FISHERIES OCEANOGRAPHY Fish. Oceunogr. 3:2, 133-141, 1994

Relationships among yellowfin and skipjack tuna, their prey-fish and plankton in the tropical western Indian Ocean

CLAUDE ROGER ORSTOM (Institut Frunsds de Recherche Scientifique pour le Diveloppement en Coopirution), B. P. 5045, 34032 Montpellier cedex 1, Frunce

ABSTRACT 1986).Nevertheless, they live in the upper layer (0-200 m or so) of tropical seas (Yuen, 1970; Dizon et al., 1978; Stomach contents of yellowfin (ThzmniG albacares) and Yang and Gong, 1987; Cayré, 1991)) which is “one of skipjack (Katsuwonus pelamis) tuna caught by trolling the most unproductive environments” (Sund et al., and purse seining in the tropical westem Indban Ocean, 1981). These well-known facts lead to the conclusion together with those of the prey-fish found in their that food availability is a key factor in determining their stomachs, have been analysed. Epipelagic fish are the abundance and distribution (Blackbum, 1969; Sund et main prey of these tunas, whereas no vertically migrat- ul., 1981; Petit, 1991; Stretta, 1991). ing fish, which inhabit subsurface layers at night, have The aim of this study is to describe the successive been found in their stomachs. These tunas are thus links of a food chain culminating with these tunas in considered day-feeders. Purse-seine-caught tunas, this region, by analysing stomach contents of tunas which belong to large schools, have a much higher together with those of the prey found in their stomachs. number of prey-fish in their stomachs than tunas caught Among the prey of these tunas (fish, cephalopods, by trolling on small schools. Similarly, prey-fish from crustaceans), only fish are suitable for reliable stomach purse-seine tunas have a much higher number of plank- contents analysis. The first step of the study is thus to tonic prey in their stomachs than those from troll- check whether prey-fish represent the main prey of caught tunas. Therefore, these tunas adopt a wandering these tunas so as to ensure that the chain plankton + strategy in small schools when food resources are scarce prey-fish is prominent in tuna feeding. The second step and form large schools when they are abundant. The is to describe the plankton used as food by these prey- planktonic organisms found in the stomachs of prey-fish fish, in order to identify the fraction of the planktonic are described by taxa and sizes; they represent the biomass actually supporting this food chain. fraction of the planktonic biomass actually supporting the stock of tuna. Size ratios between the three links tuna-prey-fish-plankton are very high, suggesting that METHODS AND MATERIAL these tunas benefit from a short food chain which is Tunas were obtained both from fishing carried out probably efficient from the energetic point of view. during the Indian Ocean Tuna Programme and from commercial purse seiners. All fishing operations were Key words: feeding, trophic relationships, Indian carried out during daylight. Working areas are indicated Ocean, yellowfin, skipjack tuna on Fig. 1. In the Mozambique area, tunas originated both from INTRODUCTION purse-seine fishing which was carried out when large schools were encountered, and from trolling on schools Yellowfin (YFT, Thunnus albacures) and skipjack (SKJ, too small to justify the use of the purse seine. In the Katsuwonlu pelurnis) tuna are very active pelagic fish Seychelles area, all samples were obtained from com- which require a large amount of energy (Kitchell et al., mercial purse seiners fishing on large schools. Troll- 1978; Crowder and Magnuson, 1981; Olson and Boggs, caught yellowfin amounted to 65 individuals, among which 5 had empty stomachs and 15 contained only unidentifiable remains; the figures are 75, 27 and 10 Received for publication 25 May 1993 respectively for skipjack. No such information is avail- Accepted for publication 1 December 1993 able for purse-seine-caught tunas, as only non-empty 01993 Bklckwell Scientific Publications, Inc. stomachs were sent to the laboratory. 9 NOV. 9994 133

6’ i

!

40 50 80 7C 20 90 100 110 120 120 140 150 - Fork length- llOcm classesi- LJ SKJ Troll Moi _. SKJ Seine Ulcz - SKJ Seine Ser

'iFT TïGli Moi -t- 'iFT Seine Sep Feeding of trupicul surface tiom 135

RESULTS Their mean length is in the range 32-42 mm and almost all of them belong to the family Engraulidae and are Stornuch contents of tunas probably of the same species. In the Mozambique area, 5 A preliminary observation should be reported. When skipjack out of 16 have fed in addition on large-sized purse-seine fishing was operated around floating logs prey-fish (mean length, 129 mm) such as flying fish, before sunrise, almost all tunas had empty stomachs. In juvenile tunas, Paralepididae, Dipluspinia sp., and the Mozambique area, 5 19 tunas from 10 different net Hemirunzphus sp. Only 14 prey-fish out of 13 I7 belonged deployments were examined; all of them, with the to this large-sized category, but their importance in exception of two yellowfins, had empty stomachs. In volume is not negligible. It should be noted that all the Seychelles area, more than 1300 tunas were exam- prey-fish are epipelagic. ined; 99% of them had empty stomachs (Stequert, pers. The stomach contents of tunas caught by trolling in comm.). This fact leads to the conclusion that these the Mozambique area showed some ditierences from tunas do not feed by night and/or do not find any food those caught by purse seine. Fish is the dominant prey, under floating material such as logs. but crustaceans account for 9% of the stomach volume The analysis of the 119 stomach contents originating in skipjack and 24% in yellowfin. Most of these crus- from trolling and purse seining carried out during day- taceans are stomatopod larvae and amphipods; small light leads to the following results. numbers of carids and niegalopa larvae have also been All tunas caught by purse seine feed almost exclus- found. Small cephalopods (mean mantle length 24 mm, ively on fish, both in the Mozambique area as in the standard deviation 19 mm) represent 46% of the Seychelles area (Table 2, Fig. 3). Prey-fish occur in stomach content volume. All prey-fish are epipelagic, large numbers in each tuna stomach: means of 53, 82 with the exception of 6 (out of 319 prey-fish) mesopela- and 135 according to species and area (Table 3, Fig. 4). gic fish in skipjack stomachs. Only 24 (out of 1122 prey-

Table 2. Stomach contents of 119 tunas.

Prey type

Fish Crustaceans Cephalopods

96 96 % Yo Yá ‘16 Fishing method and area Predator Occurrence Volume Occurrence Volume Occurrence Vulume

Trolling in Mozambique area Skipjack 100 87 47 9 16 4 Yellowfin 100 70 72 24 41 6 Purse seine in Mozambique area Skipjack 100 99 13 Trace 6 Trace Purse seine in Seychelles area Skipjack 100 99 21 Trace 7 Trace Yellowfin 100 96 67 3 33 Trace

Table 3. Number of prey-fish per stomach of tuna.

Predator

Skipjack Y ellowfin

Mean number Standard Mean number Standard Fishing method and area of prey deviation Range uf ptey deviation Range

Trolling in Mozambique area 8.4 16.8 1-100 18.5 44.8 1-300 Purse seine in Mozambique area 82.3 76.0 3-240 - - - Purse seine in Seychelles area 53.1 24.0 24-94 135.5 78.7 56-273 Prey of Troll-caught tunas Mozambique area (March-Sept.1989)

6 1 Skipjack Yellowfin

Prey of Purse-seine tunas Seychelles area (Jan.Bct.Dec.1990) T 53i

Skipjack Yellowfin 1 SIiJ Troll..lioz. 2 YFT 3oll.Lioz :3.SIid Seine Se?. 1 SKJ Sein.: hloz 5.YFT Seine %y Prey of Purse-seine Skipjack Mozambique area (May-July 1989) I, I Feeding of tropical sudace tima 137

l l Table 4. Type and size of prey-fish.

Mozambique area Seychelles area Trolling Purse seine (purse seine)

Skipjack Yellowfin Skipjack Skipjack Yellowfin

Number of stomachs of tunas analysed (total 119) 38 45 16 14 6 Number of prey-fish (total 4025) 3 19 833 1317 743 813 Type and size of prey-fish Mesopelagic" Number Occurrence (%) Mean size (mm) Large-sizedh Number 3 21 14 O O Occurrence (%) 5 22 31 - - Mean size (mm) 95 139 129 - - Other Number 310' 812" 1303" 743¿ 813" Occurrence (Yo) 100 93 81 1O0 100 Mean size (mm) 27 (a= II) 25 (a= 13) 42 (a= 11) 32 40

"5 Mgctophidne and 1 Stemupyx. "Flying fish, juvenile tunas, Paralepididae, Dipluspintu sp., Hemiramphils sp. 'Various epipelagic species: Balistidae, Dkddontidae, Ostraciunidar, Nomeidne, Engrddidae etc. 'YY% Engraulidae.

also in volume as judged from visual examination under DISCUSSION AND CONCLUSIONS binocular microscope. Other crustaceans (amphipods, ostracods, Lucifer, euphausiids and decapods) occur Fish are by far the dominant prey of surface yellowfin frequently but their quantitative importance is low. and skipjack tuna caught by trolling and purse seining in Other prey as chaetognaths, annelids and fish larvae are this region. Nearly all the prey-fish are epipelagic since also of secondary importance. Only in prey-fish sto- almost no vertically migrating fish, such as Myctophi- machs originating from yellowfin caught by trolling in dae, which rise by night in subsurface layers, were the Mozambique area, are prey other than copepods found. Many other studies have reported similar obser- significant. vations (Nakamura, 1965; Dragovitch, 1970; Drago- Lengths of copepods ranged from 0.5 to 4 mm (Fig. vitch and Potthoff, 1972; Legand et al., 1972; Grand- 6). Other crustaceans had a mean length of 4.6 mm and Perrin, 1975; Borodulina, 1982; Pelczarski, 1988; and a maximum length usually under 10 mm. Most chaetog- others) which support the generally accepted opinion naths and annelids were less than 30 mm and fish larvae that these tunas are day-feeders. less than 15 mm in length. These taxa and sizes depict Tunas fished by purse seine (therefore belonging to the plankton which supports this tuna food-chain. large schools) feed heavily on engraulid concentrations, Mean numbers of planktonic prey per stomach of and large schools of prey-fish are associated with large prey-fish are illustrated in Fig. 7(A). Important differ- schools of tuna. Tunas fished by trolling on small ences are obvious between trolling material (23.6 and schools feed to a larger extent on crustaceans and 16.4 planktonic prey per prey-fish stomach) and purse- cephalopods and contain fewer prey-fish per stomach seine material (51.2, 130.3 and 94.7 planktonic prey and these are from a wider range of taxonomic groups. per prey-fish stomach). Obviously these tunas are in search of richer areas and in 4L' 353

'11.Y Ht) i '44h 11 ni3 54.4 ni. 3 L17.5 33.h o. 9; 1 .o1

14.3 l2.i I5 34 h. 1 5.-5 h '+n ti.- 1 X.ii I I Ih

33.3 42.3 11 3 ;O 4. I i.6 (5.3) n. I 4

0 0

3.0 I 1O.J - 1510 51.7 i;6.31 Feeding of tropicul surface tunu 139

Figure 5. Lengths of the 2018 prey-fish found in the 99 tunas Figure 6. Lengths of copepods found in the stomachs of caught in the Mozambique area (not including 38 large-sized prey-fish. See legend of Fig. 2 for key. prey-fish, 95-139 mm, see Table 4). See legend of Fig. 2 for 70?6 in numbers keg. Numbers , 350 I

250

200

150

100

50

O 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 O 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 Sizes (0.5” classes)

Sizes of prey-fish (5” classes) * SKJ TroLMoz. -13- YFT TrolLMoz. SKJ Seine hioz. -?$; SKJ Seine Sey. -e- YFT Seine Sey. + Mean Troll-caught SKJ = Troll-caught YFT 0 Seine-caught SKJ Total Table 6 reports on mean lengths of tunas, their prey- fish and the copepod prey of the latter. The length ratios 1974; Roger and Grandperrin, 1976). According to this between these three links attain very high values. It view, the biomass of plankton supporting this tuna thus appears that yellowfin and skipjack tuna here take food-chain would thus comprise the above cited organ- advantage of a short food-chain which is probably isms which remain by day in the 0-200 m layer. The efficient from the energetic point of view. biomass and taxonomic composition of this fraction of By multiplying the mean number of prey-fish per tuna the zooplankton is markedly different from that of the stomach by the mean number of planktonic prey per zooplankton as a whole (Roger, 1982, 1986). prey-fish stomach (Table 7), we obtain the mean num-

? Table 6. Length ratios between tunas, prey-fish and copepods.

1 2 3 Tuna, method Mean length Mean length Ratio Mean length Ratio Ratio and area of tuna (cm) of prey-fish (mm)” 1: 2 of copepods (mm) 2 :3 1: 3

Skipjack, trolling, Mozambique 51.0 27 19 0.92 29 554 Yellowfin, trolling, Mozambique 70.3 25 28 1.31 19 53 7 Skipjack, purse seine, Mozambique 60.2 42 14 0.97 43 621 Skipjack, purse seine, Seychelles 58.1’ 32 18 0.81 40 717 Yellowfin, purse seine, Seychelles 125.7 40 31 1.05 38 1197

I “The 38 (out of4025) large-siied prey-fish are excluded. 140 C'. RIIXL'I

Table 7. Niinibers titprey per

Tuna, methnd Numher I it" prey-tkh Nunihrr of plnntmic Numhrr (.if planktonic md aren per tuna stt-ininch prey per prey-fih \tc imach prey per rima stnm:ich

Skipjack, tnillinr, hiimmhiqrie S.4 13.h 1% YelLin fin, milling, h1ii:nmhiytie 18.5 Ih.4 303 Skipjxk, pi1r.c: seine, htoxmhique 81.1 51.' 4714 Skipidq pirw .;eine, Seychelles 53.1 I 3d. 3 h91'1 l'ellciivfin~ptirw xine. Scychclles 135.5 "4.7 1'831

when teeding cunditinns Lire poor (tmprey-tidi for tuna md few planktonic prey kir prey-fshl and torni lxge \chcicils m.1ii.n teejinq ccinjitiiins are gciod. U5in.g ;i 140 , 14 cc ,niplctely &fierent str;ite.q in their research, hnsecl on theorericd vptinirim energetic yield, Petit ( l'NI ) ;tnd Strett;ì ( 19Q1) reachd the ,same concluaion.

ACKNOWLEDGEMENTS Thanks are due to the crew ot thc purse-seiner hl~simji tor their exzellent cooperation, tii my cc~lleague~from La Riunicm and h1ndag;rscx (CNRO) {Or pitrticipatinp in the sampling qvrarions find to ni)- collrag~iesfrom Seychelles md hlauritius for providing tiinit .tomachs troni commcrci,il purse-seinerh.

1 7 3 i 5 TROLLING MOZ. SEINE W(1. SEINE PEY 1:SKJ Troll.hloz. ?.TFT Troll.híoz :3.X Seine Iloz. 4 SKJ Seine Sey 5 YFT Seine Sey. Feeding of tropical sicrface tmu 141

de longdigne duru le Pacifique Sud Ouest tropicul. These de pkheries thonieres tropicales de stuface (aspect &diutif), Thèse Doctorat $Etat, Universite de Provence, Marseille. 296 pp. Universite Paris VI. 129 pp. Kitchell, J.F., Neill, W. H.,Dizon, A. E. and Magnuson, J. J. Roger, C. (1974) Les euphaicsiacés du Pucifique de'quatorial et snd- (1978) Bioenergetic spectra of skipjack and yellowfin tunas. tropicul: zoogéographie, icologie, biologie et sitirution trophique. In: The Physiologicul Ecology of Tutm. G. Sharp and A. Dizon Thèse de Doctorat d'Etat, Université de Provence, Marseille. (eds). New York: Academic Press, pp. 357-368. Mémoires ORSTOM 71: 265 pp. Legand, M., Bourret, P., Fourmanoir, P., Grandperrin, R., Roger, C. (1982) Macroplancton et micronecton de l'Atlantique Gueredrat, J. A., Michel, A., Rancurel, P.,Repelin, R. and tropical I: Biomasses et composition taxonomique. Oc&". Roger, C. (1972) Relations trophiques et repartitions verti- Trop. 17(1):85-96. cales en milieu pélagique dan l'océan Pacifique intertropical. Roger, C. (1986) Macroplancton et micronecton dans le Pacifi- Cah. ORSTOM, Sir Ociunogr. X(4):303-393. que tropical sud-ouest. Océanogr. Trop. 2 l(2):153-165. Nakamura, E. L. (1965) Food and feeding habits of skipjack tuna Roger, C. and Grandperrin, R. (1976) Pelagic food webs in the (Kutsuzc~on~cspeluniis) from the Marquesas and Tuamotu tropical Pacific. Limnol. Ocennogr. 21:731-735. Islands. Trans. ,4m. Fish. Soc. 94(3):236-242. Stretta, J. M. (1991) Contribution de la tr'lde'dde'tectionaérosputiale d Olson, R. J. and Boggs, C. H.(1986) Apex predation by yellowfin l'éluboration des buses de l'lulierttiqiuopérationnelle: l'exemple des tuna (Thitnnus albacares): independent estimates from gastric pêcheries thoni2res tropicules de surfuce (aspects pr6dictifs). Thèse evacuation and stomach contents, bioenergetics and cesium de Doctorat d'Etat, Université Paris VI. 125 pp. concentrations. Can. J. Fish. Aquat. Sci. 43:1760-1775. Sund, P. N.,Blackbum, M. and Williams, F. (1981) Tunas and Parin, N.V. (1968) Ichthyofauna of the epipelagic zone. Truns- their environment in the Pacific Ocean: a review. Oceanogr. luted by Israel Progrum for Scientific Translations 1970, IPST Mur. Biol. 19:443-512. cat.nO5528: 206 pp. Yang, W.S. and Gong, Y. (1987) The vertical distribution of Pelczarski, W. (1988) Examination of food of yellowfin tuna tunas and billfishes and fishing efficiency between Korean (Thunnus albucares) and bigeye tuna (Thimnics obessw) from regular and deep longlines in the Atlantic ocean. ICCAT the open waters of the Central Atlantic. ICC.4T XXV1,SCRS 86/81:184-187. XXVII1,SCRS 87/55:5&73. Yuen, H. S.H. (1970) Behaviour of skipjack tuna, Katsawonus Petit, M. (1991) Contribution de la télédde'tection oérosputiak d pelamis, as determined by tracking with ultrasonic devices. J. L'r'labmation des buses de l'hulieutiqiu oph&mwlle: L'exemple des Fish. Res. Bd Can. 27:2071-2079.