ir e nt », Marine Fisheries é =. t *, = » ” ares or Monterey Bay Rockfish Marine Fisher 3 Sr W. L. Hobart, Editor J. A. Strader, Managing Editor On the cover: Blue rockfish, Sebastes mystinus, is one of the many rockfish species in Monterey Bay, California. Photo by Robert R. Lauth. Articles 60(3), 1998 Review of a Small-scale Pelagic F. H. V. Hazin, J. R. Zagaglia, M. K. Broadhurst, Longline Fishery off Northeastern Brazil P. E. P. Travassos, and T. R. Q. Bezerra Bottlenose Dolphins, Tursiops truncatus, Removing By-Catch from Prawn-trawl] Codends During Fishing in New South Wales, Australia M. K. Broadhurst Declining Rockfish Lengths in the Monterey Bay, California, Recreational Fishery, 1959-94 Janet E. Mason Mortality of Lingcod, Ophiodon elongatus, Related to Capture by Hook and Line Douglas Albin and Konstantin A. Karpov U.S. DEPARTMENT OF The Marine Fisheries Review (ISSN 0090-1830) is the Office of Management and Budget. COMMERCE published quarterly by the Scientific Publications Of- The NMFS does not approve, recommend, or endorse fice, National Marine Fisheries Service, NOAA, 7600 any proprietary product or proprietary material men- William M. Daley, Secretary Sand Point Way N.E., BIN C15700, Seattle, WA 98115. tioned in this publication. No reference shall be made Annual subscriptions are sold by the Superintendent of to the NMFS, or to this publication furnished by the Documents, U.S. Government Printing Office, Wash- NMFS, in any advertising or sales promotion which ington, DC 20402: The annual subscription price is would indicate or imply that the NMFS approves, rec- NATIONAL OCEANIC AND $10.00 domestic, $12.50 foreign. For new subscriptions ommends, or endorses any proprietary product or pro- ATMOSPHERIC ADMINISTRATION write: New Orders, Superintendent of Documents, P.O prietary material mentioned herein, or which has as its D. James Baker, Under Secretary Box 371954, Pittsburgh, PA 15250-7954. purpose an intent to cause directly or indirectly the ad- Publication of material from sources outside the vertised product to be used or purchased because of for Oceans and Atmosphere NMFS is not an endorsement, and the NMFS is not re- this NMFS publication. POSTMASTER: Send address sponsible for the accuracy of facts, views, or opinions changes for subscriptions for this journal to: Marine of the sources. The Secretary of Commerce has deter- Fisheries Review, c/o Superintendent of Documents, U.S National Marine Fisheries Service mined that the publication of this periodical is neces- Government Printing Office, Washington, DC 20402. sary for the transaction of public business required by This issue, volume 60 number 3, was printed and dis- Penelope D. Dalton, law of this Department. Use of the funds for printing tributed in July 1999. Assistant Administrator for Fisheries this periodical has been approved by the Director of Review of a Small-scale Pelagic Longline Fishery off Northeastern Brazil F.H. V. HAZIN, J. R. ZAGAGLIA, M. K. BROADHURST, P. E. P. TRAVASSOS, and T. R.Q. BEZERRA Introduction gan to operate from Natal (Fig. 1) us- Argus, Rio Turi, and Soloncy Moura) ing relatively small vessels to target tu- operating off northeastern Brazil from The pelagic longline fishery off nas and other large pelagic species of 1983 to 1997 (Fig. 1). All vessels used northeast Brazil began in 1956 with sev- fish. This fleet expanded throughout the similar configurations of Japanese-style eral Japanese longliners (leased by a following 10 years and by 1997 con- multifilament longlines (Shapiro, 1950; Brazilian company) mainly targeting sisted of 10 boats ranging in size from Suzuki et al., 1977). Each longline con- tunas (Thunnus spp.). This fleet con- about 16 to 26 m. sisted of a multifilament mainline (Fig. sisted of 12 boats in 1959 (Paiva and As part of management regulations 2A) with secondary lines (Fig. 2B) at- Le Gall, 1975), and although quite good governing the present fishery, operators tached in clusters of 6—7 (termed “bas- catches were achieved, fishing opera- have been required to complete logsheets kets”) over approx. 360 m. Styrofoam tions were suspended in 1964 owing to for each fishing trip. Information re- bouys, each attached to a 25 m line were economic and political reasons. During quested include location of fishing tied to the mainline after every cluster 1976 and 1977, the fishery experienced grounds, number of hooks used on the of hooks (Fig. 2A). The types of hooks a brief revival through the leasing of two longline, time of setting and retrieving used varied among 3 main brands (de- Korean longliners; however, there was the gear, and composition of catches. pending on availability) however, rela- no significant effort until 1983 when a In addition, at the end of fishing trips, tive sizes remained similar throughout Brazilian company, “Norte Pesca,” be- biological data on the various species the period examined (Fig. 2C). The captured have routinely been collected mean number of hooks (+ SE) set per by researchers. Although some of these vessel per day was 975 + 7. Fishing The authors are with the Universidade Federal data have been used in studies examin- methods and operations were similar Rural de Pernambuco-UFRPE, Departamento de ing biological aspects of the main spe- across the fleet with the mainline-set Pesca, Laboratorio de Oceanografia Pesqueira, Av. Dom Manuel de Medeiros, s/n, Dois Irmaos, cies, including their relative distribution begining at about 0200 h and ending at Recife-PE, Brazil, CEP: 52.171-900. and abundance (Hazin et al., 1990; dawn. The gear was then left to fish for 1994a), reproductive biology (Hazin et about 6 h, before retrieval began at noon al., 1994b), and feeding habits (Hazin and ended at dusk. The primary bait was ABSTRACT—The annual catches of four et al., 1994c), no studies have been done the Brazilian sardine, Sardinella brasi- small longliners operating off northeast Brazil from 1983 to 1997 were examined describing temporal and spatial vari- liensis, although some other species, across different areas and locations. The to- abilities in composition of catches. including flying fish, Cypselurus cyan- tal catch comprised tunas (30%), sharks To provide a brief overview of the opterus, and squid, Loligo sp., were (54%), billfishes (12%), and other fish spe- changes in fishing strategy and catches, occasionally used. pcioessi t(i4o%n) .o f Fciasthcihnegs ssthraotweegdy laanrdg e ansnpautaila lc oamn-d including some analysis of relative Due to different oceanographic and temporal variabilities with the dominant abundance and distribution of the main biological conditions (Paiva and Le catches alternating among yellowfin tuna, species, our aims in this paper were to Gall, 1975; Hazin') and for the purposes Thunnus albacares; gray sharks, Carcha- collate yearly catch data from four ves- of this study, the total area fished (Fig. rhinus spp.; and blue shark, Prionace sels that have operated in the fishery 1) was divided into the following three glauca. Catches of blue and gray sharks since 1983 and compare these across subareas for analysis: showed a significant interaction among sea- mounts, with gray sharks occurring in maxi- defined areas and locations. mum abundance around those seamounts that had relatively deep summits and low- Material and Methods ' Hazin, F. H. V. 1993. Fisheries-oceanographi- sloping depth profiles. Results are discussed This study was done using data from cal study on tunas, billfishes and sharks in the in terms of the various factors that may have southwest equatorial Atlantic Ocean. D. Sc. The- influenced distribution of effort. the logsheets of four longliners, (A/fa, sis, Tokyo Univ. Fish., 286 p 60(3), 1998 Subarea I (north of lat. 5°S and west (CPUE), defined as the number of fish this study, these were grouped under the of long. 35°W) contains several large caught per 100 hooks per year (no. of category of “gray sharks,” with the ex- seamounts comprising the North Bra- fish/100 hooks/year). Catch records in- ception of C. maou, due to its easily iden- zilian Chain and is influenced by the cluded identification at the species level, tifiable characteristics. Data for two spe- North Brazil Current. with the exception of sharks, which cies of mako sharks (/surus oxyrinchus Subarea II (north of lat. 5°S and east were collectively grouped prior to 1986. and /. paucus) were also combined. of long. 35°W) contains Atol das Rocas, Most fishermen included catches of in- Analysis of Data Fernando de Noronha, and the Archi- dividual species during subsequent pelago of St. Peter and St. Paul and is trips, although it wasn’t until 1990 that To examine temporal and spatial fluc- influenced by the South Equatorial all provided these data. Consequently, tuations in relative abundance and dis- Current. while the CPUE of total sharks was cal- tribution throughout the fishery, yearly Subarea III (south of lat. 5°S) in- culated using all hooks set in each year estimates of CPUE for groups of spe- cludes a deep oceanic area with no sea- (Table 1), the CPUE of individual spe- cies were calculated for each of the three mounts or islands and is influenced by cies of sharks was derived using the areas. Further, because significant dif- the Brazil Current. number of hooks pooled across only ferences were detected in the CPUE of those trips that included a complete sharks between areas (see Result§ and Data from the four boats were tabulation of catches (Table 2). In addi- Discussion) the yearly CPUE’s of gray grouped together (to provide a larger tion, because of difficulties in distin- and blue sharks, Prionace glauca, were dataset) and total catches were standard- guishing some of the species of the ge- calculated at seven seamounts located ized to yearly catch per unit of effort nus Carcharhinus, for the purposes of in subareas I and II (Fig. 1, Table 3). In deriving these data, longlines were con- 2N sidered to be in the vicinity of a sea- mount whenever they were located Archipelago of St. @ within 5 n.mi. of the 1,000 m isobath. Peter and St. Paul Subarea | Yearly CPUE values for relevant groups and species from each area and for gray and blue sharks from each sea- Subarea II mount were analyzed using Cochran’s test for homogeneity of variances. Data were transformed if necessary and then analyzed in appropriate one and two- factor analyses of variance, respectively Fortaleza @ @ Fernando de (Underwood, 1981). Significant differ- Noronha ences detected in these analyses were investigated using Tukey’s multiple comparisons of means test. The arith- metic mean yearly CPUE’s used in these analyses are presented with their associated standard errors. Total yearly CPUE’s combined across all areas, Recife @ depths, and locations for all groups and species that showed sufficient catches were also calculated and graphed. Subarea III Results and Discussion The yearly catch data collected dur- ing the period examined are provided in Tables | and 2. Tunas and sharks, and in particular, yellowfin tuna, Thunnus Salvador albacares, and gray and blue sharks were the dominant groups, accounting for almost 84% of the total catch. oe ae T There were no significant differences 42 W 38W 34 W 30 W in the arithmetic mean yearly CPUE’s of total tunas and billfishes across the Figure |.—Location of the area fished during the period examined, including the approximate position of seamounts: Aracati (A), Dois Irmaos (DI), Fundo (F), Sirus three subareas examined (Fig. 3A, B, (SI), Pequeno (P), Leste (L), and Sueste (SU). Table 4). While differences were de- Marine Fisheries Review —_ , gies >on — oar or—o e— —_e—e tal ee e e } 25 m(o 6mm, PA/ PP, multi) } }| 360 m (0 6 mm, PA/ PP, multi) 15m(o45mm PA/PP, multi) 2m (o 1.5mm, wire) Figure 2.—Diagrammatic representation of typical (A) longline configuration; (B) secondary line: and (C) hooks used by the four vessels over the period examined. @ = diameter, mono = monofilament, multi = multifilament, PA = polyamide, PP = polypropylene Table 1.—Annual number of hooks used and catches (from 1983 to 1997) for various species and groups Tunas Billfisnes Other fist Albacore Swordfist Sailfish White marlir > ' ombined tected between mean CPUE’s of other der of catches among the three subar- sharks was significantly greater in sub- fish combined (Fig. 3C, Table 4), eas (Fig. 3C). In contrast. analyses area! than in subareas II and III (differ- Tukey’s tests showed no definitive or- showed that the mean CPUE of total ence between meanso f up to 68%) (Fig 60( 3}, 1998 Table 2.—Annual number of hooks used and catches (from 1986 to 1997) for individual species of sharks. Crocodile Total no Biue Gray Bigeye thresher Mako Ocean whitetip (Pseudocarcharias Other Year of hooks Prionace glauca) (Carcharhinus spp.) (Alopias superciliosus) (/surus spp.) (C. maou) kamoharai) shark species 1986 147,237 715 221 2 41 0 w o 286 1987 338,704 345 973 1 74 41 1988 399,430 2,503 489 103 26 1889 274,126 945 934 56 37 1990 273,700 986 7714 56 19 1991 101,670 577 1992 130,912 2 392 8S6=ao un 1993 89,222 3 212 1994 116,964 3 122 1995 172,100 8 082 1996 229,888 73 6 231 1997 74,060 1,455 oo0 0 o 3D, Table 4). This increase in CPUE Table 3.—The approximate location (midpoint) of seamounts examined, depth at the summit, and area contained within 5 n.mi. beyond the 1,000 m isobath. Seamounts in bold represent those located in subarea I. may be attributed to the significant in- crease in catches of gray sharks in sub- Seamount area I and the effects on total catches of Item Aracati Dois Irmaos Fundo Sirus Pequeno Leste Sueste sharks, since their mean CPUE was 9 Location (lat.) 3°20'S 3°20'S 3°52'S 4°00'S 3°50'S 3°45'S 4°16'S times lower in subarea II and almost (long.) 37°30'W 36°38'W 35°22'W 35°55' W 34°44'W 33°12'W 33°15'W zero in subarea III (Fig. 3E, Table 4). Depth at summit (m) 254 370 214 233 124 176 38 Although much lower, the catches of Area (km*) 4,837 1,110 2,815 5,075 833 952 833 blue sharks displayed the opposite trend and were significantly greater in subar- eas IT and III than in subarea I (Fig. 3F, Table 4.—Summaries of F ratios from analysis of variance to determine effects on CPUE due to different areas Table 4). for total tunas, sharks, billfishes, and other fish combined between 1983 and 1997 and for blue and gray sharks between 1987 and 1997. Because 5 and 3 replicate CPUE data were missing from each analysis respectively, we Possible explanations for the ob- substituted means of the remaining replicates and reduced the degrees of freedom accordingly. The transforms used to stabilize variances (if required) are also listed. ‘significant (P = 0.05); **significant (P = 0.01). served anomaly between the CPUE of gray and blue sharks may be the fish- Source of Totai Total Total Other fish Blue Gray variation df tunas billfishes sharks combined df shark shark ing location in subarea I and species- specific variabilities in habitat prefer- In(x+1) In(x+1) In(x+1) Area 0.37 14.54°° 22.50°° 36.60°* ence (Hazin et al., 1990). For example, Residua! unlike subareas II and II, which mainly comprise deep ocean, many of the sets in subarea I were done in the vicinity between means of up to 95%) at those Table 5.—Summaries of F ratios from analysis of vari- of seamounts that were relatively shal- seamounts that had relatively deep sum- ance to determine effects on yearly CPUE for two groups of sharks and different seamounts. Because low over large areas (see Table 3). Be- mits (e.g. 233-370 m) and were shal- 10 CPUE data were missing, we substituted means of cause previous studies have shown that low over a large area (i.e. Aracati, Dois the remaining replicates and reduced the degrees of freedom accordingly. The sqrt(x+1) transform was used the blue shark typically is an oceanic Irmaos, Fundo, and Sirus) than those to stabilize variances. **significant (P = 0.01). species, its relative abundance might be characterized by shallow summits (38- Source of variation df CPUE expected to decline across these areas 176 m) and steep depth profiles (i.e. (Strassburg, 1958; Hazin, et al., 1990). Pequeno, Leste, and Sueste). In the ab- Seamount 6 6.38"* Shark group 1 97.57°° In support of this, the mean CPUE of sence of any data describing oceano- Interaction 6 14.36°° blue sharks was consistently low across graphic conditions and habitats at these Residual 130 individual seamounts (Fig. 4) and different types of seamounts, it is diffi- showed little variability since Tukey’s cult to determine possible causes for the test for a significant interaction detected observed increase in CPUE at the sea- Columns” (Rogers, 1994; Travassos et between seamounts and groups of mounts with deeper summits and low- al., In press). The extent to which these sharks (Table 5) failed to detect any sig- sloping depth profiles. Previous stud- sorts of processes may facilitate primary nificant differences in mean CPUE of ies have shown, however, that sea- production and consequent cascading this group among the seamounts exam- mounts in this category may have more effects throughout the trophic chain is ined (Fig. 4, Table 5). turbulence, due to the interaction be- unknown but, given the observations in Conversely, the CPUE of gray sharks tween oceanic currents and the subma- the present study, they may have had did show an effect due to seamounts and rine relief, that results in the formation some influence that contributed to the was significantly greater (difference of upwellings and possibly “Taylor greater relative abundance of gray Marine Fisheries Review A) Total tunas B) Total billfishes i 7 C) Other fish combined D) Total sharks aaron I Cof(hfnioP oso.Uhk /Es1/ 0y0e ar) _E) Gray shark F) Blue shark po my Subarea Figure 3.—Differences in arithmetic mean yearly CPUE (+SE) of (A) total tunas, (B) total billfishes, (C) other fish combined, (D) total sharks, (E) gray shark, and (F) blue shark across each of the three areas. <, >, and = indicate direction of differ- ences detected in Tukey's comparison of means test. sharks in subarea I| and the significant 1.44 in 1989 to 0.10i n 1997 (Fig6).. T he 7). Although total catches of billfishes increase in their CPUE. CPUE of albacore. Thunnus alalunea, remained fairly stable (Fig. 5), individual The data presented in Figures 5, 6, 7, showed less variability but steadily de- catches of swordfish, Xiphias gladius. and 8 show quite large temporal fluc- creased from 0.56 in 1983 to 0.001 in steadily increased from 1991 (Fig. 8). tuations of total yearly CPUE (com- 1993, whilst the yearly CPUE of big- Many of these temporal variabilities bined across all areas and depths) of eye tuna, 7: obesus, was always very low can be explained with respect to changes most groups and species examined. For (Fig. 6). Catches of gray and blue sharks in overall fishing strategy over the past 14 example, with the exception of the first showed similar fluctuations with blue years, summarized in 5 distinct periods: year of the fishery, yellowfin tuna was shark most abundant from 1986 until June always the most abundant species of 1991 before the CPUE of gray sharks Period | (July 1983—June 1986). Tu- tuna, with a yearly CPUE varying from began to increase, peaking in 1995 (Fig. nas (i.e. mainly yellowfin tuna) and 60(3), 1998 Ee] Gray shark [| Blue shark A>B B B S a: B is 5 Cofh(finoP osoU.hk /Es1/ 0y0e ar) B B B B =. aaa | | a | —_ Aracati Dois Irmaos Fundo Sirus Leste Pequeno Sueste Seamount Figure 4 Differences in arithmetic mean yearly CPUE (+SE) of (A) gray shark, and (B) blue shark across different seamounts. <, >, and = indicate direction of differences detected in Tukey's comparison of means test. billfishes (swordfish, sailfish, /stio- ity. In contrast to many international CPUE of total tunas (Fig. 5) and also phorus albicans, white marlin, Tetrap- longline fisheries, which tend to remove resulting in a slight drop in the CPUE turus albidus, and blue marlin, Makaira the fins and discard the carcass at sea, of sharks (1988-90) due to a shift in nigricans) were the main target groups in this fishery all shark carcasses are effort away from their areas of maxi- whilst sharks were avoided. landed and sold at the local market. mum abundance (i.e. the seamounts in Period 2 (July 1986—December 1987). However, because of the low price of subarea I) (Fig. 5). Along with tunas and billfishes, sharks their meat, prior to June 1986 the re- After 1992 (during period 4), vessels (mainly the blue) became a target group. turn paid to fishermen was half that for concentrated around the many sea- Period 3 (January 1988—June 1991). tunas and billfishes. In July 1986, the mounts mainly located in subarea I to Fishing grounds were discovered off the fishing company standardized the return target gray sharks (Fig. 7). Contributing Archipelago of St. Peter and St. Paul, for all species caught after it realized factors towards the greater shift in effort with associated increases in catches of that although sharks were worth less on this group were |) the discovery of yellowfin tuna, shifting effort towards than other species, their abundance large abundances of these individuals this species. meant that production could be more above those seamounts in subarea | that Period 4 (July 1991—June 1996). than doubled. Facilitating this diversi- are shallow over a large area (see discus- Fishing effort was concentrated over fication of effort was an established lo- sion above) and 2) an increase in the price shallow seamounts, where species com- cal market for shark products, (e.g. fro- of shark fins for international markets. prising gray sharks were abundant. zen fillets) that had developed from the Prompted by the leasing of an Ameri- Catches were mainly dominated by steady production of shark meat during can swordfish longliner, vessels began to these sharks. the first 3 years of the fishery. shift fishing effort again during mid 1996 Period 5 (July 1996—June 1997). Al- From January 1988 and during the (period 5), using light sticks above the though gray sharks were stull the domi- first quarter of each consecutive year hooks and squid as bait to target sword- nant group, the CPUE of swordfish rose (period 3), vessels began to operate in fish. Recent modifications to gear, particu- sharply owing to a concentrated effort the vicinity of the Archipelago of St. larly the use of monofilament mainlines, involving the use of light sticks above Peter and St. Paul to target yellowfin should see this trend increase as fisher- the hooks and squid as bait. tuna that aggregate in large numbers to men adopt the latest technology and meth- feed on dense schools of spawning fly- ods developed in other fisheries (Bjordal, The initial diversification in effort ing fish (Hazin'). As a result, the CPUE 1989: Lokkeborg and Bjordal, 1992). that resulted in sharks being included of this species more than tripled in the It is apparent that a number of fac- as a target group (period 2) mainly oc- first 2 years after the new fishing tors have contributed to changes in fish- curred because of an increased aware- grounds were discovered (Fig. 6). This ing strategy during the past 14 years. ness of their abundance and availabil- contributed to an overall rise in the Many of the shifts in effort appear to Marine Fisheries Review 4] Oo Total tunas Total billfishes Total sharks > oO Other fish combined ie) °o |e ee fCoh(finoP oso.Uhk /Es1/ 0y0e ar) Year Figure 5.— Yearly CPUE estimates of total tunas, billfishes, sharks, and other spe cies combined throughout the period examined — Yellowfin tuna —is— Albacore oy wo —O— Bigeye tuna ® > ~” x ° \2) x o oO rr ~ & 2 = —_ co) ° = Lu a ou O 14 985 Year Figure 6.—Yearly CPUE estimates of yellowfin tuna, albacore, and bigeye tuna throughout the period examined sustainability of the fishery, that monitor have been either market orientated, re- related, given some of the more recent flecting variabilities in consumer de- developments (e.g. use of monofilament) ing of catch and effort data 1s maintained. mand for various species, or as a result and that most of the fishing areas have Acknowledgments of the discovery of new fishing grounds been fully explored, these sorts of changes and stocks. While few, if any, of the are likely to affect CPUE in the future. It This study was funded by the Comissao changes in the first 13 years are gear is important, therefore, for the continued Interministerial para os Recursosdo Mar 60(3), 1998 O1 — Blue chek (CCIIRR M 1) thrroouuggh h the the PrPoroggrarmaa maN acNaicoinozn il de Avaliagado dos Recursos Vivos da —Ls Gray shark ; . ee : : Ree Zona Economica Exclusiva (REVIZEE) —O— All other sharks combined and the Conselho National de Ensino e Pesquisa (CNPq). Thanks are extended to Norte Pesca for their assistance in compiling the catch data and to Geoff Gordon for helpful discussions. Literature Cited Bjordal, A. 1989. Recent developments in longline fishing - catching performance and conservation aspects. In Proceedings of the World Symposium on Fishing Gear and Fish- ing Vessel Design, St. Johns, Newfoundland, oO i. Canada, November 21-24, 1988, p. 19-24 L\, Hazin, F. H. V., A. A. Couto, K. Kihara, K hCof(finoP oso.Uhk /Es1/ 0}y o0I se a| |r ) \ \ \ Oatbsuunkdaa,n caen d ofM .p elIasghiicn o. sha1r9k9s0. in Ditshet risbouuttiho-nw esatn d x l yf \ pS = f= = —— a > ho Atlantic. J. Tokyo Univ. Fish K | 8 Hazin, F. H. V., C. E. Boeckman, E. C. Leal, R +149 86 41 987 14908 8 1989 coz] P. T. Lessa, K. Kihara, and K. Otsuka. 1994a. Distribution and relative abundance of the blue shark, Prionace glauca, in the southwestern equatorial Atlantic Ocean. Fish. Bull. 92:474 Figure 7.—Yearly CPUE estimateso f gray shark, blue shark, and other sharks com- 480 bined throughout the period examined .K. Kihara, K. Otsuka, C. E. Boeckman, and E. C. Leal. 1994b. Reproduction of the blue shark Prionace glauca in the south-west ern equatorial Atlantic Ocean. Fish. Sci. (To kyo) 60(5):487—49 | . R. Lessa, and M. Chammas. 1994¢ White marlin First observation on stomach contents of the blue shark, Prionace glauca, from southwest = —4— Swordfish ern equatorial Atlantic. Rev. Bras. Biol © —~O— Sailfish 154(2):195-198 ®© = : Lokkeborg, S., and A. Bjordal. 1992. Species and ” —* Blue marlin size selectivity in longline fishing: a review x Fish. Res. (Amst.) 13:311-322 ° 2) Paiva, M. P., and J. Y. Le Gall. 1975. Catches ot Aan tunas and tuna like fishes in the longline fish So ery area off the coast of Brazil Arq Ciénc © Mar. 15(1):1-18 x~= o n iF p \ | RogeArdsv,. DM.a r.D . Bi1o9l9,4 . 30T:h3e0 5b-i3o5l0o gy of seamounts 2 Shapiro,G . D. 1950. The Japanese longline fish — Sac9noe) 0.24 || | \ igs e1r-y 26 for tunas. Commer. Fish. Rev. 12(4): £ Strassburg, D. W. 1958. Distribution, abundance \ and habitats of pelagic sharks in the central Lu t , rm : Pacific Ocean. Fish. Bull. 58:335-361 Aa | UJ i ; Suzuki, Z.. Y.W arashima, and M. Kishida. 1977 ou / beg QD The comparison of catches by regular and O o~ e deep tuna longline gear in the western and ’ : central equatorial Pacific. Bull. Far Seas Fish Res. Lab. 15:51-89 9 Travassos,P . E. P. F.. F.H . V. Hazin, J. R. Zagaglia, 2) R. A. Rocha, and J. Schober. In press. Ther- mocline structure around seamounts and islands Year _ of Northeast Brazil Arch Fish. Mar. Res. Underwood, A. J. 1981. Techniques of analysis ; 7 oes of variance in experimental marine biology Figure 8.—Yearly CPUE estimates of white marlin, swordfish, sailfish, and blue and ecology. Oceanogr. Mar. Biol. Annu. Rev marlin throughout the period examined. 19:513-605 Marine Fisheries Review