EXPERIMENTAL SURFACE GILL NET FISHING FOR SKIPJACK IN HAWAIIAN WATERS se Biological Laboratory L 1 03 K. j:v, i* Tf MAR3i lt53 WOODS HOLE, MASS. SPECIAL SCIENTIFIC REPORT: FISHERIES No. 90 UNITED STATES DEPARTMENT OF THE INTERIOR FISH AND WILDLIFE SERVICE Explanatory Note The series embodies results of investigations, usually of restricted scope, intended to aid or direct management or utilization practices and as guides for administrative or legislative action. It is issued in limited quantities for the official use of Federal, State or cooperating agencies and in processed form for economy and to avoid delay in publica- tion. Washington, D. G< November 19?2 United States Department of the Interior, Oscar Lo Chapman^ Secretary Fish and Wildlife Service^ Albert M.. Day., Director EXPERIMENTAL SURFACE GILL NET FISHING FOR SKIPJACK (Katsuwonus pelaniis) IN HAWAIIAN WATERS By Walter M, Matsumoto Fishery Research Biologist Pacific Oceanic Fishery Investigations Special Scientific Reports Fisheries Noo 90 WASHIlGTONs NOVEMBER 1952 CONTENTS Page History of gill netting for tuna „ » o <> o , » o , o « » o » o o o » o o o o o . o o = 1 L/ao UX^X^ wlOn UX ^@a>X^ 0000000040000 ooooouoooaooooooooooooeoooo t " XollXIig UpcsX^ci ui QllS ooooooooooooo&ooooooooooooooooooooooooooo y Fishing grounds and seasons oooooooo<,»..o„o. 0000000000000000 lU iJjLJ wU.O SXOn o ooooooooooOo • 00000000 OOOoooooooooOO 0000000000000 cJLO o 1 iTTuTirt. iry siici concxusxons 000000*0000000 000000000000000 0000000 X-y XjX USro. t» IXrO ex U©Q •ooooooooooooooooooooooooooooooooooooeoo*** C\J ILLQSTEIATIONS Figure gage lo Plan of gear in operation^ o oo o oooo o o oo o o . ■ . . h 2c Details of individual shackleu o o o o o ooo o oo ooo o o o i.o ooo o 5 3„ Lee areas in the Hawaiian Islands ^ _ . . 10 lie Hawaiian skipjack landings by monibBj, iPU^-^-iPSl" o o o o o 16 EXPERIMENTAL SURFACE GILL NET FISHING FOR SKIPJACK (Katsuwonus pelamis) IN PIAWAIIAN WATERS By Vifalter lie Matsumotc Fishery Rfc-sesirch Biologxst The tuna canning industry of Hawaii,, utilizing the skipjack or striped tunac Katsuwonus pelamis (Linnaeus 1758), is limited by its bait si^jply t^elsh 19W^) and by the seasonal fluctuations in the catch„ Recognizing the possibility that any drastic bait shortage could seriously curtail the skipjack fisheryo local members of the Tuna Industry Advisory Committee for the Pacific Oceanic Fishery Investigations suggested a trial of gill nets as a possible means of taking these fish in commercial quantities without the use of baito Accordingly 5, experimental fishing with gill nets was done intermittently during the period from July 2 to October 19? 1951 ^ on the Territorial Division of Fish and Gajne's vessel Makua and the Fish and Wildlife Service's vessel John Ro Manning o This experiment was a cooperative venture of the Fish and Wildlife Service and the Di-i/ision of Fish and Game of the Territory of Hawaii » The author wishes to thank the Division of Fish and Game, Territory of Hawaii for their cooperation and the use of the Makua-i, and Dro William F„ Royce, who offered valuable advice and encouragement in planning the experiment and in preparing the manuscriDto HISTORY OF GILL NETTING FOR TUNA Although the g-^ll net is one of the oldest forms of nets en^iloyed in commertd al fishing, the available literature on its application to fishing for tuna contains only meager information,, and there is no report of Its being used to any extent in this fishery until the early part of the twentieth centuryo In northern Japan drift net fishing for black tuna^, Thunnus ori entails (Temminck and Schlegel);, began in 1905 at Kushiroj Urakawaj and Muroran_ when this species was first captured in drift nets set for sharks „ By 1527 the commercial fleet exceeded 200 vessels with a reported catch of over 8 million poondSo In the 1930*s the fishery reached a flourishing condition at Kushiro and Urakawa (Kawana 193U)j, but in recent yeajrs it has died out almost completely with the failure of the black tuna runs and a shift to longliningo In Peruj where the gill net is used successfully for bonitOj Sarda chilensiSc, and -skipjack;, atten^ts at gill netting boni to were made by the Uo So Fishery Mx-ssion to Peru in 19Ul with the vessel Pacific Queen (Fiedler j, JarviSj and Lobell 19lil) Approxi- mately UOO pounds of fish were caught during the 82 hours of the experiment, averaging 12 pounds per net and Uo? pounds per- houro In general, the experiments indicated that it was uneconomical to fish 'vfith gill nets from a boat the size of the Pacific Queeno More recently, nylon gill nets were used in an experiment in the Mediterranean Sea by a leading French nylon manitfacturer (Anonymous 1950)0 Fishing was done at the dark of the moon<, 5 to 6 fathoms below the surface » The tuna caught weighed from 26 to 110 pounds. Off the west coast of North America^ gill net experiments for albacore were conducted by the California Division of RLsh and Game on t'oa No B„ Scofield (Anonymoua 1951) o This experiment was made durxng the off season along the southei'n California coast to determine the whereabouts of the albacore rather than to test the method foi' possible commercial use^ and consequently only a few albacore were caught o Farther to the north the tuna explorations of the Uo So Fish and Wildlife Service vessel Jghn No Cobb, from June 12 to Septem- ber 28, 1950,9 produced a catch of 169 albacore taken in one set (Powell 195'0)o Aside from this one sizable catch., the experiment in general did not show any xmmediate promise of success on a com- mercial scale o From July 6 to August 9s 1951^ the gill net was again tried by the John No Cobbo Albacore were csaught in only 3 of the 13 sets made, with a total catch of 28 fisho In Hawaii gill nets were used previous to 1883s, ^^^ their use was limited to catching reef fishes and lobsters (Beckley 1886) o l^ to the time of the present experiment the only recorded attempt to use a drift net for tuna in this area was made by the California Division of Fish and Game vesselj No Bo Scofieldc. on its cruise te to the Hawaiian Islands from July 21 to September 139 19U8 (Godsil and Greenhood 19k9) o Gill net fishing was carried out on an experimental scale but only a few fish were caught. DESCRIPTICW OF GEAR The drift gill net is designed to catch pelagic fishes swim- ming close to the surface of the water. The depth at which the nets are to fish can be regulated by adjusting the float lines o In effects the net is an upright wall of netting made of a suitable material with an appropriate mesh to permit fish of a certain size to pass only part way through. Usually the fish is caught when the mesh catches bsnsath its gill covers and prsYents its swimming either forward or backvj-ardo Often when the fish strikes the net it may penetrate the mesh until it gets caught at the thicker part of the body, provided the girth is larger than the size of the mesh. The fish may also be caught by becoming entangled in the meshes without being gilled. The gill net (figo i) used in the present fishing experimentV for slcipjack in Hawaiian waters was constructed on the pattern of the albacore net used on tlie John N„ Cobb during its Ip^O testSo Eight to 111 "shackles" of net^, each measuring 50 fathoms in length and 100 meshes deep, were connected in a "string" at each settingo A l6-inch air-inflatad rubber float was attached .to the float line between shackles and at each end of the stringo A "sea anchor," made hj lashing together three or four rope fenders around a lead weight, was attached to the lead line at the free end of the first shackle; wb.ile a 60- to ?0-fathom bridle was attached to the float and lead lines of the last shackle o The lead line was kept from rolling up on the net by a 25-pound lead ball which was attached at the junction of tfco bridle and lead line^ D'uring night fishing buoy lights were attached to the l6~inch rubber floats at the center and both ends of the neto A total of l6 ohacklesj, eight of 8-ply salmon-lay nylon and eight of iiO/8-ply linen^ ^era constructed in stretched-mesh sizes CI U, 5j 6j ?j 9s lOj llj and 12 incheso The individual shackle (figo 2) was made up of tiie following parts s (1) Kettingo The netting was made of double-lcnctted nylon and linen X7hich measured 100 fathoms in length and 100 m.e3hes deep with double selvage » The -Septh in fathoms ranged from 3-1/3 for the u-inch mesh to l6-p- for the 12~inch mesho The 100-fathom netting was hung on a 50-fathom float line (2 si hanging ratio) ^ and was attached at every second mesh vrith UO/ll-ply linsn twine. (2) gloat Ixn?o Tt-^e float line consisted of SO fathoms of _, thread sof t-laid Manila, 17/32 inch in diameter, with an additional 20 inches at each end, (3) Floats, Tae 6-irich diameter round glass floats were tightly wrapped in l^-inch mesh webbing of No, 2? medium-laid cotton thread, Tnese float? were attached to the float line with 2 61 -thread hard-laid cotton straps 13 inches long at 2~fathom Tae mcdiiication and construction of the nets "-vere supervised by Kalfred Yea, formerly Fishery Methods and Equipment Special- ist, Pacific Oceanic Fishery Investigations, U, S, Fish and Wildlifs Service, ^■^^ < cr UJ O < o < Q. 6 < I CO < 9 > LU intervalso At first glass floats were used on all the shackleso Hcweverj upon discovering that these floats too easily became entangled in the larger meshesy they were replaced by 8 x 10=inch metal floats on liie nets of 10 and ll»inch mesh^ and the use of the 12 -inch mesh was discontinuedo (h) Lead lineo The lead line consisted of 50 fathoms of l8=thread soft=-laid Manila^ 15/32 inch in diameter^ with a 20-inch exten- sion at both endso Ihis was weighted with U-ounce seine leads spaced 1 fathom apart » ^^^ Breast linsn A breast line of 6=thread soft-laid Manila, J inch in diameter and equal in length to the depth of the net ting J was used at each end of the shackle = The ends of this line were spliced on to the float and lead lineso Straps of 26l-thread hard=°laid cotton line 10 inches in length were used to tie the shackles together and these were placed 1 fathom apart on the breast line^ All the linen nets were treated with net preservatives s the Uo 6, 9.» and ll^inch mesh sizes with Dura Nett 200^ an acid=free plastic preservative^ and the ^p 7s 10, and 12=inch mesh sizes with the New Lodge ProcesSj a combination of a plastic and copper napthenateo The nylon nets were first used without any preserva= tivBo After these nets failed to catch a single fish,, they were treated with preservatives, more for their staining than for their preserving effect, as the latter was unnecessary with the nylono The ij.^ 7s and 11 -inch mesh sizes were treated with the New Lodge Process J the 5s 6, and 10-inch mesh sizes with Net Life (green) (, a preservative with a soluble copper bases and the 9 and 12 -inch mesh sizes irvith Dura Nett 200, The preservatives increased the weight of the neos by 28 to 50 percent o The detailed account of the cost and the time consumed in completing the nets is shown in the following table £;o Table lo—Itemized cost of supplies and materialsV ^or the construction of 16 shackles of gill neto Item Quantity Cost/ unit! Cost linen netting 8 pieces $2 75 o 00 $2s200o00 Nylon net Ling 8 pieces 560.63 U,U85.oi; Cotton netting 52 Ibso 2o35 122o20 6-inch glass float kSO 0o32 lijlioOO 8 X 10-inch metal float 32 3o85 123o20 l6-inch rubber float 16 6eii5 I03o20 it -ounce lead seine sinker 200 IbSo 0o23 i;6oOO Bxxoy light h 15..00 6o„oo Buoy light bulb 18 Ooll lo98 Buoy light battery 8 2,50 2O0OO 6-thread soft-laid Manila rope 3 coils 12«60 37o80 18-thread soft-laid Manila rope li coils 3lio81; 139,36 2l4-thread soft-laid Manila rope h coils 3I4.8U 1390 36 UO/B-ply linen tmne 9 Ibso iio25 38,25 ilO/ll-ply linen twine 6 IbSo ho25 25.50 8'"ply nylon twine 20 Ibso 7o85 i5?oOO Dura Nett preservative (50=gal, drums) 2 driams 117,00 23iioOO New Lodge Process preservative C^O-gal, drijms) 2^ driim=> 37oOO 83o25 Net Life preservative (50-galo drums) Total i-1/3 drume 7?»00 96066 $8,257,60 1 ■' All prices are as delivered at Honoluiuj T, H, roller., hard-rubber V-rollerj and construction not included. Cbsts of the net of the net box are liable 2(a),,— Itemized construction time per shackle Man.=day.3 per shackle Description of work. With glass lloats With metal floats Attaching float line Attaching lead line Attaching breast line Attaching floats Treating luth preservatives . 5 5 1 it 1 2 5 5 1 i Total 13 12 (b)o—- Time required to complete entire net Item Noo Ifeji. =days/3hackle Total man- days Shackles with glass floats Shackles with metal floats 13 h 13 12 169 U8 Total man=- 5 UJ UJ -I UJ i^ X g£ ^ ?^t^' OlAj < rlco ■ X ^M! I o On the retrieve, the net was hauled in from the starboard side of the vessel o It was passed over the net roller; across a tablet, and through a V-ghaped roller on the wincho From the V= shaped roller it was passed into the net box, where it was replied in preparation for the next setting „ The entire procedure c, from the time the net was hauled out of the water until it came to rest in the net box^ was done in one continuous operation, stopping only to remove fish from the nets this was done on the table o The time required to haul in 10 to lij. shackles averaged 1-|- hourso The John Ro Manning fished only off the leeward coasts of Oahu, Kauai; and Hawaii since they presented the largest lee areas (figo 3)0 Off Oahu fishing operations were attempted between Maile and Kaena Point from 3 to l5 miles offshore o Off Kauai fishing was carried out in the area bet»Ksi Port Allen and Mana Point from 2 to Bi miles offshore „ At Hawaii, although the leeward area was very extensive ^ fishing was done only off Kealakekua Bay from 5 to 7 miles offshore (table 3)o Depth of waters at the time the net was set ranged from 200 to l^^TO fathoms, however.^ the net frequently drifted inshore to depths of 28'- 30 fathoms o Table 3o— Dates and areas fished by both vessels Vessel Date fished Hours fished Area Total catch (all species) Makua July 5-19 125 Hawaii 15 Manning Septo 26=29 5h^ Oahu h Octo 16=18 381 Oahu k Octo >5 39-3/U Kauai 2 Octo 11-12 26h Hawaii 2 The net fished a total of 159 hours from the Manning with only ij hours of this fishing time during daylight o This daylight set was made on a trial basis to observe the setting and retrieving operations, and to plan the retrieving procedure as efficiently as possible^ The catches of the two vessels are lis+ed m table ko It shoiild be noted that most of the tuna were taken in tne smaller meshes (6 inches and less) and that these con^jrised approximately 11 C H c a id fi s a ® o c g s B s i 1 p T (- 1 o m a « ffl iH d rH H r-\ c c H e S ^ •H H 3 O E! g r o iH o O 0 ^ o O o O O { to H =Q to iC-. XI •9 1 rH C^ CV « H CO CV CO CO (D 1 Wi 21 Xt raj M' aj| (0 03j i ^ N ^ ^ f-^ ■^1 6 s gg P« (Ql CO (D m CO p 1^ R rHo rH s^ , «H p «a a> °dl g| (D u 1 ■H 4> 9) Q 21U-.. O -p oj! j3 -« ^ s "Hi s CO « aS nj ^il (Q i- o m 03 en P •r* xa a 0 3 4 >» u 2S fl Ch H >! S a) iH ^ CJ J 0 E M Ci i CO <0 (C ^ eol O O U 03 J a ca «l! a o(^^ W MF w w! OT M fa <; ■^^g •g "gg "g •gg ag 00 OA K cS sD nrS -# CD kO, ffl ID ■p -*3 -P ■€ •i* •*» +' 5 •a ^ •§, 00 -8. •a >» "g, >» >» >>, EH "H SJ "H H '►i sd m °H cS (0 « S &s a % as K n S3 n Q « ■p a) H v£) ^ ^ ^ O H J^ rH to rH ON H O [ rH a ' 5 -^ -;:> ^N^, "^ ^v^ \ •^ c^ C^ i> O t> O - u > c •H rH ^ >> ^ 0) CO 1 H C » 0 fi rH is; H ^ 5 5 H O 0 o 0 O 0 o 0 O g 0 0 0 TJ C3 CO ^ ^ 5 ir> c- ITs -t « CO R H B ■P -p -p •P -P 43 0 iH Oi O f" ■^ :i 3 3 s 3 to O S 0 o o o O o o 0 0 0 c »A . x> ^ ^ «a x> 4J iH (H UOi r-i CV cn as CO CO CO 00 § iS -g CO 1 ^ •5 ^ m ^ g i oi CO 03 s n J3 O (0 £ . 4D .^ ffl ^ ^ ^ t m 4> +» d fl CO •a cy CO *=*"^ b OT « -P •O 0 -P g •H 03 C iH ^ P< c, :d ® ro c o O. G , 0 C ^ s p. o s ffl H 3 3 (fl g >t SH a, o s s >» ^ +5 P CO ! as ^-■ri d CO a CO CD yi-l s of g m s ^ ehI; Pt< s Q CO W i-q Wil m CO S S ^ E-i S SI 5 3 H 0] jcj 3 § •H O 3 CO j3 "H CO ■3 o ^ «> J«! (0 +3 o. O ■♦-» ® ^ « a> 1 cft C a 0 ^ § ® o H 1^ ^ q* 0^ fr, o CD 0 i c3 W «M ^ 0) ^ ^ ce CO >H Snt 0 CO 1-1 ««-l 2.' & s «H « %^ »H & ^ o O O jS s & o 3 O "H S & c £ CO ►^ •H e CO ^ ^ (Q W ■^ ■^ d E-) ^ B O M ^ ^ ( HCM >a c^ Jhl OJ S Hn CO o Hos CO sn CO D (0 -* sH (H -* f^W Oi P< «"- n H m •^m - ■' •+' g J3 W s ^ s g; s s s s s s e s r-i Si o ^ S5 *A M "^ v^ ^ vO vO nO !>• S sO e^ O to O c* in J~* rH J-f rH •z rv jT^ ■1^ ^ ^ "->, ^v. ^-. ^>^. ^^ ^v. '^ *^, CT <« ^\ ^^ ^>. '^ ^^ O O O o 0 0 0 0 J a o o C^ ON ON r-\ rH r-i „ H _ rH r-i rH rH 13 half the total amount of netc Ihe total hours spent in fishing by both vessels wa'? 29U, with a total of 196 shackles set.. The catch per unit of effort was Ool3 fish including 0 OU tunas per shacklec. but if allowance is made for half of the mesh being too large, the catch per shackle is double that indicatedo Several factors affected the fishing operations., the most noticeable of which were wind,, current, and sea conditLonso Although the upper limits of these factors were not investigated thoroughly <■ it was possible to fish with the wind velocity up to 30 kno&So TUB-nds of this force caused heavy strain on the gear and e^ecially on the line by which the net was secured to the vessels The strain was somewhat reduced by the addition of more line; Strong currents were a cons^ant source of concerns Although the net was usually set across or obliquely to the direction of the cxrrrent, it often swung arcund to a position parallel to the current during xho set-. In localities where ths corrent changed its direction of flow several times during the night j the net folded upon itself ^ thus pressiting only a part of its laigth to actuail fishingc. Drifting over 15 miles during the night was not uncommon^ and in several instances where the current was extremely rapid^ the net drifted into the wind e^^-en though the wind velocity exceeded 28 knots. Tne current also affected the vertical angle of the neto During one of the cross=current sets, a current of approximately 1/2 knot caused the net to fish at a 35=^0 degree angle with the s^^rface. This was prob- ably due to the shear between the surface and the lower water layer j or to the effect of the wind acting on the floats in a direction opposite to the ciKrentg or possibly to a combination of bothc However, the velocity of the wind was only about 10 knots on this particular seto Sea conditions affected the retrieving operation greatly^ Ihe ship's roll, due to large swells j caused great strain on the net, and often this resulted in nuaierous tears in the webbing alcng the lead arrL float lines , The Manrang worked in swells up to 7 feet in height, A smaller vessel probably could have performed this operation in much rougher water with less damage to the gear^ It was evident from the first few sets that the net could only fish safely from the Manning in areas of calm water o FiaiING (SOUNDS AND SEASCWS The observations made on the fishing areas showed that in the Hawaiian Islands caJa waters are found only in the leeward areas of the islands cf Hawaii, Oahu,- and Kauai o ttily at twc islands.; Hawaii and Oahu^ dees this lee extend much beyond 15 miles from shore » At Kauai it extends only L cr 5 mileSo Ihe lee at Mauij, although it is Ih relatively calnij is broken up into several large channels swept by strong currents^ conseqaently it is not feasible to use the gill net in this area. Drift net fxshing can usually be done with safety to the gear in the calmer waters bordering Hawaii. Oahu, and Kauai during the entire year and during the sunmer off the islands of Lanai and Molokaio Along these leeward areas the wind shifts from NE to Sff at night and back to HE as mornajig approaches,, Ihe general pattern at all the islands is nearly identical^ but the ext-ent of this shift varies greatly between islands depending on the size and formation of the mo\mtainso Ihe velocity., too; often vanes from 0 to 30 knots during the night o Ihe wind by itself does not seriously hanger the opera- tions j but when it occurs simultaneously with strong currents and adverse sea conditions j fishing becomes very difficult. Ihe currents themselves are also variable both with respect to time and intensity at the various localities and even within the same locality: Of two sets made on September 26 and 27 in approximately the same locality, the first set drifted mor-e than 15 miles while the latter drifted only about 7 miles and somewhat obliquely to the drift of the first sets Consequently, the allowance for the drift, the prevailing leeward wind velocity and direction; and the anticipated morning sea conditions tend to circumscribe the area of operationo The numerous small craft and barges plying between the island? and the activities of commercial and sport fishing vessels at times make the operations hazardousc Hovrevers this situation could be overcome by fishing the net deeper; if the gill net should be used commercially. The fishing seasons, as determined from the skipjack landings by the live bait fishery, show considerable changes during the year and from season to season Cfi-go k) - ^le first sizable catches of skip= jack by the commercial fishery are usually made in kay: After increas- ing to a maximum m J'ane or July^ the catches drop off gradually until October or November, small catches, however p are made in most years throughout the yearc The John E, L^anning fished extensively off the leeward coast of Oahu in the Tfaianae-Kaena Point area, the most productive skipjack ground in the Hawaiian Islands, during a period when fish should have been fairly abundant (tables 3 and 5)= The occurrence of skipjack schools was recorded regularly on this cruise as follows = During a period of 17 days with an average of 5 hours per day spent in scout= ing for tuna, a total of 12 fish schools were sighted under bird flocks and one school without any accompanying birds. Nine flocks of birds were also sighted, and although no fish were seenc the types and behavior of the birds indicated the presence of skipjacka Half of the schools were sighted in open water where it was inipossible to 15 JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUS SEP OCT NOV DEC FIG 4. HAWAIIAN SKIPJACK LANDINGS BY MONTHS, 1945-1951 16 o m m 0 u (3 0) A ■>t to oa O. G) 1 g tn 2