5.76 6.65 6.80 12 5 1.50 3.88 4.41 5.02 5.94 6.69 6.98 7.24 Mean 1.49 3.84 4.35 4.84 5.80 6.66 6.89 7.24 SD 0.18 0.40 0.40 0.49 0.42 0.46 0.57 0.40 Growth increase 0.91 0.60 0.50 0.49 0.47 0.41 0.22 0.35
B
Mean radius (mm) from focus to each bandBand
class I II III IV V VI VII VIII IX X
0 1 1.47 2 1.32 2.16 3 32 1.28 4 43 1.31 3.51 5 27 1.32 3.65 4.05 6 22 1.32 3.71 4.11 4.47 7 9 1.39 3.75 4.28 4.62 8 8 1.33 3.82 4.26 4.66 5.37 9 4 1.25 3.92 4.29 4.70 5.37 10 5 1.40 3.93 4.41 4.80 5.47 6.06 Mean 1.34 3.76 4.23 4.65 5.40 6.06 SD 0.17 0.30 0.31 0.33 0.35 0.38 Growth increase 1.01 0.58 0.48 0.41 0.36 0.20 Sample XI X 2 86 76 3.09 2.43 3.08 2.34 3.26 2.39 3.14 2.35 3.33 2.40 5.20 3.47 2.60 5.39 3.18 2.36 5.39 6.26 3.35 2.37 5.33 6.21 3.23 2.43 5.23 6.25 3.30 2.44 5.39 6.29 3.24 2.41 5.32 6.25 0.34 0.23 0.40 0.33 0.84 0.48 0.46 Sample size 2 84 57 3.06 2.36 3.04 2.32 3.11 2.32 3.21 2.44 3.21 2.41 4.98 3.24 2.52 5.02 3.35 2.25 5.05 5.74 3.16 2.36 5.11 5.76 3.17 2.35 5.04 5.75 0.26 0.22 0.32 0.39 0.82 0.39 0.35
ference in size ranges of LJFL sampled for studying the sex ratio (Mejuto, et al., 1995). Most of the LJFL in our sample ranged between 100 and 185 cm, close to Arocha and Lee’s (1995) middle size range within which the sex ratio was also almost 1:1 (Arocha and Lee, 1995). Besides the size range difference, the differences in geographical areas and seasons can also affect the sex ratio (Hoey, 1991; Mejuto et al., 1991). The proportion of females in our study, which increased to 100% at 210 cm and thereafter, was similar to those described by Turner et al. (1996), Stone and Porter (1997), and DeMartini et al. (2000).
Several genetic studies (Grijalva-Chon et al., 1994; Rosel and Block, 1996; Chow et al., 1997; Chow, 1998) have been unable to reject the hypothesis that swordfish comprise a single, homogenous population in the Pa cific. However, from recent analyses of mtDNA, Reeb et al. (2000) concluded that swordfish are not homogenous in the Pacific. They found significantly different northern and southern populations in the western Pacific and sev eral overlapping swordfish populations may occur in the eastern Pacific, making swordfish genetically continuous there. Gene flow between the populations occurs through
--Figure 9
Relationship between LJFL and anal ray radius for female and male swordfish in the waters around Taiwan.
n = 334 Method I LJFL = 21.137S + 65.091, r 2 = 0.8894 — Method II LJFL = 73.754 S0.5193 , r 2 = 0.8753 0 50 100 150 200 250 300 0 6 10 n = 293 Method I LJFL = 19.966S + 68.160, r 2 = 0.8737 — Method II LJFL = 77.075 S0.4772 , r 2 = 0.8742 0 50 100 150 200 250 0 3 8 10 Female Male Lo w er ja w f o rk length (cm) Spine radius (mm) 5 4 3 2 1 7 8 9 2 1 4 5 6 7 9
a horseshoe-shaped corridor, running between the north-western Pacific, across to the eastern Pacific and back to the southwestern Pacific (Ward and Elscot, 2000). Accord ing to Reed’s studies, the swordfish in our samples can be considered a part of the northern population in the west-ern Pacific Ocean.
We found that anal-fin rays are useful for aging swordfish; they are easily sampled without reducing the economic value of the fish and can be read easily (the growth rings stand out clearly). This aging tool is espe cially important because swordfish lack scales and their very small otoliths are not amenable to traditional aging techniques (Ovchinnikov, 1971; Beckett, 1974; Tserpes and Tsimenides, 1995). Moreover, fin rays can be easily stored for future reexamination (Compeán-Jimenez and Bard, 1983). One problem associated with the fin-ray method used in our study, also indicated by Berkeley and Houde (1983) and Tserpes and Tsimenides (1995), was the pos sible existence of multiple bands and the missing first annulus in larger fish. However, González-Garcés and Fariña-Perez (1983) and Tserpes and Tsimenides (1995) noted that experienced readers could overcome the
prob-Table 2
Mean back-calculated lower jaw fork lengths at age for swordfish in the waters around Taiwan.
Back-calculated length (cm)
Method I Method II
Age (yr) Male Female Male Female
1 95.19 96.07 88.60 90.35 2 115.70 114.96 116.18 3 133.41 133.80 136.47 4 146.30 145.22 150.35 5 157.99 154.40 162.91 6 169.95 161.36 175.34 7 180.48 167.84 186.40 8 190.14 176.87 195.83 9 198.40 185.23 204.62 10 207.79 191.58 214.15 11 215.06 220.58 12 222.15 226.62 114.86 131.88 143.12 152.14 159.15 165.60 174.76 184.00 190.89
Table 4
Results of the multivariate (Hotelling’s T2) tests for difference between the estimated von Bertalanffy growth parameters of female and male swordfish in the waters around Taiwan.
Standard von Bertalanffy Generalized von Bertalanffy
growth model growth model
Method I Method II Method I Method II
T2 20715.1 1614630 6782.34 df 3, 623 3, 623 4, 623 4,623 T 2 0.01,df 11.48 13.46 13.46 99% CI for1 L∞U−L∞- 53.842 ~ 61.436** 56.452 ~ 63.390** 68.845 ~ 71.365** 85.621 ~ 89.587** kU−k- –0.044 ~ –0.034** –0.071 ~ –0.065** KU−K- –0.008 ~ –0.004** –0.050 ~ –0.042** t0U–t0- –0.216 ~ –0.014** –0.406 ~ –0.288** –0.542 ~ –0.418** –0.163 ~ –0.085** mU–m- 0.205 ~ 0.227** –0.062 ~ 0.028
1 Roy-Bose simultaneous confidence intervals around differences between parameter values.
** Indicates the parameter tested that significantly affects differences in growth between the females and the males at a significant level of 0.01.
56005.2
11.48 Table 3
Parameter estimates for the standard von Bertalanffy and the generalized von Bertalanffy growth models for swordfish in the waters around Taiwan. Numbers in parentheses are standard errors.
Standard von Bertalanffy Generalized von Bertalanffy
growth model growth model
Method I Method II Method I Method II
Parameter Male Female Male Female Male Female Male Female
L∞ 224.170 281.809 207.520 267.441 231.772 301.877 213.052 300.656 (12.802) (6.805) (8.465) (6.517) (26.937) (11.068) (19.153) (38.869) k 0.140 0.101 0.198 0.130 (0.025) (0.006) (0.031) (0.009) K 0.066 0.060 0.086 0.040 (0.103) (0.055) (0.035) (0.116) t0 –3.089 –3.204 –1.955 –2.302 –1.556 –2.036 –0.626 –0.750 (0.523) (0.171) (0.406) (0.190) (2.942) (1.263) (1.196) (2.272) m –0.625 –0.409 –0.768 –0.785 (2.388) (0.943) (1.730) (1.324)
lem of multiple bands by determining whether the bands were continuous around the circumference of the entire ray section and by judging their distance from preceding and following bands. In our study, 91 out of the 627 read-able specimens had “multiple bands” that were read with-out a problem by using these criteria. Of the 26 discarded specimens that had multiple unreadable bands, most were found in swordfish larger than 200 cm in size. The missing first band in larger fish can be estimated from observa tions of its position on young specimens where the first band is visible. Similar approaches for solving the problem
of missing band have also been used for Atlantic swordfish (Berkeley and Houde, 1983) and eastern Mediterranean swordfish (Tserpes and Tsimenides, 1995).
Results of marginal increment ratio analysis (Figs. 7 and 8) suggest one growth ring (annulus) is formed each year from July to August, which is toward the end of the spawning period for the swordfish in the north Pacific (Yabe et al., 1959). Although the timing of annulus deposi tion coincides with the swordfish’s spawning season in the north Pacific, it may also be related to swordfish migration, as suggested by Berkeley and Houde (1983) for Atlantic
swordfish and by Tserpe and Tsimenides (1995) for east-ern Mediterranean swordfish. The relationship between annulus formation and migration for the western Pacific swordfish should be investigated. Others (Nelson and Ma nooch, 1982; Beckman et al., 1990; Sturm and Salter, 1990; Ferreira and Russ, 1994; Franks et al., 1999) have com mented on the physiological nature of annulus formation and the importance of environmental factors, suggesting that reproduction may not be the sole determining factor in annulus deposition. Our results only partially validated age. Validation of ages requires either a mark-recapture study or the identification of known-age fish in the popu lation (Beamish and McFarlane, 1983; Prince et al., 1995; Tserpes and Tsimenides, 1995; Sun et al., 2001).
According to Hotelling’s T2 analysis, the female and male
swordfish in either method I or method II of both standard VB and generalized VB grew differently. Those of the gen eralized VB had a little larger r2 ranks than those of the
standard VB, although the goodness-of-fit of nonparamet ric analysis of r2 ranks did not show a significant difference
between two growth equations (P=0.22). In addition, the generalized VB appeared to fit the data well over the range of ages and it provided more realistic growth patterns for juveniles less than one year. However, the standard VB, commonly used to describe fish asymptotic growth, did not fit these data well, and generated grossly overestimated val ues for individuals less than one year (Table 3 and Fig. 10) (Ehrhardt, 1992; Ehrhardt et al., 1996).
In Table 3, the t0 values estimated for the generalized VB with method II (i.e. a power function was used to describe the relationship between ray radius and LJFL) were much closer to zero than those estimated for the generalized VB with method I (i.e. a simple linear function was used to describe the relationship between ray radius and LJFL). Also, Ehrhardt (1992), Ehrhardt et al. (1996), and Tserpes and Tsimenides (1995) favored the power function (method II) because they believed its description to be more biologi cally realistic. Therefore, the parameter estimates for the generalized VB model with method II shown in Table 3 are recommended as the most acceptable for determining the age composition of swordfish in the waters around Taiwan.
Age-length relationships of swordfish (Fig. 11) are mostly based on Atlantic specimens (Berkeley and Houde, 1983; Radtke and Hurley, 1983; Wilson and Dean, 1983; Ehrhardt, 1992; Ehrhardt et al., 1996), and a few on Pacific samples (Yabe et al., 1959; Castro-Longoria and Sosa-Nishizaki, 1998; Uchiyama et al., 1998; Castro-Longoria1; Uchiyama2). Differences in estimates of growth
parameters for males and females arise from the use of different hard parts, e.g. anal-fin rays (our study) versus otoliths (Radtke and Hurley, 1983; Wilson and Dean, 1983) or vertebrae (Esteves et al., 1995), artifacts of preparation, or interpretation. Even though we used the same method as Ehrhardt (1992), Tserpes and Tsimenides (1995), and Ehrhardt et al. (1996), observed difference could be re lated to geographical coverage of the studies. Our results were in the mid-range of previous estimates, well within the range of variation that might be expected due to the somewhat subjective nature of the processing, measuring, and interpreting of growth rings on fin rays. Therefore, we
Figure 10
Standard and generalized von Bertalanffy growth curves for female and male swordfish in the waters around Taiwan. 0 50 100 150 200 250 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 Standard VB - Method I Standard VB - Method II Generalized VB - Method I Generalized VB - Method II 0 50 100 150 200 250 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 Age (year) Standard VB - Method I Standard VB - Method II Generalized VB - Method I Generalized VB - Method II Female Male
Lower jaw fork length (cm)
believe our growth parameter estimates are appropriate for use in assessment studies of the northern swordfish population in the western Pacific Ocean.
Acknowledgments
The authors express sincere gratitude to Nancy Lo and David Au, Southwest Fisheries Science Center of the
1 Castro-Longoria, R. 2000. Personal communication of the length-at-age estimates mentioned in Castro-Longoria and Sosa-Nishizaki, 1998. Departmento de Investigaciones Cientí ficas y Tecnológicas de la Universidad de Sonora, Rosales y Niños Héroes S/N, Hermosillo, Sonora, 83000 Mexico.
2 Uchiyama, J. H. 2000. Personal communication of the length-at-age estimates mentioned in Uchiyama et al., 1998. Hono lulu Laboratory, Southwest Fisheries Service Center, NOAA, 2570 Dole Street, Honolulu, Hawaii 96822-2396.
Figure 11
A comparison of the growth curves of female and male swordfish estimated by different authors.
0 50 100 150 200 250 300 1 2 3 4 5 6 7 8 9 10 11 12 Yabe et al. (1959) -sex combined Berkeley & Houde (1983) Wilson & Dean (1983) Radtke & Hurley (1983) Tsimenides & Tserpes (1989) Ehrhardt (1992) Tserpes & Tsimenides(1995) Ehrhardt et al. (1996) Uchiyama et al. (2000) Castro-Longoria (2000) Present study 0 50 100 150 200 250 300 1 2 3 4 5 6 7 8 9 10 11 12 Age (year) Female Male
Lower jaw fork length (cm)
National Marine Fisheries Services, NOAA, for their comments on the initial draft of this paper. Our many thanks are also given to four anonymous referees for their valuable comments. This study was in part supported financially by the “Fisheries Administration, Council of Agriculture, Taiwan,” through grant 88-AST-1.4-FID-04 (04) to Chi-Lu Sun.
Literature cited
Arocha, F., and D. Lee.
1995. The spawning of swordfish from the Northwest Atlantic. ICCAT (International Commission for the Con servation of Tunas) Coll. Vol. Sci. Pap. 44(3):179–186. Bartoo, N. W., and A. L.Coan.
1989. An assessment of the Pacific swordfish resource. In Proceedings of the second international billfish sym posium; Hawaii, August 1−5, 1988 (R. H. Stroud, ed.), p. 137–151. National Coalition for Marine Conservation, Inc., Savannah, GE.
Beamish, R. J., and D. A. Fournier.
1981. A method for comparing the precision of a set of age determinations. Can. J. Fish. Aquat. Sci. 38:982–983.
Beamish, R. J., and G. A. McFarlane.
1983. Validation of age determination estimates: the forgot-ten requirement. U.S. Dep. Commer., NOAA Tech. Rep. NMFS 8:29–33.
Beckett, J. S.
1974. Biology of swordfish, Xiphias gladius L., in the North-west Atlantic Ocean. U. S. Dep. Commer., NOAA Tech. Rep. NMFS SSRF-675:103–106.
Beckman, D. W., A. L. Stanley, J. H. Render, and C. A. Wilson, 1990. Age and growth of black drum in Louisiana waters of
the Gulf of Mexico. Trans. Am. Fish. Soc. 119:537–544. Berkeley, S. A., and E. D. Houde.
1983. Age determination of broadbill swordfish, Xiphias gladius, from the Straits of Florida, using anal fin spine sections. U. S. Dep. Commer., NOAA Tech. Rep. NMFS 8: 137–143.
Bernard, D. R.
1981. Multivariate analysis as a means of comparing growth in fish. Can. J. Fish. Aquat. Sci. 38:233–236. Brothers, E. B.
1983. Summary of round table discussions on age validation. U. S. Dep. Commer., NOAA Tech. Rep. NMFS 8:35–44.
Campana, S. E.
deter-mination, including a review of the use and abuse of age validation methods. J. Fish. Biol. 59:197–242.
Campana, S. E., M. C. Annand, and J. I. McMillan.
1995. Graphical and statistical methods for determining the consistency of age determinations. Tran. Am. Fish. Soc. 124:131–138.
Castro-Longoria, R., and O. Sosa-Nishizaki.
1998. Age determination of swordfish, Xiphias gladius L., from waters off Baja California, Mexico, using anal fin rays and otoliths. U. S. Dep. Commer., NOAA Tech. Rep. NMFS 142:231–257.
Caton, A., K. Colgan, P. Sahlqvist, P. Ward, C. Ramirez, and M. Scott.
1998. Swordfish, Xiphias gladius, and the fisheries for tunas and billfishes in the Australian fishing zone. U. S. Dep. Commer., NOAA Tech. Rep. NMFS 142:11–35. Chen, Y., D. A. Jackson, and H. H. Harvey.
1992. A comparison of von Bertalanffy and polynomial func tions in modelling fish growth data. Can. J. Fish. Aquat. Sci. 49:1228–1235.
Chow, S.
1998. Genetic comparison of Pacific and Mediterranean swordfish, Xiphias gladius, by RFLP analysis of the mito chondrial D-loop region. U. S. Dep. Commer., NOAA Tech. Rep. NMFS 142:239–244.
Chow, S., H. Okamoto, Y. Uozumi, Y. Takeuchi, and H. Takeyama. 1997. Genetic stock structure of the swordfish (Xiphias gla dius) inferred by PCR-RFLP analysis of the mitochondrial DNA control region. Mar. Biol. 127:359–367.
Compeán-Jimenez, G., and F. X. Bard.
1983. Growth increments on dorsal spines of eastern Atlan tic bluefin tuna, Thunnus thynnus, and their possible rela tion to migration patterns. U. S. Dep. Commer., NOAA Tech. Rep. NMFS 8:77–86.
DeMartini, E. E., J. H. Uchiyama, and H. A. Williams.
2000. Sexual maturity, sex ratio, and size composition of swordfish, Xiphias gladius, caught by the Hawaii-based pelagic longline fishery. Fish. Bull. 98:489–506.
Ehrhardt, N. M.
1992. Age and growth of swordfish, Xiphias gladius, in the northwestern Atlantic. Bull. Mar. Sci. 50(2): 292−301. Ehrhardt, N. M., R. J. Robbins, and F. Arocha.
1996. Age validation and growth of swordfish, Xiphias gladius, in the northwest Atlantic. ICCAT (International Commission for the Conservation of Tunas) Col. Vol. Sci. Pap. 45(2):358–367.
Esteves, E., P. Simões, H. M. Da Silva, and J. P. Andrade. 1995. Ageing of swordfish, Xiphias gladius Linnaeus,
1758, from the Azores, using sagittae, anal-fin spine and vertebrae. Bull. Univ. Azores, Life and Marine Sciences 13A:39–51.
Ferreira, B. P., and G. R. Russ.
1994. Age validation and estimation of growth rates of the coral trout, Plectropomus leopardus (Lacepede 1802), from Lizard Island, Northern Great Barrier Reef. Fish Bull. 92:46–57.
Franks, J. S., J. R. Warren, and M. V. Buchanan.
1999. Age and growth of cobia, Rachycentron canadum, from the northeastern Gulf of Mexico. Fish. Bull. 97:459–471. Fraser, C. Mcl.
1916. Growth of the spring salmon. Trans. Pacific Fish. Soc. 1915, p. 29–39.
Friedman, M.
1937. The use of ranks to avoid the assumption of normality implicit in the analysis of variance. J. Am. Stat. Assoc. 32: 675–701.
1940. A comparison of alternate tests of significance for the problems of m rankings. Ann. Math. Stat. 11: 86–92. González-Garcés, A., and A.C. Fariña-Perez.
1983. Determining age of young albacore, Thunnus alalunga, using dorsal spines. U. S. Dep. Commer., NOAA Tech. Rep. NMFS 8:117–122.
Grijalva-Chon, J.M., K. Numachi, O. Sosa-Nishizaki, and J. de la Rosa-Velez.
1994. Mitochondrial DNA analysis of North Pacific sword-fish (Xiphias gladius) population structure. Mar. Ecol. Prog. Ser. 115:15–19.
Hoey, J.
1991. Sex ratio data for western North Atlantic swordfish. ICCAT Coll. Vol. Sci. Pap. 34(2):429–436.
Kendall, M. G.
1962. Rank correlation methods, 3rd ed, 199 p. Charles Griffin, Condon.
Media Cybernetics.
1997. Image-Pro Plus, version 3.0 for Windows, 480 p. Media Cybernetics, Silver Spring, MD.
Mejuto, J., B. Garcia, and M. Quintans.
1991. A preliminary analysis of the sex-ratio of the sword-fish (Xiphias gladius) in the north Atlantic by size class using space-time strata. ICCAT Coll. Vol. Sci. Pap. 35(2): 473–481.
Mejuto, J., J. M. de la Serna, and B. Garcia.
1995. An overview of the sex-ratio at size of the swordfish (Xiphias gladius L.) around the world: similarity between different strata. ICCAT Coll. Vol. Sci. Pap. 44(3):197–205. Nelson, R. S., and C. S. Manooch.
1982. Growth and mortality of red snappers in the west-cen tral Atlantic Ocean and northern Gulf of Mexico. Trans. Am. Fish. Soc. 111:465–475.
Ovchinnikov, V. V.
1971. Swordfishes and billfishes in the Atlantic Ocean, ecol ogy and functional morphology, 77 p. Israel Program for Scientific Translations. U.S. Dep. Commer., NOAA NMFS TT 71-50011.
Powers, J. E.
1983. Some statistical characteristics of ageing data and their ramifications in population analysis of oceanic pe lagic fishes. U. S. Dep. Commer., NOAA Tech. Rep. NMFS 8:19–24.
Prince, E. D., D. W. Lee, and S. A. Berkeley.
1988. Use of marginal increment analysis to validate the anal spine method for ageing Atlantic swordfish and other alternatives for age determination. ICCAT Coll. Vol. Sci. Pap. 27:194–201.
Prince, E. D., D. W. Lee, J. L. Cort, G. A. McFarlane, and A.Wild. 1995. Age validation evidence for two tag-recaptured Atlan
tic albacore, Thunnus alalunga, based on dorsal, anal, and pectoral finrays, vertebrae, and otoliths. In Recent devel opments in fish otolith research (D. H. Secor, J. M. Dean, and S. E. Campana (eds.), p. 375–396. Univ. of South Carolina Press, Columbia, SC.
Radtke, R. L., and P. C. F. Hurley.
1983. Age estimation and growth of broadbill swordfish, Xiphias gladius, from the northwest Atlantic based on external features of otoliths. U. S. Dep. Commer., NOAA Tech. Rep. NMFS 8:145–150.
Reeb, C. A., L. Arcangeli, and B. A. Block.
2000. Structure and migration corridors in Pacific popu lations of the swordfish Xiphias gladius, as inferred through analyses of mitochondrial DNA. Mar. Biol. 136: 123–1131.
Richards, F. J.
1959. A flexible growth function for empirical use. J. Exp. Bot. 10:290–300.
Rosel, P. E., and B. A. Block.
1996. Mitochondrial control region variability and global population structure in the swordfish, Xiphias gladius. Mar. Biol. 125:11–22.
SAS Institute.
1990. SAS/STAT user’s guide, version 6, fourth ed., p. 1135– 1194. SAS Institute Inc., Cary, NC.
Sakagawa, G. T.
1989. Trends in fisheries for swordfish in the Pacific Ocean. In Proceedings of the second international billfish sympo sium; Hawaii, August 15, 1988 (R. H. Stroud, ed.), p. 61–79. National Coalition for Marine Conservation, Inc., Savan nah, GE.
Stone, H. H., and J. M. Porter.
1997. Development of a swordfish sex-ratio-at-size relation-ship for catches from the Canadian fishery. ICCAT Col. Vol. Sci. Pap. 46(3):311–314.
Sturm, M. G. de L., and P. Salter.
1990. Age, growth, and reproduction of the king mackerel, Scomberomorus cavalla (Cuvier) in Trinidad waters. Fish. Bull. 88:361–370.
Sun, C. L., C. L. Huang, and S. Z. Yeh.
2001. Age and growth of the bigeye tuna Thunnus obesus in the western Pacific Ocean. Fish. Bull. 99: 502–509. Tserpes, G., and N. Tsimenides.
1995. Determination of age and growth of swordfish, Xiph
ias gladius L., 1758, in the eastern Mediterranean using anal-fin spines. Fish. Bull. 93:594–602.
Tsimenides, N., and G. Tserpes.
1989. Age determination and growth of swordfish, Xiphias gladius L., 1758, in the Aegean Sea. Fish. Res. 8:159– 168.
Turner, S. C., P. Arocha, and G. P. Score.
1996. U.S. swordfish catch at age by sex. ICCAT Col. Vol. Sci. Pap. 45(2):373–378.
Uchiyama, J. H., R. A. Skillman, J.D. Sampaga, and E. E DeMartini.
1998. A preliminary assessment of the use of hard parts to age central Pacific swordfish, Xiphias gladius. U. S. Dep. Commer., NOAA Tech. Rep. NMFS 142:261–273.
Ward, P., and S. Elscot.
2000. Broadbill swordfish: status of world fisheries, 208 p. Bureau of Rural Sciences, Canberra.
Wilson, C. A., and J. M. Dean.
1983. The potential use of sagittae for estimating age of Atlantic swordfish, Xiphias gladius. U. S. Dep. Commer., NOAA Tech. Rep. NMFS 8:151–156.
Yabe, H., S. Ueyanagi, S. Kikawa, and H. Watanabe.
1959. Study on the life-history of the swordfish, Xiphias gla dius Linnaeus. Rep. Nankai Reg. Fish. Res. Lab. 10:107– 150
Zar, J. H.
1999. Biostatistical analysis. 4th ed., 929 p. Prentice-Hall Inc., Englewood Cliffs, NJ.
