CATCH, FISHING EFFORT AND CATCH PER UNIT EFFORT OF TAIWANESE LONGLINE FISHERY FOR BLUE MARLIN AND WHITE MARLIN IN THE ATLANTIC

(1)

Col. Vol. Sci. Pap. ICCAT, 53: 307-317. (2001)

SCRS/2000/078

CATCH, FISHING EFFORT AND CATCH PER UNIT EFFORT

OF TAIWANESE LONGLINE FISHERY FOR BLUE MARLIN

AND WHITE MARLIN IN THE ATLANTIC

Chien-Chung Hsu

1

ABSTRACT

Catch, fishing effort and catch per unit effort of white marlin

(Tetrapturus albidus)

and blue

mar-lin

(Makaira nigricans)

are verified and updated to 1998 for Taiwanese longline fishery. The nominal

CPUE of white marlin and blue marlin were standardized by general linear model with Year,

Quarter and Area factors, and with Quarter:Area interactions for two time periods (1968-1980

and 1981-1998). These analyses were done for the total Atlantic and separately using south and

north Atlantic CPUE data. The results show similar decreasing tendencies in standardized CPUE

from 1968 to 1992 for white marlin and from 1968 to 1989 for blue marlin, which then increase to

the level of the early 1970s and become stable from 1990 onwards.

RESUMEN

Se comprueban y actualizan hasta 1998 los datos de captura, esfuerzo de pesca y captura por

unidad de esfuerzo de aguja blanca

(Tetrapturus albidus)

y aguja azul

(Makaira nigricans)

correspondientes a la pesquería palangrera de Taipei Chino. La CPUE nominal de la aguja blanca

y la aguja azul se estandarizaron por medio del modelo lineal generalizado, con los factores Año,

Trimestre y Zona e interacciones Zona/Trimestre durante dos periodos (1968-1980 y 1981-1998).

Estos análisis se llevaron a cabo para el conjunto del Atlántico y por separado, usando los datos

de CPUE del Atlántico norte y Atlántico sur. Los resultados muestran tendencias descendentes

similares en la CPUE estandarizada para la aguja blanca, de 1968 a 1992 y para la aguja azul,

de 1968 a 1998, que posteriormente se elevan hasta el nivel de principios de la década de los 70

y se estabilizan a partir de 1990.

RÉSUMÉ

Les données de capture, d’effort de pêche et de capture par unité d’effort concernant le makaire

blanc

(Tetrapturus albidus)

et le makaire bleu

(Makaira nigricans)

ont été vérifiées et actualisées

jusqu’en 1998 pour ce qui est de la pêche palangrière du Taiwan. Les CPUE nominales du makaire

blanc et du makaire bleu ont été standardisées par un modèle linéaire généralisé assorti des

facteurs Année, Trimestre et Zone, et des interactions Trimestre:Zone pour deux périodes temporelles

(1968-1980 et 1981-1998). Ces analyses ont été effectuées pour l’ensemble de l’Atlantique et

séparément à l’aide des données de CPUE de l’Atlantique nord et sud. Les résultats indiquent des

tendances à la baisse similaires de la CPUE standardisée de 1968 à 1992 pour le makaire blanc,

et de 1968 à 1989 pour le makaire bleu, qui augmentent ensuite pour atteindre le niveau du début

des années 1970 et se stabilisent à partir de 1990.

KEYWORDS

Tuna fisheries, High seas fisheries, longlining, By catch, Logbooks, Fish catch statistics, Catch/

effort, Multivariate analysis, Abundance

(2)

INTRODUCTION

Taiwan is one of the major nations using longline gear fishing for tunas and tuna-like species in the

Atlantic. In addition, tunas such as albacore, bigeye tuna, and yellowfin tuna are the target species, and

the billfishes - swordfish, white marlin, blue marlin and sailfish - are caught incidentally by Taiwanese

fishermen.

The Taiwanese longline fishery initiated in 1913 and operated in the waters off Taiwan, and extended

into the Atlantic about 1962. The dominant species for the Taiwanese longline fishery in the Atlantic

were albacore from the 1970s and then bigeye tuna starting in the early 1990s. This indicates that a

transformation in the fishing pattern occurred at the end of the 1980’s (Hsu and Chern 1996). This

transformation resulted in increasing catch of tropical tunas along with an increase of swordfish and

billfish catches (in particular, white marlin and blue marlin). The causes of the increased billfish catches

were not verified. However, the amounts harvested annually are significant among longline fishing

na-tions in the Atlantic. Therefore, Taiwanese fishing activities including number of vessels, fishing

pat-terns and fishing grounds, catch, effort, catch per unit effort are important for billfish species assessment.

Hence, a review of Taiwanese billfish catches in the Atlantic and the standardized catch per unit effort

have been reported previously (Hsu 1998) during the Third Billfish Workshop (Miami, FL, USA, July

11-20, 1996).

In this document, an update of standardized catch per unit effort series for blue and white marlin

based on the Taiwanese longline fishery operating in the Atlantic are provided using a general linear

model (GLM).

MATERIALS AND METHODS

Basic data

The landings for white marlin (Fig. 1) and blue marlin (Fig. 2) are compiled by similar procedures to

those of of Hsu (1998), from sources of Fisheries Yearbooks of the Taiwan Area (Anonymous,

1966a-1999a), Annual Catch Statistics of Taiwanese Tuna Longline Fishery (Anonymous, 1967b-1993b), and

the tuna longline logbooks which are submitted by captains of longline fleets and collected by the

De-partment of Fisheries, Kaohsiung Municipal Government. Logbooks are the main source used to

com-pile the catch. Catches of the two species, white marlin and blue marlin, are verified and provided by

Oversea Fisheries Development Council (OFDC) from 1994 to 1999; the 1999 catch is preliminary.

Detailed data compilation of Taiwanese distant-water longline fishery was described in Hsu and Lin

(1996), which includes raw data used, process of compilation and debugging.

Standardized abundance indices

In using the GLM as the standardization method, the single factors, year, quarter, and area were used

with a quarter*area interaction to adjust the nominal catch per unit effort of white marlin and blue

marlin. Thus, the GLM model used was:

log(CPUE+10% of grand mean) = u + Year effect + Quarter effect + Area effect + Quarter*Area

interaction +error,

where CPUE is in number of billfish caught per 1000 hooks, u is the overall mean, and the error term is

assumed to be normally distributed with mean 0 and standard deviation £m. The F-test was used on all

effects and interactions to determine the significance of the contribution of each factor to the model.

(3)

Area was classified according to fishing effort distributions (Fig. 3) and nominal catch per unit effort

of white marline (Fig. 4) and blue marlin (Fig. 5), indicating where the heavy fishing effort is employed,

and where billfish CPUE is highest. The goal of the area classification is to attempt to eliminate as much

noise as possible other than abundance itself. The final sub-area stratification is given in Fig. 6.

For the analysis of Taiwanese data, two time frames were defined according to the attributes of the

data. One is from 1968 to 1980, with 5x5 degree-square aggregated catch and fishing effort data. The

other is from 1981 to 1998, using daily logbook data. Quarter was assigned as the calendar quarter of the

year. Observations having daily fishing effort smaller than 3000 hooks were excluded from the analysis

following the reasoning of Uosaki (1998) and Uozumi (1998). However, observations with zero catch

were retained in the analysis.

RESULTS AND DISCUSSION

Catch, effort and nominal catch per unit effort

Fishing effort, catch, mean weight and nominal CPUE for white marlin and blue marlin are given in

Tables 1 and 3 respectively. White marlin (Tables 1 and 2), and blue marlin (Tables 3 and 4) were

analyzed for the total Atlantic and also separately as north and south Atlantic stocks (separated at 5

o

_N

latitude).

The catches of white marlin indicate that there are two high-catch periods from 1969 to 1976, and

from 1987 to 1995, which then decline from the historically-high level (Fig. 1). Those catches almost

come from the south Atlantic (Table 2) where conventional longline Taiwanese vessels operated to target

albacore (in particular since 1987). The increasing catch may relate to the albacore targeted by

Taiwan-ese longline fleets. A similar situation occurs for blue marlin (Fig. 2 and Table 4).

Abundance indices

Nominal and adjusted catch per unit effort in the total Atlantic are given in Tables 1 and 3 for white

marlin and blue marlin respectively, and also by sub-areas (Tables 2 and 4 for white marlin and blue

marlin respectively) from 1968 to 1998. The standardized CPUE trends with 95% confidence intervals

for 1968-1994 are given in Table 1 for white marlin and in Table 3 for blue marlin.

Adjusted catch per unit effort was standardized by the GLM method with year, quarter, and area

factors, and quarter*area interaction as ICCAT species working groups used in previous assessments.

The significance tests of standardized catch per unit effort on the time frame 1968-1980 are given in

Tables 5 and 7, and for 1981-1998 they are given in Tables 6 and 8 for white marlin and blue marlin,

respectively. Model fits to white marlin and blue marlin in the total Atlantic and blue marlin in the north

Atlantic were not significant. All effects and interactions were included in estimating yearly

standard-ized catch per unit effort. The models explained 37% and 48% of the variation in CPUE for white marlin

and blue marlin, respectively for the total Atlantic.

The adjusted catch per unit effort trend of white marlin for the Taiwanese longline fishery shows

great fluctuation from 1968 to 1994 for the total Atlantic and north Atlantic series. In contrast, the south

Atlantic series (Figure 7) showed a decreasing trend from 1968 to 1977, was relatively flat through 1986,

and then fluctuated through 1994. The blue marlin index showed a decreasing trend from 1968 to 1975,

was relatively flat from the late 1970s through the 1980s, and then increased through 1994 for the total

Atlantic series (Figure 8). Trends for the north and south series (Figures 9 and 10) were similar to the

total Atlantic results.

(4)

The Taiwanese distant-water tuna longline fishery catches billfishes incidentally in the Atlantic.

However, targeting is suspected for Taiwanese longline fleets operating joint venture in the waters off

Brazil during the period from 1990 to 1992. This may be one of the possible reasons for the increasing

billfish catch during the 1990s. There were between several and 20 or more Taiwanese conventional

longline vessels operating in the southeastern and northeastern waters off Brazil, especially in the

south-eastern waters during the period 1990 to 1992. Those vessels were joint ventures with Brazil, and targetted

mainly albacore. Whether or not targeting on either swordfish or billfishes is suspected, the amounts of

swordfish and billfish catches were highly significant. It is obvious that the billfish catches in Fisheries

Yearbooks, Taiwan Area, increased during the mentioned period, and it did in Brazil-Taiwan statistics

(Hsu 1998). However, to investigate the catches, the Taiwan Fishery Authority has regulated the catches

of white marlin and blue marlin under the level of 550 t annually to reflect their by-catch status.

The 1967 Task II data were not included during standardization of catch per unit effort series because

of the low coverage of 5x5 square blocks in Task II, which also pertained only to the last half of that year

(from July to December). Therefore, the adjusted catch per unit effort in 1967 may not represent the

abundance index appropriately. Thus, only the data from 1968 to 1998 were used for the standardization

purpose. Further, complete 1999 data were not available. However, the standardized 1998 index may be

carried over to 1999 if needed because the fishing pattern of the Taiwanese longline fishery in the

Atlan-tic was similar.

Finally, the standardized catch per unit effort of white marlin (Figs. 7 and 8; Tables 1 and 2) and blue

marlin (Figs. 9 and 10, Tables 3 and 4) were illustrated and tabulated. Those trends show similarly

decreasing tendency from 1968 to 1992 for white marlin and to 1989 for blue marlin, increasing to the

level of early 1970s and stable trend from 1990 afterward.

REFERENCES

ANONYMOUS. 1966a-1999a. Fisheries Yearbook, Taiwan Area. Taiwan Fisheries Bureau, Department of

Agricul-ture and Forestry, Provincial Government of Taiwan.

ANONYMOUS. 1967a-1993b. Annual catch statistics of Taiwanese tuna longline fishery. Institute of

Oceanogra-phy, National Taiwan University.

HSU, C.C. 1998. A review of catch, fishing effort and catch per unit effort of Taiwanese

longline fishery for billfishes

in the Atlantic.

Col. Vol. Sci. Pap. ICCAT, 47:267-278.

HSU, C.C. and Y.C. Chern. 1996. Recent status of Taiwanese longline fisheries in the Atlantic. Col. Vol. Sci. Pap.

45:62-67.

HSU, C.C. and M.C. Lin. 1996. The recent catch estimating procedures of Taiwanese longline fishery. Col. Vol. Sci.

Pap. ICCAT, 43:171-178.

H U N A G, H.C. 1974. Billfish fishery of Taiwan. Pages 332-335

in

R.S. Shomura and F.

W illiams

(ed.): Proceedings

of the International Billfish Symposium, Kailua-Kona, Hawaii, US, 9-12 August 1972, part 2. Review and

contributed papers, NOAA Tech. Rep. NMFS SSRF-675.

UOSAKI, K. 1998. The CPUE trend for Atlantic blue marlin caught by Japanese longline fishery. Col. Vol. Sci. pap.

ICCAT, 47:245-254.

UOZUMI, Y. 1998. Standardized of CPUE for white marlin caught by Japanese longline fishery in the Atlantic. Col.

Vol. Sci. Pap. 47:255-265.

(5)

Year Fishing effort (1000hoo ks) Catch (mt) Weight Mean (kg) Nominal CPUE (no./1000 hooks) Adjusted CPUE (no./1000 hooks) Lower CI of adjusted CPUE (no./1000 hooks) Upper CI of adjusted CPUE (no./1000 hooks) 1967 3678 181 30.64 0.0710 1968 56606 385 16.99 0.3142 0.16099 0.10211 0.24562 1969 72562 583 25.20 0.3188 0.18681 0.12629 0.27041 1970 70723 612 24.21 0.3020 0.11805 0.07652 0.17539 1971 102994 438 21.36 0.1754 0.12426 0.07984 0.18626 1972 98397 713 23.10 0.1117 0.10346 0.06324 0.16056 1973 108549 599 22.80 0.2418 0.12471 0.07888 0.18938 1974 98669 537 21.15 0.2572 0.15009 0.09902 0.22101 1975 90820 507 20.46 0.2728 0.11906 0.07484 0.18149 1976 110031 519 18.97 0.0713 0.01542 0.00144 0.03514 1977 142341 163 24.77 0.0221 0.00775 -0.00358 0.02351 1978 116006 277 25.73 0.0414 0.01027 -0.00220 0.02787 1979 84044 217 20.17 0.0494 0.04936 0.02233 0.08971 1980 87092 250 19.62 0.0996 0.05579 0.02807 0.09622 1981 104863 310 22.52 0.1224 0.04246 0.01850 0.07768 1982 134997 361 22.89 0.0652 0.02972 0.01062 0.05729 1983 96825 290 21.68 0.0657 0.03969 0.01693 0.07294 1984 99032 220 18.93 0.0636 0.02379 0.00604 0.04970 1985 172034 300 22.70 0.0598 0.02070 0.00263 0.04805 1986 256645 515 20.24 0.0992 0.04028 0.01700 0.07449 1987 220672 766 24.05 0.1443 0.05225 0.02404 0.09454 1988 155941 565 18.75 0.1758 0.02808 0.00521 0.06483 1989 144898 983 19.52 0.3474 0.06038 0.02532 0.11672 1990 199600 895 20.40 0.0950 0.01759 0.00159 0.04111 1991 165268 803 18.06 0.1915 -0.00683 -0.01470 0.00452 1992 185315 598 17.52 0.1292 0.00018 -0.01011 0.01521 1993 258161 616 22.58 0.1055 0.09319 0.05119 0.15627 1994 146607 1350 32.34 0.2847 0.14706 0.09336 0.22367 1995 142893 907 31.22 0.2033 0.05642 0.03014 0.09372 1996 80793 566 35.85 0.1954 0.10568 0.06586 0.16162 1997 78902 441 35.73 0.1564 0.07299 0.04295 0.11500 1998 231466 506 46.03 0.0475 0.02078 0.00422 0.04479 1999 212000 464 - - - -

Table 1.

The

catches, fishing effort, nominal catch per unit effort (CPUE) and adjusted catch per unit effort of white

marlin by Taiwanese longline fishery in the whole Atlantic from 1967 to 1999. The CPUE was adjusted by GLM

with year, quarter, area factors and quarterarea interaction as described in the text. '' indicates a suspected figure.

Year Catch

(north) (south) Catch (north) Effort (south) Effort Nominal CPUE (north) Nominal CPUE (south) Adjusted CPUE (north) Adjusted CPUE (south) 1967 47 134 1771 1907 0.0536 0.0871 1968 58 327 16997 39610 0.2143 0.3570 0.09678 0.23747 1969 135 448 19792 52770 0.3231 0.3172 0.13496 0.24198 1970 104 508 28343 42380 0.1278 0.4184 0.11184 0.12434 1971 178 260 31562 71432 0.2633 0.1365 0.16131 0.10098 1972 244 469 20363 78034 0.2172 0.0841 0.11616 0.09217 1973 135 464 36491 72058 0.1854 0.2704 0.10761 0.14404 1974 252 285 33913 64756 0.3704 0.1978 0.21627 0.09857 1975 125 382 29894 60926 0.2207 0.2984 0.12621 0.10957 1976 142 377 25729 84302 0.0430 0.0799 0.00886 0.02521 1977 44 119 35488 106852 0.0130 0.0251 0.00263 0.01378 1978 79 198 18409 97597 0.0368 0.0423 0.00741 0.01371 1979 62 155 17737 66307 0.0660 0.0449 0.06814 0.03488 1980 105 145 19185 67907 0.2180 0.0661 0.07554 0.03870 1981 174 136 25276 79587 0.2896 0.0693 0.10724 0.02183 1982 134 227 39385 95612 0.1172 0.0438 0.06462 0.02424 1983 203 87 52030 44795 0.0880 0.0398 0.06846 0.03677 1984 96 124 65236 33796 0.0430 0.1033 0.03557 0.03959 1985 128 172 75683 96351 0.0689 0.0527 0.03894 0.02776 1986 319 196 120109 136536 0.1396 0.0637 0.08007 0.02444 1987 153 613 47611 173061 0.1537 0.1417 0.08879 0.03941 1988 0 565 10433 145507 0.0000 0.1884 0.00787 0.09381 1989 4 979 7029 137869 0.0046 0.3648 0.02513 0.11233 1990 85 810 29144 170456 0.0212 0.1076 0.04412 0.02398 1991 13 790 31773 133495 0.0059 0.2357 -0.00280 0.01476 1992 92 506 29726 155589 0.0262 0.1489 0.00847 0.01811 1993 123 493 56280 201881 0.1970 0.0800 0.16138 0.08036 1994 506 844 38944 107662 0.3194 0.2722 0.16311 0.15685 1995 241 666 32847 110046 0.2266 0.1963 0.07214 0.06842 1996 122 444 19874 60919 0.1511 0.2099 0.09633 0.14427 1997 80 361 15997 62905 0.1361 0.1616 0.08826 0.08530 1998 105 401 44663 186803 0.0468 0.0477 0.03693 0.03189 1999 96 368 40790 171460 - -

Table 2.

The catches (mt), fishing effort (1000 hooks), nominal catch per unit effort (CPUE, no./1000 hooks) and

adjusted catch per unit effort (no./1000 hooks) of white marlin by Taiwanese longline fishery in the north and south

Atlantic from 1967 to 1999. The CPUE (no./1000 hooks) was adjusted by GLM with year, quarter, area factors and

quarter*area interaction as described in the text.

(6)

Year Fishing effort (1000hooks) Catch (mt) Weight Mean (kg) Nomial CPUE (no./1000 hooks) Adjusted CPUE (no./1000 hooks) Lower CI of adjusted CPUE (no./1000 hooks) Upper CI of adjusted CPUE (no./1000 hooks) 1967 3678 291 70.98 0.2477 1968 56606 803 76.39 0.1859 0.14956 0.10846 0.20430 1969 72562 1364 76.23 0.1861 0.21423 0.16254 0.28091 1970 70723 929 85.11 0.1338 0.10381 0.07698 0.13843 1971 102994 935 71.49 0.1271 0.10302 0.07560 0.13867 1972 98397 928 88.73 0.0540 0.07153 0.05048 0.09928 1973 108549 692 68.96 0.0604 0.06071 0.04256 0.09928 1974 98669 552 59.07 0.0584 0.03764 0.02551 0.08453 1975 90820 527 62.98 0.0291 0.02536 0.01562 0.05335 1976 110031 409 16.79** 0.0481 0.01338 0.00649 0.03816 1977 142341 171 83.41 0.0138 0.00864 0.00309 0.02241 1978 116006 258 85.42 0.0082 0.00547 0.00047 0.01584 1979 84044 190 63.50 0.0146 0.01284 0.00515 0.01204 1980 87092 289 47.65 0.0381 0.03224 0.01995 0.02340 1981 104863 202 67.11 0.0253 0.03698 0.02312 0.05582 1982 134997 250 58.81 0.0189 0.02129 0.01197 0.03377 1983 96825 172 59.14 0.0156 0.02050 0.01112 0.03318 1984 99032 172 55.32 0.0227 0.02451 0.01409 0.03859 1985 172034 313 59.56 0.0177 0.01508 0.00647 0.02705 1986 256645 215 53.43 0.0157 0.01541 0.00724 0.02651 1987 220672 317 66.74 0.0215 0.01873 0.00930 0.03175 1988 155941 292 57.25 0.0250 0.02332 0.01111 0.04113 1989 144898 473 53.03 0.0451 0.02215 0.01030 0.03941 1990 199600 1704 62.73 0.0510 0.06163 0.04127 0.08928 1991 165268 1672 55.27 0.0620 0.0926 0.00299 0.01765 1992 185315 824 69.14 0.0353 0.01780 0.00913 0.02950 1993 258161 685 53.37 0.0498 0.05518 0.03563 0.08220 1994 146607 663 59.27 0.0763 0.08306 0.05878 0.1153 1995 142893 467 71.98 0.0454 0.06233 0.04339 0.08736 1996 80793 660 88.83 0.0920 0.06324 0.04457 0.08771 1997 78902 1478 89.48 0.2093 0.06710 0.04775 0.09235 1998 231466 578 81.01 0.0308 0.03682 0.02340 0.05486 1999 212000 486 - - -

Table 3.

The catches, fishing effort, nominal catch per unit effort (CPUE) and adjusted catch per unit effort of blue

marlin by Taiwanese

longline fishery in the total Atlantic from 1967 to 1994. The CPUE was adjusted by GLM with

year, quarter, area factors and quarter -area interaction as described in the text. '*' denotes a suspected figures; and

'**' denotes a previous year mean weight was used in the estimation of adjusted CPUE in weight.

Year Catch (north) (mt) Catch (south) (mt) Effort (north) /1000 Effort (south) /1000 Nominal CPUE (north) Nominal CPUE (south) Adjusted CPUE (north) Adjusted CPUE (south) 1967 131 160 1771 1907 0.4872 0.0252 1968 374 429 16997 39610 0.3202 0.1283 0.18887 0.12529 1969 348 1016 19792 52770 0.1856 0.1862 0.19655 0.23253 1970 369 560 28343 42380 0.1142 0.1469 0.10104 0.10669 1971 193 742 31562 71432 0.1053 0.1368 0.07810 0.12930 1972 300 628 20363 78034 0.0609 0.0522 0.06431 0.07913 1973 155 537 36491 72058 0.0513 0.0651 0.07138 0.05056 1974 183 369 33913 64756 0.0800 0.0471 0.03945 0.03539 1975 105 422 29894 60926 0.0369 0.0253 0.01989 0.03368 1976 169 240 25729 84302 0.0566 0.0455 0.00876 0.02019 1977 64 107 35488 106852 0.0239 0.0105 0.00902 0.00812 1978 81 177 18409 97597 0.0073 0.0083 0.00217 0.00984 1979 51 139 17737 66307 0.0162 0.0142 0.00629 0.02108 1980 160 129 19185 67907 0.1160 0.0161 0.04953 0.01858 1981 98 104 25276 79587 0.0430 0.0196 0.03392 0.05031 1982 100 150 39385 95612 0.0338 0.0128 0.01659 0.03313 1983 125 47 52030 44795 0.0190 0.0095 0.01602 0.03570 1984 102 70 65236 33796 0.0226 0.0228 0.01349 0.05319 1985 148 165 75683 96351 0.0220 0.0143 0.00843 0.03711 1986 117 98 120109 136536 0.0202 0.0118 0.00985 0.03451 1987 52 265 47611 173061 0.0204 0.0218 0.01023 0.04428 1988 26 266 10433 145507 0.0281 0.0248 0.01847 0.04508 1989 11 462 7029 137869 0.0048 0.0471 0.01015 0.05694 1990 937 767 29144 170456 0.1429 0.0351 0.08379 0.07174 1991 716 956 31773 133495 0.0882 0.0558 0.00566 0.02147 1992 336 488 29726 155589 0.0428 0.0338 0.00935 0.03554 1993 281 404 56280 201881 0.1528 0.0211 0.08443 0.05433 1994 180 483 38944 107662 0.0969 0.0689 0.06025 0.11899 1995 97 370 32847 110046 0.0415 0.0466 0.03882 0.11410 1996 148 512 19874 60919 0.0925 0.0918 0.04026 0.10059 1997 37 110 15997 62905 0.2454 0.2002 0.03959 0.10998 1998 80 498 44663 186803 0.0243 0.0324 0.01938 0.06094 1999 68 418 40790 171460 - -

Table 4. The catches (mt), fishing effort (1000 hooks), nominal catch per unit effort (CPUE, no./1000hooks) and

adjusted catch per unit effort (no./1000 hooks) of blue marlin by Taiwanese longline fishery in the north and south

Atlantic from 1967 to 1994. The CPUE (no./1000 hooks) was adjusted by GLM with year, quarter, area factors and

quarter*area interaction as described in the text.

(7)

(1) Total Atlantic

Sources DF Sum of Square

Mean Square F value Pr >F Model 51 4578728 8.9779 9.51 0.0001 Year 12 134.5806 11.2151 11.87 0.0001 Quarter 3 1.6928 0.5643 0.60 0.6171 Area 9 125.6934 13.9659 14.79 0.0001 Quarter-area 27 138.3005 5.1222 5.42 0.0001 Error 362 341.9222 0.9445 Corrected total 413 799.7951 R2 C.V. (%) Root MSE 0.5725 -44.2076 0.9719 (2) North Atlantic Sources DF Sum of Square

Mean Square F value Pr >F Model 31 220.9476 7.1273 6.33 0.0001 Year 12 79.3242 6.6103 5.87 0.0001 Quarter 3 29.4360 9.8120 8.71 0.0001 Area 4 42.8664 10.7166 9.51 0.0001 Quarter-area 12 33.7689 2.8141 2.50 0.0048 Error 175 197.1132 1.1264 Corrected total 206 418.0608 R2 C.V. (%) Root MSE 0.5285 -48.7896 1.0613 (3) South Atlantic Sources DF Sum of Square

Mean Square F value Pr >F Model 31 252.4035 8.1420 11.10 0.0001 Year 12 69.9865 5.8322 7.95 0.0001 Quarter 3 27.8093 9.2698 12.64 0.0001 Area 4 75.7306 18.9327 25.82 0.0001 Quarter-area 12 44.5340 3.7112 5.06 0.0001 Error 175 128.3183 0.7332 Corrected total 206 280.7218 R2 C.V. (%) Root MSE 0.6630 -38.5569 0.8563

Table 5.

A N O VA results for the factors in the GLM procedure foradjusted catch per unit effort for white marlin in

the Atlantic,1968-1980.

(1) Total Atlantic

Sources DF Sum of

Square Mean Square F value Pr >F

Model 56 438.8705 7.8370 7.04 0.0001 Year 17 128.7696 7.5747 6.81 0.0001 Quarter 3 7.3463 2.4488 2.20 0.0871 Area 9 211.9647 23.5516 21.17 0.0001 Quarter-area 27 78.5987 2.9111 2.62 0.0001 Error 492 547.3389 1.1125 Corrected total 548 986.2094 R2 C.V. (%) Root MSE 0.4450 -39.6872 1.0547 (2) North Atlantic Sources DF Sum of

Model 36 199.1861 5.5329 4.31 0.0001 Year 17 86.5608 5.0918 3.97 0.0001 Quarter 3 7.4633 2.4878 1.94 0.1239 Area 4 79.6601 19.9150 15.53 0.0001 Quarter-area 12 23.3548 1.9462 1.52 0.1186 Error 229 293.6380 1.2823 Corrected total 265 492.8241 R2 C.V. (%) Root MSE 0.4042 -43.8818 1.1324 (3) South Atlantic Sources DF Sum of Square

Mean Square F value Pr >F

Model 36 288.1117 8.0031 9.62 0.0001 Year 17 86.7808 5.1048 6.14 0.0001 Quarter 3 24.8601 8.2867 9.96 0.0001 Area 4 118.8026 29.7006 35.71 0.0001 Quarter-area 12 33.0466 2.7529 3.31 0.0002 Error 246 204.6246 0.8318 Corrected total 282 492.7363 R2 C.V. (%) Root MSE 0.5847 -33.3756 0.9120

Table 6.

A N O VA results for the factors in the GLM procedure foradjusted catch per unit effort for white marlin in

(8)

(1) Total Atlantic

Sources DF Sum of

Model 51 613.5362 12.0301 20.45 0.0001 Year 12 221.3072 18.4423 31.35 0.0001 Quarter 3 5.4758 1.8253 3.10 0.0267 Area 9 227.4461 25.2718 42.96 0.0001 Quarter-area 27 83.7882 3.1032 5.28 0.0001 Error 362 212.9302 Corrected total 413 826.4664 R2 C.V. (%) Root MSE 0.7424 -27.3527 0.7669 (2) North Atlantic Sources DF Sum of

Square Mean Square F value Pr >F Model 31 302.9902 9.7739 15.59 0.0001 Year 12 121.1708 10.0976 16.10 0.0001 Quarter 3 17.8062 5.9254 9.47 0.0001 Area 4 103.3224 25.8306 41.20 0.0001 Quarter-area 12 20.5776 1.7148 2.73 0.0002 Error 175 109.7260 0.6270 Corrected total 206 412.7162 R2 C.V. (%) Root MSE 0.7341 -28.6170 0.7918 (3) South Atlantic Sources DF Sum of Square

Table 7.

A N O VA results for the factors in the GLM procedure foradjusted catch per unit effort for blue marlin in the

Atlantic,1968-1980.

Table 8.

A N O VA results for the factors in the GLM procedure foradjusted catch per unit effort for blue marlin in the

Atlantic,1981-1998.

(1) Total Atlantic

Sources DF Sum of

Model 56 379.3962 6.7749 9.62 0.0001 Year 17 96.4730 5.6749 8.06 0.0001 Quarter 3 2.8449 0.9483 1.35 0.2587 Area 9 158.1803 17.5756 24.95 0.0001 Quarter-area 27 76.7587 2.8429 4.04 0.0001 Error 492 346.5839 0.7044 Corrected total 548 725.9801 R2 C.V. (%) Root MSE 0.5226 -27.8631 0.8393 (2) North Atlantic Sources DF Sum of

Model 36 262.1989 7.2833 5.90 0.0001 Year 17 80.8947 4.7585 3.85 0.0001 Quarter 3 7.2888 2.4296 1.97 0.1195 Area 4 92.5526 23.1381 18.74 0.0001 Quarter-area 12 42.99348 3.5779 2.90 0.0009 Error 229 282.6842 1.2344 Corrected total 265 544.8832 R2 C.V. (%) Root MSE 0.4812 -32.7932 1.1110 (3) South Atlantic Sources DF Sum of Square

(9)

White marlin 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1962 1965 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 Year Ca tc h (m t) Total (mt) South (mt) North (mt)

Blue marlin

0 150 300 450 600 750 900 1050 1200 1350 1500 1650 1800 1962 1965 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 Year Ca tc h ( m t) Total (mt) South (mt) North (mt)

Figure 1.

Annual catches of white marlin by Taiwanese

longline fleets in the Atlantic. Catches are separated

into assumed north and south Atlantic stocks at 5

o

N

and total Atlantic.

Figure 2.

Annual catches of blue marlin by Taiwanese

longline fleets in the atlantic. Catches are separated

into assumed north and south Atlantic stocks at 5

o

N

and total Atlantic.

Figure 3. Effort distributions of Taiwanese

longline fishery operating in the Atlantic Ocean.

Data shown are aggregated from 1968-1980

(top) and from 1981-1998 (bottom)

(10)

Figure 4. Nominal catch per unit effort distribution of white marlin for Taiwanese longline fishery in the Atlantic.

Data shown are aggregated from 1968to 1980 (left) and from 1981 to 1998 (right).

Figure 5. Nominal catch per unit effort distribution of blue marlin for Taiwanese longline fishery in the Atlantic.

Data shown are aggregated from 1968 to 1980 (left) and from 1981 to 1998 (right).

(11)

Figure 6.

Area division used for general linear model analysis on the standardization of catch per unit effort of white

marlin and blue marlin by Taiwanese longline fishery in the Atlantic.

White marlin 0 0.05 0.1 0.15 0.2 0.25 0.3 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998

Year

C P U E (N o. /1 000 h ook s) White marlin 0 0.05 0.1 0.15 0.2 0.25 0.3 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 Year C P U E ( N o. /1 000 ho oks ) Blue marlin 0 0.05 0.1 0.15 0.2 0.25 0.3 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 Year C P U E ( N o. /1 000 ho oks ) Blue marlin 0 0.05 0.1 0.15 0.2 0.25 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 Year CP U E ( N o. /1 00 0 ho ok s)