AN ASSESSMENT OF THE SKIPJACK AND BAITFISH RESOURCES OF SOLOMON ISLANDS

A.W. Argue and R.E. Kearney

Skipjack Survey and Assessment Programme Final Country Report No. 3

South Pacific Commission Noumea, New Caledonia November 1982 ^fv^ Hf*.

AN ASSESSMENT OF THE SKIPJACK AND BAITFISH RESOURCES OF SOLOMON ISLANDS

A.W. Argue and R.E. Kearney

Skipjack Survey and Assessment Programme Final Country Report No.3

South Pacific Commission Noumea, New Caledonia November 1982 iii

PREFACE

The Skipjack Survey and Assessment Programme was an externally funded part of the work programme of the South Pacific Commission. Governments which provided funding for the Programme were Australia, France, Japan, New Zealand, United Kingdom and the United States of America.

The Skipjack Programme has been succeeded by the Tuna and Billfish Programme which is receiving funding from Australia, France, New Zealand and the United States of America. The Tuna Programme is designed to improve understanding of the status of the stocks of commercially important tuna and billfish species in the region. Publication of final results from the Skipjack Programme is continuing under the Tuna Programme. Papers referred to as manuscripts in this final country report will be released over the duration of the Tuna Programme.

The staff of the Programme at the time of preparation of this report comprised the Programme Co-ordinator, R.E. Kearney, Research Scientists, A.W. Argue, C.P. Ellway, R. Farman, R.D. Gillett, P.M. Kleiber, W.A. Smith and M.J. Williams; Research Assistants, Susan Van Lopik and Veronica van Kouwen; and Programme Secretary, Carol Moulin.

The Skipjack Programme is indebted to the people of Solomon Islands, particularly the officials of the Fisheries Division in the Ministry of Natural Resources, for making its stay more productive and enjoyable.

For supplying catch statistics, sampling statistics and tag returns from the Solomon Islands fishery, the authors wish to thank Doug Gibson, Chief Fisheries Officer, David Evans, Fishery Biologist, and Christine Mansfield, Biometrician, of the Solomon Islands Ministry of Natural Resources, Honiara.

Tuna Programme South Pacific Commission V

CONTENTS

Page PREFACE iii

LIST OF TABLES vii

LIST OF FIGURES ix

1.0 INTRODUCTION 1

1.1 Background to the Tuna Fishery 1

2.0 SKIPJACK PROGRAMME RESEARCH PLAN 6

3.0 VESSEL AND CREW 6 4.0 METHODS 6 4.1 Skipjack Fishing, Tagging and Biological 7 Sampling 4.2 Baitfishing 8 4.3 Data Compilation and Processing 8 4.4 Data Analysis 8 5.0 SUMMARY OF FIELD ACTIVITIES 8'

6.0 RESULTS AND DISCUSSION 12 6.1 Baitfishing 12 6.1.1 General observations 12 6.1.2 The Slot and vicinity 19 6.1.3 Other baiting localities 22 6.2 Tuna Fishing Results 23 6.2.1 General observations 23 6.2.2 Ontong Java 25 6.2.3 Santa Cruz Islands 26 6.3 Skipjack Population Biology 27 6.3.1 Maturity and juvenile recruitment 27 6.3.2 Skipjack diet 28 6.3.3 Skipjack growth 29 6.3.4 Population structure 30 6.4 Skipjack Tagging Results 32 6.4.1 Solomon Islands tagging 32 6.4.2 International migrations from 35 Solomon Islands 6.4.3 International migrations to 37 Solomon Islands 6.4.4 Mortality, production and fishery 42 interactions 6.5 Yellowfin Biology and Tagging Results 50

7.0 CONCLUSIONS 52 7.1 Baitfish 52 7.2 Skipjack 53 REFERENCES 55 VI

APPENDICES A. Scientists, observers and crew on board the research 61 vessels

B. Bait species composition, percentage of bouki-ami hauls containing a particular species, and estimated total 63 catch, for the 1977 and 1980 Skipjack Programme surveys in the waters of Solomon Islands

C. Listing of tag and recovery data for each tagged skipjack and yellowfin that made an international migration out 65 of or into the waters of Solomon Islands

D. Stomach contents of yellowfin sampled by the Skipjack 73 Programme in the waters of Solomon Islands vii

LIST OF TABLES

Table Page

1 Annual tuna and baitfish catch and effort statistics for the locally based pole-and-1ine fishery in the 3 waters of Solomon Islands

2 Summary of daily field activities in the waters of 9 Solomon Islands

3 Summary of Skipjack Programme baitfishing effort and 10 catch in the waters of Solomon Islands

4 Summary of numbers of fish sampled for biological data 13 from the waters of Solomon Islands

5 Stomach contents of skipjack sampled in the waters of 16 Solomon Islands

6 Incidence of tuna juveniles in the stomachs of skipjack and other predator species from the waters of Solomon 17 Islands

7 The dominant ten species of baitfish caught with bouki-ami gear by the Skipjack Programme in Solomon 18 Islands

8 A comparison of Skipjack Programme baitfish catches for repeat visits to the same baiting localities in Solomon 20 Islands

9 Average bait catch and species composition for baiting areas fished by the Skipjack Programme in the vicinity 21 of The Slot, Santa Cruz Islands and Ontong Java atoll

10 Catch per fishing day for the Solomon Islands pole-and-line fleet in November and June, 1976 to 1980, „ and for the Skipjack Programme tagging vessel in November 1977 and June 1980 in the commercial fishing area

11 Tuna school sightings per hour per fishing day and total catches per fishing day for both surveys by the 24 Skipjack Programme research vessel in the waters of Solomon Islands

12 Indices of fishing performance for tuna schools fished by the Skipjack Programme research vessels during 1977 26 and 1980 surveys in the waters of Solomon Islands

13 Estimates of skipjack growth rates for several countries in the SPC region, by size at release and 29 time-at-large vm

14 Releases of tagged skipjack during 1977 and 1980 in the waters of Solomon Islands, numbers of recoveries made 33 inside and outside the Solomon Islands fishery zone, and numbers of recoveries from unknown recovery locations

15 Numbers of skipjack tagged, total recovered and percentage of tags recovered for sub-areas in the J^ waters of Solomon Islands

16 International recoveries of tagged skipjack for 35 subdivisions of the Solomon Islands tagging area

17 Monthly skipjack tag recoveries by all fishing gears in the waters of Solomon Islands from tag releases in •" other countries

18 Percentage of tag recoveries in the waters of Solomon Islands from releases in Papua New Guinea at locations 40 increasingly distant from the Solomon Islands fishery

19 A comparison of average length of skipjack in the Solomon Islands pole-and-line catch with length of 41 tagged skipjack (recovered in Solomon Islands) at time of tagging and at time of recovery

20 Monthly tag returns, catch and effort for the Solomon 43 Islands pole-and-line fishery

21 Estimates of skipjack population size, turnover rate, throughput and catchability coefficients for the 46 Solomon Islands pole-and-line fishery

22 Fishery interaction results between Solomon Islands and 49 Papua New Guinea

23 Estimated contribution to Solomon Islands throughput by immigrants from tagging locations outside the Solomon 50 Islands fishing area

24 Numbers of yellowfin tagged, total recovered and percentage of tags recovered for sub-areas in the 51 waters of Solomon Islands IX

LIST OF FIGURES

Figure Page

The area of the South Pacific Commission Inside front cover

Straight line representations of movements of skipjack Inside back tagged by the Skipjack Programme and subsequently cover recovered

Survey area and baitfishing localities for the 1977 and 2 1980 Skipjack Programme surveys in the waters of Solomon Islands

Monthly tuna catch by Solomon Islands pole-and-line 5 vessels, 1976 to 1981

Length frequency distributions for tagged and sampled 14 skipjack and tagged and sampled yellowfin for both Skipjack Programme surveys in the waters of Solomon Islands

Length frequency distribution of female skipjack from 15 Solomon Islands biological samples and distribution of maturity stages of female skipjack Skipjack school serum esterase gene frequency versus 31 longitude of the sample location

Length frequency distributions for sampled skipjack 34 from the Solomon Islands pole-and-line catch during the month of tagging, contrasted with size frequency distributions of tagged skipjack that were released in the commercial fishing area

7 Migration arrows for 18 of 23 Solomon Islands tagged 36 skipjack that made international migrations

8 Migration arrows for 41 of 77 tagged skipjack that made 38 international migrations to the waters of Solomon Islands

9 Relationship between tag recovery rate by Solomon 40 Islands pole-and-line vessels and average distance travelled by tagged skipjack originating from countries (areas) external to the Solomon Islands fishing area

10 Length frequency distributions for skipjack sampled 42 from the June 1980 commercial catch and for tagged skipjack (one school excluded) from the SPC vessel catch, June 1980

11 Solomon Islands monthly skipjack tag returns from Table 45 20, and estimated monthly tag returns from the fitted tag attrition model X

12 Length frequency distributions for tagged yellowfin 52 released during 1977 and 1980 in the waters of Solomon Islands AN ASSESSMENT OF THE SKIPJACK AND BAITFISH RESOURCES OF SOLOMON ISLANDS

1.0 INTRODUCTION

The Solomon Islands pole-and-line fishery has developed rapidly in the last twelve years. With a present annual catch of more than 20,000 tonnes, predominantly skipjack (Katsuwonus pe1amis), it now is one of Solomon Islands' three leading industries, as well as the country's major source of foreign exchange accounting for over 23 million Solomon Island dollars of exports in 1980 (Anon 1981a). Clearly the health of this fishery and its interaction with other regional fisheries are of paramount importance to the Government of Solomon Islands. It was concerns such as these, which were shared by many countries in the region, that led to development of the South Pacific Commission's Skipjack Survey and Assessment Programme.

While in the waters of Solomon Islands it was the purpose of the Skipjack Programme to assess the status of the skipjack stock through tagging and biological sampling in both fished and unfished areas, and to enlarge on the Solomon Islands baitfish data base, in particular to assist with species identification in fished areas and to conduct exploratory fishing in remote areas. Three years of tagging over a broad area of the central and western Pacific, and amongst all locally based fishing fleets, provided the data necessary to assess fishery interactions, such as that between Solomon Islands and Papua New Guinea pole-and-line fisheries.

The Skipjack Programme conducted two surveys in the waters of Solomon Islands. The first, from 1 November to 4 December 1977, followed the Programme's original survey in the waters of Papua New Guinea. Preliminary results for the first Solomon Islands survey were reported in Kearney and Lewis (1978). The timing of the second survey, from 25 May to 28 June 1980, allowed for examination of seasonal effects between years. The surveys ranged from Ontong Java atoll in the north to the Santa Cruz Islands in the south and included considerable time in The Slot and nearby waters (Figure 1). This report presents final results from both surveys and discusses management implications of these findings.

1.1 Background to the Tuna Fishery

The fishery zone of Solomon Islands is seventh largest in the South Pacific Commission (SPC) region, encompassing approximately 1.3 million square kilometres (Sevele and Bollard 1979), much of it enclosing productive oceanic fishing grounds. The population of Solomon Islands, approximately 225,000 in 1980 (South Pacific Commission, unpublished data), is spread over seven provinces and 28,530 square kilometres of land area. The population is expected to increase to 435,000 by the year 2000 (Anon 1980a).

Solomon Islands is among the wettest countries in the world (Brookf ield 1969); recorded annual rainfall ranges from 218 cm in Honiara to 435 cm at Graciosa Bay in the Santa Cruz Islands (1970-1979 average from data in Anon 1980a). This, in part, is a result of Solomon Islands' location near the Equator (4°S to 16°S, 153°E to 173°E) in the region of the intertropical convergence zone (Donguy and Henin 1978 and 1981), and in part is a result of the presence of six large, high elevation islands. These islands partially enclose The Slot where much of the locally based pole-and-line catch is taken. FIGURE 1. SURVEY AREA AND BAITFISHING LOCALITIES (NUMBERS) FOR THE 1977 A PROGRAMME SURVEYS IN THE WATERS OF SOLOMON ISLANDS. Inset map shows area for Solomon Islands.

155E 156E 157E 158E 159E 160E 161E

BOUGAINVILLE IS.

SHORTLAND IS.

VELLA C\lAVELLA IS

NEW GEORGIA GROUP flOTO

/*. Ontong Java * Poncador Re

s Duff Islands ..

«« • tt

155E 156E 157E 158E 159E 160E 161E 3

The pole-and-line fishery began in 1971 after a successful year of exploratory fishing surveys by Taiyo Fishing Company Limited of Japan. The Government of Solomon Islands then signed a joint venture agreement with Japan which led, in 1972, to the formation of Solomon Taiyo Limited, a Solomon Islands registered fishing company (Anon 1975). Soon afterwards, the company established a fishing base at Tulagi, which now has freezing, cold storage, canning and smoking facilities. A second base was established in 1976 at Noro. Until 1980, the company was restricted, by government regulation, to use of pole-and-line gear for capturing surface tunas. Solomon Taiyo Limited presently operates 19 pole-and-line vessels and one small purse-seine vessel, while a second company, the National Fisheries Development Corporation Limited, established in 1978 under Solomon Islands ownership, operates a fleet of four ferro-cement pole-and-line vessels.

Commercial pole-and-line catches rose from under 5,000 tonnes in 1971 (Table 1) to a peak catch of just over 23,000 tonnes in 1979. The single Solomon Islands seine vessel harvested 2,823 tonnes of surface tunas during its first year of operation in 1981 (Solomon Islands Ministry of Natural Resources, unpublished data). Small subsistence fisheries for surface tuna operate from many Solomon Islands villages. Their catches are poorly documented, but are undoubtedly very small in relation to the commercial catch.

TABLE 1. ANNUAL TUNA AND BAITFISH CATCH AND EFFORT STATISTICS FOR THE LOCALLY BASED POLE-AND-LINE FISHERY IN THE WATERS OF SOLOMON ISLANDS*

Tuna Catch Days Tonnes Bait Catch Nights Buckets (tonnes) Fished per Day (buckets) Fished p er Night

1971 4,666 NA NA NA NA NA 1972 7,641 NA NA NA NA NA 1973** 6,446 1,909 3.38 118,804 1,722 70.0 1974 10,332 2,180 4.74 91,371 1,503 60.4 1975 7,146 2,149 2.95 131,430 1,563 86.2 1976 15,800 3,495 4.52 169,560 1,974 87.1 1977 12,155 4,409 2.75 212,271 2,756 81.8 1978 17,455 4,656 3.75 238,965 3,579 59.8 1979 23,807 5,085 4.68 303,743 4,858 62.9 1980 21,936 4,993 4.39 325,645 4,903 63.1 1981 22,171 NA NA NA NA NA

* Source: 1971--1972 - Anon 1975 1973--1981 - Unpublished data, Ministry of Natural Resources, Honiara, Sol omon Islands. One bucket is assumed to contain 2 .5 kg of bait.

** No data for January through March period NA No data available

Over 90 per cent of the commercial catch is taken within the outer boundary of the territorial sea between Shortland Islands and the southern tip of Guadalcanal Island. Within this area catches are concentrated in The Slot and in waters just west of the New Georgia Islands. Over 99 per cent of the pole-and-line catch is reported as skipjack; however, small yellowfin (Thunnus albacares). less than 6 kg, are generally not separated from 4 skipjack in the landings, hence the percentage of yellowfin in the catch is somewhat higher than the one per cent shown by landing statistics (Doug Gibson, personal communication). It may be as high as the 10 per cent level reported for Papua New Guinea pole-and-line catches (Ellway and Kearney MS). Solomon Taiyo's pricing structure for tuna discourages the capture of small skipjack, less than 1.5 kg, since small fish produce lower yields after processing.

Peak catches (Figure 2) and catch per unit of effort may occur anytime from June to November. January through April is the "off-season", when crews from the 23 pole-and-line vessels (most in the 59- to 100-tonne range) usually take a month's holiday. The upward trend in total annual catch, which has increased by approximately 2,000 tonnes per year since 1971, is closely correlated with increasing effort. Annual catch per unit of effort (CPUE) has fluctuated between 2.75 tonnes per fishing day in 1977, also a poor year in the Papua New Guinea fishery (Ellway and Kearney MS), and 4.74 tonnes per day in 1974. There was no correlation between annual CPUE and effort over the 1973 to 1980 period, a feature common to all major fisheries for skipjack (Joseph and Calkins 1969, Kearney 1979).

Since 1971, Solomon Islands Ministry of Natural Resources has operated a port sampling programme to estimate monthly length frequency distribution of the commercial catch. Catch statistics are collected from the local fleet by month and by one-half degree square.

The catch of bait species for pole-and-line fishing has increased roughly in proportion to increased tuna catches and now exceeds 800 tonnes per year (Table 1). Fishing for bait with "bouki-ami" nets takes place in lagoon areas where traditional fishing is active. There has been concern that in heavily fished areas the relatively effective commercial method could reduce catches by traditional fishing methods, and as well could possibly over-exploit the baitfish resource (Anon MS). In recognition of these concerns, a royalty is paid to the villagers for the privilege to fish for baitfish with bouki-ami gear, and permission to fish for bait can be withheld if there are conservation or allocation concerns.

Most skipjack caught by the locally based pole-and-line fishery are exported frozen, principally to the canneries in Pago Pago, American Samoa, and to canneries in Mexico (until 1980), in Puerto Rico, and in the continental United States (Anon 1980a). Local sales represent only a small portion of the total pole-and-line catch. Utilisation of the pole-and-line catch, averaged over 1978 and 1979, was as follows: frozen exported - 85 per cent, canned - 10 per cent, smoked - 4 per cent, and local sales of frozen fish/other - 1 per cent.

Foreign-based tuna fisheries have operated for a number of years in the waters of Solomon Islands. Statistics published by the South Pacific Commission for Japanese distant-water pole-and-line vessels for 1972 to 1978, Japanese longline vessels for 1962 to 1977, Taiwanese longline vessels for 1967 to 1977, and Korean longline vessels for 1975 and 1976, all indicate considerable effort within the Solomon Islands 200 nautical mile fishery zone (Klawe 1978, Skipjack Programme 1980, 1981a). Japanese pole-and-line vessels were present in each of the years for which catch statistics are available. Their landings of skipjack reached 19,475 tonnes in 1976 (98 per cent skipjack), but fell to 259 tonnes in 1978 largely as a result of changes in access arrangements between Japan and Solomon Islands. The Japanese pole-and-line catch of skipjack in international waters just to the north and east of Solomon Islands ranged between 107 tonnes in 1975 and 6,670 tonnes in 1976 (Tuna Programme, South Pacific Commission, unpublished data). 5

00 0-- •• •. • © oo o 0) H vO

CO W co V •© CO w ©. o > 03 w 0) z H •J I .©* P •©• . • ©• w o•J © 0) CO Q ©• 0) © . •J • ©• CO z .©• I ©•• . o © ohJ 00 CO ©• .•; !* ••©-. 0) • ••© CwO ©• Zz o H .1 •©• pa .. © o ©•• H © ©• 0) • • JO S3 o • © H •©. :©

H ©• • • Z ©•• o- § o •. (0 ©••• •o ©•.. CO w • • •©*.. ©. & o o o o o o o o o CM

(sauuo}) H01V0 A1H1NOH 6

Japanese, Korean and Taiwanese longliners were present in the Solomon Islands zone each year from 1962 to 1977, though catches of skipjack were negligible, as expected for this type of gear. However, the longline catch of other tuna and billfish species, principally large yellowfin, albacore (Thunnus alalunga) and bigeye tuna (T. obesus), and various marlin species, averaged over 5,000 tonnes annually during the mid to late 1970s.

Recently expanded purse-seine fisheries by several countries in waters to the north and west of Solomon Islands may presently be harvesting approximately 45,000 tonnes of skipjack annually (Tuna Programme, South Pacific Commission, unpublished data).

2.0 SKIPJACK PROGRAMME RESEARCH PLAN

The objectives of the Skipjack Programme were to survey the skipjack and baitfish resources of all countries within the area of the South Pacific Commission and to assist with the assessment of the status of the stocks and the degree of interaction between individual fisheries for skipjack within the region and beyond. These assessments would provide a basis for both rational development of skipjack fisheries throughout the region and for sound management of the resource. The Programme's field work spanned almost three years, from October 1977 to August 1980 inclusive, and incorporated visits to all of the countries in the area of the South Pacific Commission and also New Zealand and Australia (see Figure A). Eight hundred and forty-seven days of chartered vessel time were spent in the region and 25 countries and territories were visited. Seventy-one days were spent in the waters of Solomon Islands in 1977 and 1980. At the request of the Government of Solomon Islands, the Programme made a special effort to include surveys of the tuna and bait resources of Ontong Java and of the Santa Cruz Islands.

3.0 VESSEL AND CREW

Two Japanese commercial fishing vessels were chartered separately by the Skipjack Programme. The Hatsutori Maru No.l. of 192 gross tonnes, was used during the first visit to Solomon Islands, and the Hatsutori Maru No. 5. of 254 gross tonnes, during the second. Both vessels were chartered from a commercial fishing company, Hokoku Marine Products Company Limited, Tokyo, Japan, and were slightly modified to accommodate the requirements of fisheries research work. Details of both vessels are given by Kearney (1982).

The Hatsutori Maru No.1 was operated with at least three Skipjack Programme scientists, nine Japanese officers and between nine and twelve Fijian crew. For the Hatsutori Maru No.5. between 13 and 15 Fijian crew were employed. Observers from the Ministry of Natural Resources of Solomon Islands were on board for varying times throughout the survey. All personnel and details of the times scientists and observers spent on board are given in Appendix A.

4.0 METHODS

Visual scanning and exploratory fishing for tunas and baitfish were the primary survey techniques. Tagging and biological sampling, including blood and parasite studies, were the basic tools used to assess skipjack resources. Analysis of results from these studies, together with available catch and effort data, constitute the basis for assessment of the resource of skipjack. 7

Baitfish resources were surveyed by exploratory fishing, predominantly at night. Assessments of the baitfish resources were based on these results, estimates of the magnitude of suitable baitfish habitat, and knowledge of the utility of the common species as skipjack bait.

4.1 Skipjack Fishing. Tagging and Biological Sampling

Both vessels used by the Skipjack Programme were commercial live-bait pole-and-line fishing vessels and the basic strategy of approaching and chumming schools normally employed by these vessels was not changed. As for commercial vessels, minor variations in technique were tried from day to day depending upon the behaviour of skipjack schools and the quantity and quality of live-bait carried.

The number of crew on the Hatsutori Maru No. 1 and No. 5 was less than either of these vessels carry when fishing commercially. As at least one crew member was required to assist each scientist in the tagging procedures, the effective number of fishermen was further reduced. Moreover, the need to pole skipjack accurately into the tagging cradles reduced the speed of individual fishermen. Clearly these combined effects would decrease the fishing power of the research vessel. During the first survey in the waters of Fiji (26 January - 10 April 1978), the Hatsutori Maru No.l fished commercially for approximately one month, under an agreement between the Programme and the vessel's owners. From comparison of survey and commercial catches at this time, it was estimated that the fishing power of the Hatsutori Maru under survey conditions was 28 per cent of its fishing power during commercial fishing (Kearney 1978).

As tagging was the primary tuna research tool, attempts to tag large numbers of fish normally dominated the fishing strategy. The tagging techniques and alterations to commercial fishing procedures have been described in detail by Gillett and Kearney (1982).

Specimens of all tuna and other pelagic species which were poled or trolled, but not tagged and released, were routinely analysed. Data collected included length and weight measurements, sex, gonad weights, stages of sexual maturity, and records of stomach contents and fullness. In addition, a log was maintained of all fish schools sighted throughout the Programme. Where possible the species composition of each school was determined. Records were kept of the chumming response and catch by species from each school. Argue (1982) describes methods used for the collection of these data.

Blood samples for genetic analysis were collected according to the methods described by Fujino (1966) and Sharp (1969), and were frozen and packed on dry ice for air freighting to the Australian National University, Canberra, Australia, where they were electrophoretically analysed (Richardson MS).

During the second visit to Solomon Islands, skipjack body cavities were examined for the presence of macro-parasites, and complete sets of gills and viscera were taken from five fish from each school (up to a maximum of three schools per day), frozen, and subsequently air freighted to the University of Queensland, St Lucia, Australia, for detailed examination for the presence of parasites. 8

4.2 Baitfishing

Baitfishing carried out by the Programme in the waters of Solomon Islands employed a "bouki-ami"' net set at night around bait attraction lights. Procedures were similar to those used by commercial vessels, but were modified where necessary to meet the Programme's special requirements. In some countries beach seining during daylight was used as an alternative bait catching technique. Beach seining was not attempted in Solomon Islands. Details of both techniques and all modifications employed by the Skipjack Programme are given by Hallier and Gillett (1982).

4.3 Data Compilation and Processing

Five separate logbook systems formed the basis for compiling data accumulated during the fieldwork outlined in Sections 4.1 and 4.2. The techniques used to enter data into computer files and to process data are discussed by Kleiber and Maynard (1982). Data processing was carried out on the Programme's Hewlett Packard 1000 computer in Noumea. Electrophoretic data from blood samples and parasite identifications from gill and viscera specimens were also coded and entered into files on the computer.

4.4 Data Analysis

Assessment of the skipjack resource and possible interactions among skipjack fisheries in Solomon Islands and those in other countries were approached from several viewpoints. Studies of the migration of tagged skipjack, using analytic techniques described in Skipjack Programme (1981b) and Kleiber (MS), have formed the basis of investigations on movement patterns and fishery interactions. Methods employed in biological studies of growth are described in Lawson and Kearney (MS), and for juvenile abundance, in Argue, Conand and Whyman (MS). Comparison of fishing effectiveness between different baitfish families follows on procedures described in Argue and Hallier (MS). Evaluation of population structuring across the whole of the western and central Pacific has centred on a comparison of the blood genetics work with tagging results (Anon 1980b, 1981b). Occurrence and distribution of skipjack parasites has also been evaluated (Lester 1981).

5.0 SUMMARY OF FIELD ACTIVITIES

During the two surveys 71 days were spent in the waters of Solomon Islands of which 53 were spent fishing, 11 steaming, and 7 in port (Table 2). Included in Table 2 are two days when the vessel fished in the waters of Solomon Islands during the first survey of Papua New Guinea.

Figure 1 shows the areas surveyed for tuna and baitfish. During the first visit, Ontong Java atoll, the waters of the Solomon Islands/Papua New Guinea border area near Shortland Islands, The Slot and the Santa Cruz Islands were surveyed. Survey work during the second visit took place in or near The Slot and amongst the Santa Cruz Islands. Though skipjack fishing could not be attempted at Tikopia Island and amongst the Duff and Reef Islands due to lack of bait, time was spent evaluating the tuna resource by sighting.

Baitfishing activities are summarised in Table 3. Eighteen separate localities were fished using the bouki-ami night-baiting technique (see baiting locality numbers in Figure 1). Baiting was not attempted in the Duff and Reef Islands since, at the time of the surveys, charts suitable for navigating the reef passages and lagoons were unavailable to the Programme. 9

TABLE 2. SUMMARY OF DAILY FIELD ACTIVITIES IN THE WATERS OF SOLOMON ISLANDS. Schools sighted are given by species: SJ = skipjack or skipjack with other species except yellowfin; YF = yellowfin or yellowfin with other species except skipjack; S+Y = skipjack with yellowfin or skipjack with yellowfin and other species; t)T = other species without skipjack or yellowfin; UN = unidentified.

Date General Area Principal Bait Hours School s S ighted Fi sh Tagged Fish Caught Total Activity Carried Fishing (numbers) (numbers ) (kg) Catch (kg) and SJ YF S+Y OT UN SJ YF OT SJ YF (kg) Sighting

25/10/77* Bougainville Is Fishing 270 5 1 0 0 0 0 113 0 0 574 0 574 29/10/77* Shortland Is Fishing 50 8 3 0 0 0 0 155 0 0 561 0 563 01/11/77 NW New Georgia Steaming 0 10 2 0 0 0 10 0 0 0 9 0 9 02/11/77 Honiara In Port 0 0 ------03/11/77 Santa Ysabel Is Steaming 0 5 0 0 0 0 7 - -. - - - - 04/11/77 The Slot Fishing 195 11 3 0 0 0 3 69 0 0 232 0 232 05/11/77 Kolokofa Hbr Fishing 308 12 3 1 1 0 2 69 0 0 227 0 231 06/11/77 Shortland Is Fishing 158 9 2 2 1 1 3 10 80 3 17 158 197 07/11/77 Gizo Is Fishing 98 11 0 3 1 0 3 1 1 0 1 11 15 08/11/77 Kolokofa Hbr Fishing 60 7 0 2 2 0 0 0 0 0 0 0 0 09/11/77 The Slot Fishing 300 11 1 2 1 0 1 531 0 0 1648 7 1659 10/11/77 Honiara In Port 0 0 ------11/11/77 Guadalcanal Is Fishing 143 3 0 0 2 0 0 0 0 0 0 0 0 12/11/77 The Slot Fishing 120 13 7 0 1 0 1 236 0 0 825 29 858 13/11/77 The Slot Fishing 36 11 0 0 1 0 0 0 0 0 0 0 0 14/11/77 Mborokua Is Fishing 188 12 0 0 3 0 0 484 37 0 1578 326 1906 15/11/77 Mborokua Is Fishing 75 9 4 1 1 0 2 3 0 0 10 0 10 16/11/77 Mborokua Is Fishing 390 8 3 0 1 2 0 0 0 0 0 0 8 17/11/77 S Guadalcanal Is Fishing 270 11 1 1 1 0 0 92 0 0 355 . 0 355 18/11/77 S Guadalcanal Is Fishing 105 10 0 2 1 0 1 4 0 0 13 0 13 19/11/77 Indispensable Fishing 323 13 0 0 1 0 1 32 0 0 127 26 153 20/11/77 S New Georgia Fishing 263 13 1 0 1 1 0 1 0 0 8 0 10 21/11/77 Santa Ysabel Is Fishing 194 9 1 0 0 1 0 0 0 0 0 0 2 22/11/77 Ontong Java Fishing 188 12 1 1 1 0 2 0 0 0 7 11 18 23/11/77 Ontong Java Fishing 165 10 0 3 0 0 4 0 0 0 0 21 21 24/11/77 Roncador Reef Fishing 327 12 2 1 0 0 1 0 0 0 0 0 0 25/11/77 Mborokua Is Fishing 234 13 3 1 0 0 0 273 0 0 927 0 927 26/11/77 Honiara In Port 0 0 ------27/11/77 Honiara In Port 0 0 ------28/11/77 Marau Sound Steaming 0 0 ------29/11/77 Marau Sound Baiting 63 0 ------30/11/77 San Cristobal Is Fishing 123 13 4 1 0 0 2 0 0 0 0 0 0 01/12/77 V Santa Cruz Is Fishing 80 10 0 0 1 0 0 194 0 0 862 0 862 02/12/77 Utupua Is Fishing 134 7 1 1 0 0 0 0 0 0 0 0 0 03/12/77 W Vanikoro Is Fishing 227 12 3 1 0 2 1 223 0 0 1037 0 1041 04/12/77 Graciosa Bay Fishing 35 5 0 1 0 0 0 0 0 0 0 19 19

25/05/80 Tikopia Is Steaming 0 11 0 1 0 0 13 - - - _ - - 26/05/80 Santa Cruz Is Steaming 0 10 0 0 0 0 18 ------27/05/80 San Cristobal Is Steaming 0 7 0 0 0 0 7 ------28/05/80 Honiara In Port 0 0 ------29/05/80 Honiara In Port 0 0 ------30/05/80 Honiara-Tulag1 Steaming 0 0 ------31/05/80 Malaita Is Fishing 119 7 0 0 0 0 1 0 0 0 0 0 0 01/06/80 Malaita Is Fishing 228 11 0 0 0 0 2 0 0 0 0 0 0 02/06/80 Nendo Is Fishing 366 9 0 0 3 0 6 2 152 2 100 606 712 03/06/80 Nendo Is Fishing 135 8 3 0 0 1 1 11 0 0 44 0 78 04/06/80 Utupua Is Fishing 47 5 0 0 1 0 7 68 1 0 239 29 268 05/06/80 Vanikoro Is Fishing 201 10 2 0 1 0 6 105 42 0 355 193 547 06/06/80 Utupua Is Fishing 152 10 0 1 2 1 3 0 0 0 0 0 0 07/06/80 San Cristobal Is Fishing 38 11 0 0 1 0 10 0 0 0 0 0 0 08/06/80 Honiara-Santa Baiting 0 5 1 0 0 0 9 ------Ysabel Is 09/06/80 The Slot Fishing 197 10 1 0 0 1 2 99 0 0 119 0 119 10/06/80 The Slot Fishing 68 7 2 0 0 0 0 46 0 1 62 0 70 11/06/80 The Slot Fishing 483 5 1 0 1 1 3 288 18 0 753 59 846 12/06/80 The Slot Fishing 519 10 1 0 1 3 7 139 40 0 629 173 890 13/06/80 Vella Lavella Is Fishing 810 9 0 0 1 2 5 24 139 0 127 884 1042 14/06/80 Vella Lavella Is Fishing 416 5 0 0 2 1 2 122 149 0 425 557 1032 15/06/80 Vella Lavella Is Fishing 705 6 0 0 4 1 0 37 55 0 164 262 506 16/06/80 The Slot Fishing 239 11 1 0 0 0 1 0 0 0 0 0 0 17/06/80 The Slot Fishing 68 6 3 0 0 0 0 246 0 0 842 0 842 18/06/80 Santa Ysabel Is Fishing 492 6 0 0 1 4 0 1533 26 1 3167 114 3567 19/06/80 Santa Ysabel Is Fishing 224 10 0 0 0 4 3 0 0 0 0 0 224 20/06/80 Indispensable Fishing 413 8 3 0 1 2 2 345 42 0 1228 105 1431 21/06/80 Indispensable Fishi'ng 227 ' 7 1 0 3 2 2 459 76 0 1475 255 1858 22/06/80 Indispensable Fishing 158 6 1 0 0 1 1 204 0 0 987 0 1001 23/06/80 Guadalcanal Is Fishing 441 5 2 0 0 1 1 3 0 0 13 0 15 24/06/80 Honiara In Port 0 0 ------25/06/80 Honiara Steaming 0 0 ------26/06/80 Santa Cruz Is Steaming 0 0 ------27/06/80 Duff Is Steaming 0 5 0 0 0 0 0 ------28/06/80 E Duff Is Steaming 0 10 0 0 0 0 3 ------TOTALS 525 68 26 43 32 159 6221 858 7 19747 3844 24731

* During the first survey in Papua New Guinea waters a portion of two days f ishing, near the Papua New Guinea-Solomon Islands border, was spent in the waters of Solomon Islands. 10

TABLE 3. SUMMARY OF SKIPJACK PROGRAMME BAITFISHING EFFORT AJJD CATCH IN THE WATERS OF SOLOMON ISLANDS

Locality Anchorage Time Number Dominant Species Est. Av. Mean Other Common Species Number of of Catch Length Hauls Hauls per Haul (mm) (kg)

Thousand Ships Bay Stolephorus devisi 105 54 Pellona ditchela 08°26'S Night 1 Stolephorus heterolobus 81 61 Trichiuris haumela 159°41'E Herklotsichthys punctatus Stolephorus indicus

Thousand Ships Bay Stolephorus buccaneeri 78 55 Pellona ditchela 08°26'S Night 2 Rastrelliger faughni 23 93 Rastrelliger kanagurta 159°40'E Stolephorus heterolobus 6 67 Dussumieria sp.

Thousand Ships Bay Stolephorus heterolobus 60 69 Rastrelliger brachvsoma 08°27'S Night 8 Stolephorus devisi 49 57 Sardinella sirm 159°40'E Stolephorus buccaneeri 32 59 Selar boops

Thousand Ships Bay Stolephorus devisi 53 57 Rastrelliger faughni 08°27'S Night 2 Stolephorus buccaneeri 32 64 Rastrelliger brachysoma 159°41'E Stolephorus heterolobus 25 63 Pellona ditchela

Thousand Ships Bay Stolephorus devisi 118 62 Stolephorus buccaneeri 08°26'S Night 2 Stolephorus heterolobus 94 69 Leiognathus bindus 159°39'E Sardinella melanura 7 78 Secutor sp.

Kolokofa Harbour Stolephorus devisi 85 56 Pterocaesio sp. 07°42'S Night 5 Stolephorus heterolobus 26 61 Pterocaesio pisang 158°34'E Gymnocaesio gymnopterus 5 63 Dipterygonotus leucogrammicus

Lorau Point Bregmaceros sp. 14 Stolephorus devisi 07°00'S Night 1 Spratelloides gracilis 14 Stolephorus heterolobus 155°46'E Thrissina baelama 11 Pseudamia sp.

Ghizo Harbour Herklotsichthys punctatus 14 65 Apogon(Rhabdamia) cypselurus 08°04'S Night 1 Spratelloides gracilis 8 49 Archamia lineolata 156°49'E Spratelloides delicatulus 5 49 Benthosoma fibulatum

Macquitti Bay Spratelloides gracilis 15 Dipterygonotus leucogrammicus 09°02IS Night 1 Herklotsichthys punctatus 8 Pterocaesio tile 159°06'E Gymnocaesio gymnopterus 3 Stolephorus devisi

Marovo Lagoon Spratelloides gracilis 100 61 Stolephorus devisi 08°30'S Night 2 Spratelloides (new species) 72 61 Dip.terygpnotus leucogrammicus 158°02'E S.tplephorus heterplpbus 47 57 Pterocaesio pisang

Danae Bay Gymnocaesio gymnopterus 108 55 Sardinella sirm 09°49'S Night 4 Herkipt8ichthys punctatus 51 48 Thrissina baelama 160°49'E Spratelloides (new species) 14 Spratelloides gracilis

Ontong Java Spratelloides gracilis 233 43 Hyppatherina sp. 05°33'S Night 1 Hyppatherina ovalaua Archamia lineolata 159°42'E Apogon(Rhabdamia) cypselurus Apogon fragilis

Haoto Bay Herklotsichthys punctatus 20 Dussumieria sp. 09°45'S Night 1 Gymnocaesio gymnopterus 17 Sardinella sirm 160»47'E Dipterygonotus leucogrammicus 17 Stolephorus devisi

10 Byron Bay Stolephorus devisi 18 45 Herklotsichthys punctatus 11°15'S Night 6 Thrissina baelama 10 78 Stolephorus bataviensis 166°30'E Hyppatherina ovalaua 7 56 Spratelloides (new species)

11 Mangadai Bay Spratelloides (new species) 40 85 Stolephorus heterolobus 11°39'S Night 4 Herklotsichthys punctatus 23 54 Stolephorus devisi 166°55'E Spratelloides gracilis 5 43 Leiognathus bindus

12 Gavutu Harbour Stolephorus heterolobus 45 52 Rastrelliger kanagurta 09°06'S Night 2 Spratelloides gracilis 8 50 Gazza minuta 160°11'E Stolephorus buccaneeri 8 50 Pellona ditchela 12

Skipjack fishing activities, including school sightings, tag releases and catches, are summarised in Table 2. On fishing days an average of 9.1 hours were spent searching and fishing. A total of 328 schools were sighted, for an average of 0.62 schools sighted per hour. Of the 169 identified schools, 66 per cent had some skipjack and 21 per cent had some yellowfin, many schools having both.

A total of 19.8 tonnes of skipjack and 3.8 tonnes of yellowfin were caught and accounted for over 95 per cent of the total catch of 24.7 tonnes. This represents an average catch of 485 kg per fishing day. Other species in the catch included mackerel tuna (Euthynnus affinis) , frigate tuna (Auxis thazard) and rainbow runner (Elegatis bipinnulatus).

Due to comparatively poor fishing conditions during the first visit to Solomon Islands, only 2,611 tagged tuna were released (95 per cent skipjack), 2,194 in and near The Slot and 417 amongst the Santa Cruz Islands. During the second visit, the Hatsutori Maru No.5 did as well or better than l*ocal pole-and-line vessels and tagged 4,475 tuna (83 per cent skipjack), of which 4,092 were tagged in and near The Slot and 383 amongst the Santa Cruz Islands. As of 1 March 1982, the SPC had received 607 tag returns from a total of 7,079 skipjack and yellowfin that were tagged in the waters of Solomon Islands. There have been no recoveries from seven releases of tagged bigeye tuna.

A summary of numbers of fish sampled for biological data is given in Table 4. The size distribution of tagged skipjack (Figure 3) shows a range of 29 to 73 cm fork length. The average length was 50.5 cm, which is almost identical to the average length of sampled skipjack (Figure 3) and to the Skipjack Programme's overall average of 50.4 cm for tagged skipjack. Tagged yellowfin averaged 56.1 cm; sampled yellowfin averaged 55.2 cm (Figure 3). Skipjack maturity data is summarised in Figure 4. Skipjack diet items are listed in Table 5. The incidence of tuna juveniles in the stomachs of sampled skipjack, yellowfin and other species is given in Table 6. A single skipjack blood sample of 113 specimens was taken from a school near Thousand Ships Bay on 18 June 1980.

6.0 RESULTS AND DISCUSSION

6.1 Baitfishing

6.1.1 General observations

Results from the Programme's baiting at each of 18 localities in Solomon Islands are detailed in Table 3. For larger localities, such as Thousand Ships Bay and Austria Sound, results are shown separately for all anchorages which were further than one nautical mile apart. Table 7 presents a summary of these data for the 10 dominant bait species. Over one hundred species were identified from 59 bait hauls in the waters of Solomon Islands (Appendix B); however, the 10 species in Table 7 were estimated to account for 85 per cent of the total catch of 8,966 kg. Note that in this and subsequent tables the total catch includes a small percentage of bait (~ 4%) that was discarded while loading the bait on board the research vessel (e.g., dead bait, bait in excess of baitwell holding capacity). 11

Locality Anchorage Time Number Dominant Species Est. Av. Mean Other Common Species Number of of Catch Length Hauls Hauls per Haul (mm) (kg)

13 Are Are Lagoon Sardinella sirm 59 150 Stolephorus bataviensis 09°25'S Night 2 Rastrellieer fauehni 36 137 Stolephorus devisi 161°12'E Decapterus maruadsi 16 124 Spratelloides delicatulus

14 Graciosa Bay Herklotsichthvs ounctatus 6 90 Selar boops 10°45'S Night 2 Stolephorus bataviensis 3 89 Alepes sp. 165°48'E Franesus oineuis 3 76 Thrissina baelama

15 Wickham Harbour Haul aborted 08°43'S Night 158°03'E

16 Austria Sound Rastrellieer kanaeurta 102 StoleDhorus devisi 07 °40'S Night 2 Sardinella sirm 62 150 Spratelloides gracilis 158°29'E StoleDhorus heterolobus 28 58 Pranesus Dineuis

Austria Sound Stolephorus heterolobus 70 55 Rastrellieer brachvsoma 07"39'S Night 2 Rastrellieer kanaeurta 58 169 Gvmnocaesio evmnoDterus 158°29'E Herklotsichthvs ounctatus 58 97 Spratelloides eracilis

17 Labaka Bay Rastrellieer kanaeurta 313 167 Spratelloides (new species) 07°22'S Night 2 Scomberoides tol 66 Spratelloides delicatulus 15717'E Sardinella sirm 42 152 Selar crumenophthalmus

18 Sandfly Bay Herklotsichthvs ounctatus 153 78 Gazza minuta 07°15'S Night 6 Stolephorus bataviensis 44 76 HvDoatherina ovalaua 156°34'E Sardinella melanura 19 91 Rastrellieer fauehni

Explanatory Notes

Anchorage Recorded positions are truncated to the nearest minute. For large bays there may be more than one position tabulated.

Time of Hauls Day hauls 0600-1759 hrs inclusive Night hauls 1800-0559 hrs inclusive

Number of Hauls Number of hauls at the anchorage position, either day or night as specified. A haul is defined as any time the net was placed in the water.

Species Those species that made up at least one per cent of the numbers caught from one or more bait hauls at a particular location.

Average Catch (species) Total catch includes bait loaded, bait discarded alive and bait discarded dead at the location. The average catch in kilograms per haul is the product of total catch in kilograms and weighted numerical percentage of the catch for a particular species, divided by the total number of hauls at the location. The weighted numerical percentage is the product of numerical percentage, a constant, and the cube of the species' average standard length. (In the absence of a mean SL for the species, the numerical percentage itself is used). The sum of the weighted percentages equals the sum of the numerical percentages. In this way the smaller (numerically abundant) fish are suppressed in their contribution to the catch while the less common, larger fish are proportionally enhanced in their representation. Catches are expressed in kilograms for the dominant three species; thus, the sum of the average species catches will often be less than the average location catch.

Mean Length Weighted by numerical abundance when there were multiple hauls at the same location.

Baitfish Taxonomy The taxonomic nomenclature for several of the above baitfish species has recently been reviewed, and pending publication, the following changes are anticipated:

Herklotsichthvs punctatus to Herklotsichthvs quadrimaculatus. and Pranesus pineuis to Atherinomorus lacunosa. 13

TABLE 4. SUMMARY OF NUMBERS OF FISH SAMPLED FOR BIOLOGICAL DATA FROM THE WATERS OF SOLOMON ISLANDS

Species Total No. Total No. Total No. Total No. Total No. Measured Weighed Examined Examined Examined for Sex for Stomach for Tuna Content Juveniles

Skipjack 855 494 555 287 492 Katsuwonus pelamis

Yellowfin 266 154 173 92 164 Thunnus albacares

Mackerel Tuna 155 72 78 41 79 Euthvnnus affinis

Frigate Tuna 119 97 96 45 95 Auxis thazard

Bigeye Tuna 6 0 0 5 5 Thunnus obesus

Rainbow Runner 168 165 119 58 130 Elegatis bipinnulatus

Dolphin Fish 11 11 11 9 11 Corvphaena hippurus

Dogtooth Tuna 1 1 1 1 1 Gvmnosarda unicolor

Spanish Mackerel 3 3 3 3 3 Scomberomorus commerson

White-spotted Triggerfish 5 0 0 0 0 Canthidermis rotundatus

Barracuda 1 1 1 1 1 Sphvraena sp.

TOTALS 1590 998 1037 542 981 FIGURE 3. LENGTH FREQUENCY DISTRIBUTIONS FOR TAGGED AND SAMPLED SKIPJACK (LEFT SAMPLED YELLOWFIN (RIGHT) FOR BOTH SKIPJACK PROGRAMME SURVEYS IN THE W ISLANDS. Mean lengths, standard deviations of the means (SD) and samp indicated.

500 Tagged Skipjack 75 n Mean 50.5 SD 5.99 400 N 5730 RF

50- 300- a m Z O E 25 < ~^>100 O Q. LJ if) n >- o •i' 20 30 40 50 60 70 80 20 30 40 50 60 O o 801 Sampled Skipjack 25 Mean 50.3 if) SD 5.99 K. N 855 LJ m 20

15

10 I m n n 50 60 70 80 20 30 40 50 60 FORK LENGTH IN CENTIMETRES 15

LENGTH FREQUENCY DISTRIBUTION OF FEMALE SKIPJACK FROM SOLOMON ISLANDS BIOLOGICAL SAMPLES (UPPER GRAPH) AND DISTRIBUTION OF MATURITY STAGES OF FEMALE SKIPJACK (LOWER GRAPH). Stippling in the upper graph shows those females that were maturing or mature (.> maturity stage 3)»

301 Stippled: Mature Clear: Immature

20-

'•..•-.•L M

JW ' - * .* * 30 40 50 60 70 FORK LENGTH IN CENTIMETRES

N 252

.:'; /„•.' .V.".7-\ JS-S. 2 3 4 5 MATURITY STAGE 16 TABLE 5. STOMACH CONTENTS OF SKIPJACK SAMPLED IN THE WATERS OF SOLOMON ISLANDS

Item Diet Item Number of Percentage No. Stomachs Occurrence Fish and Invertebrates

1 Chum from Hatsutori Maru 200 69.69 2 Fish remains (not chum) 111 38.68 3 Squid (Cephalopoda) 64 22.30 4 Alima stage (Stomatopoda) 58 20.21 5 Anchovy juvenile (Engraulidae) 36 12.54 6 Stolephorus buccaneeri (Engraulidae) 30 10.45 7 Acanthuridae 28 9.76 8 Empty stomach 26 9.06 9 Siganidae 25 8.71 10 Exocoetidae 21 7.32 11 Tuna juvenile (Scombridae) 19 6.62 12 Shrimp (Decapoda) 14 4.88 13 Blue goatfish (Mullidae) 12 4.18 14 Chaetodontidae 12 4.18 15 Juvenile fish 10 3.48 16 Aluteridae 10 3.48 17 Synodontidae 10 3.48 18 Megalopa stage (Decapoda) 9 3.14 19 Pteropoda (Gasteropoda) 8 2.79 20 Unidentified fish 8 2.79 21 Holocentridae 8 2.79 22 Balistidae 7 2.44 23 Euphausiid (Euphausiacea) 7 2.44 24 Gempylidae 6 2.09 25 Leptocephalus (Anguilliformes) 5 1.74 26 Decapterus sp. (Carangidae) 4 1.39 27 Stomatopoda 4 1.39 28 Carangidae 4 1.39 29 Phyllosoma stage (Decapoda) 4 1.39 30 Myctophidae 2 .70 31 Trichiuridae 2 .70 32 Dactvlopterus orientalis (Dacylopteridae) 2 .70 33 Caesiodidae 2 .70 34 Crustacean remains 2 .70 35 Apogonidae .35 36 Stomiatidae .35 37 Argonauta (Cephalopoda) .35 38 Feather tuna jig .35 39 Mulloidichthys samoensis (Mullidae) .35 40 Paralepidae .35 41 Rastrellieer sp. (Scombridae) .35 42 Syngnathidae .35 43 Menidae .35 44 Octopus (Cephalopoda) .35 45 Chiasmodon sp. (Chiasmodontidae) .35 46 Plant material .35 47 Leiognathidae .35 48 Bramidae .35 49 Tunicate (Urochordata) .35 50 Callionymidae .35 51 Amphipoda .35 52 Coelenterata .35 53 Tetrodontidae .35

Total Stomachs Examined 287 TABLE 6. INCIDENCE OF TUNA JUVENILES IN THE STOMACHS OF SKIPJACK AND OTHER PRE FROM THE WATERS OF SOLOMON ISLANDS

Predator Predators Prey Species No. of Pr edators Prey per Examined (tuna Prey wi th Prey 100 juveniles) Predators

Skipjack 492 Skipjack 46 23 9.35

Yellowfin 5 3 1.02

Mackerel Tuna 2 2 .41

Frigate Tuna 17 10 3.46

Yellowfin 164 Skipjack 1 1 .61

Mackerel Tuna 1 1 .61

Frigate Tuna 95 Skipjack 23 3 24.21

Dolphin Fish 11 Yellowfin 7 3 63.64

Rainbow Runner 130

Mackerel Tuna 79

Bigeye Tuna 5

Dogtooth Tuna 1

Spanish Mackerel 3

Barracuda 1

TOTALS 981 102 TABLE 7. THE DOMINANT TEN SPECIES OF BAITFISH CAUGHT WITH BOUKI-AMI GEAR B PROGRAMME IN SOLOMON ISLANDS

Bait Species 1977 1980 Total

Total Kg per % per Total Kg per % per Total Kg per Kg Haul Haul Kg Haul Haul Kg Haul

Stolephorus devisi 676 32 23 769 20 12 1445 25

Herklotsichthvs punctatus 241 12 9 1140 30 19 1381 23

Stolephorus heterolobus 305 14 10 1008 27 17 1313 22

Rastrelliger kanaeurta - - - 945 25 16 945 16

Spratelloides gracilis 476 23 17 57 2 1 533 9

Gvmnocaesio evmnopterus 471 22 17 22 1 1 493 8

Spratelloides (new species) 286 14 10 85 2 1 371 6

Stolephorus buccaneeri - - - 498 13 8 498 8

Sardinella sirm 39 2 1 322 8 5 361 6

Stolephorus bataviensis - - - 279 7 4 279 5

Total Loaded Alive 2772 132 5634 148 8406 143 Grand Total Caught 2882 137 87** 6084 160 84 8966 152 Hauls 21 38 59 Nights 14 19 33 Total Catch per Night 206 320 272

* Average of 1977 and 1980 kg per haul. ** The above ten species acounted for 87 per cent of the grand total caught. 19

Column 11 of Table 7 presents the average of 1977 and 1980 catch per haul figures. The Programme's bouki-ami catches averaged 148 kg per haul over the two surveys; anchovies (38%), followed by sardines (17%) and then sprats (13%) dominated the catch. Two stolephorid anchovies, Stolephorus devisi and S. heterolobus. both very effective bait species, accounted for approximately 32 per cent of the catch from the two visits. Another effective bait species, the gold-spot herring (Herklotsichthvs punctatus) . accounted for a further 14 per cent. Together, catches of these three species averaged 67 kg per haul and were among the dominant species in the catch at 13 of the 18 baiting locations. Two species of sprats in the genus Spratelloides. both very effective bait for pole-and-line fishing, made up just over 13 per cent of the catch. The scombrid, Rastrelliger kanagurta, accounted for eight per cent of the average catch per haul. Large specimens of this species are of little value as bait, and fortunately they were only abundant at one location, Austria Sound, during the second survey. In general, species from the anchovy, sardine and sprat families, which dominated catches, exhibit behavioural characteristics that make them effective baitfish for pole-and-line fishing (Baldwin 1977, Smith 1977, Argue and Hallier MS).

The Programme covered a variety of baiting locations in Solomon Islands, some of which are seldom, if ever, fished by the commercial fleet. As a consequence of this exploratory strategy, only three localities were visited in both years, Thousand Ships Bay at the southern end of Santa Ysabel Island, and Byron Bay and Mangadai Bay in the Santa Cruz Islands. Catches per haul, and species composition, did vary somewhat between years at each location (Table 8), but taken together, results were reasonably similar for the two visits.

Bait catches by the commercial fleet in the waters of Solomon Islands have shown little seasonal variation. Based on 1976 to 1980 data supplied by the Solomon Islands Ministry of Natural Resources, catches per boat-night averaged 167 kg from January through March, 159 kg from April through June, 192 kg from July through September and 196 kg from October through December. The average bouki-ami catch by the Skipjack Programme's research vessel in 1977 was 206 kg per night, quite similar to the fleet average of 227 kg for November 1977. During the second visit the Programme's vessel averaged 320 kg per night, more than double the estimated fleet average of 134 kg for June 1980 (Solomon Islands Ministry of Natural Resources, unpublished data). The difference between surveys arose in part because the research vessel focused on the more productive baiting sites in 1980 (20 of 38 sets were made in Sandfly Bay and Thousand Ships Bay).

6.1.2 The Slot and vicinity

Baiting results for both years are combined in Table 9 to illustrate the Programme's average bait catch and species composition for six general areas in the vicinity of The Slot: New Georgia Group, Russell and Florida Islands, Choiseul Island, Santa Ysabel Islands, Guadalcanal Island, and Malaita Island. The figures for each area in Table 9 were averaged over the sites fished within each area (Table 3), in order to give equal weight to each site regardless of the numbers of sets per site.

The Programme's bait catches at sites in the New Georgia Group, on Santa Ysabel Island and on Guadalcanal Island, were generally of sufficient quantity and comprised a good mixture of species to enable high daily pole-and-line catches when skipjack were abundant. Thousand Ships Bay produced consistent catches of stolephorid anchovies almost to the exclusion TABLE 8. A COMPARISON OF SKIPJACK PROGRAMME BAITFISH CATCHES FOR REPEAT VISI BAITING LOCALITIES IN SOLOMON ISLANDS

Baiting Local ity 1977 1980

Bait Species Average Bait Species Catch per Haul (kg) (%)

Thousand Ships Bay Stolephorus devisi 105 (48) Stolephorus devisi (S Santa Ysabe 1 Is) Stolephorus heterolobus 81 (37) Stolephorus hetero Stolephorus buccan 1 haul Total 219 (85)* Rastrelliger faugh Sardinella melanur

14 hauls

Byron Bay Stolephorus devisi 53 (79) Thrissina baelama (Utupua Is) Pranesus pinguis 4 ( 6) Hypoatherina ovalau Hvpoatherina temmincki 4 ( 6) Herklotsichthvs pun

2 hauls Total 67 (91) 4 hauls

Mangadai Bay Spratelloides (new species ) 45 (80) Herklotsichthvs pu (Vanikoro Is) Stolephorus devisi 7 (13) Spratelloides (new Spratelloides gracilis 3 ( 5) Spratelloides grac

2 hauls Total 56 (98) 2 hauls

Average Catch per 114 Haul per Site

* The two Stolephorus species accounted for 85 per cent of the average total catch of 219 kg per haul of all bait spe cies. 21

TABLE 9. AVERAGE BAIT CATCH AND SPECIES COMPOSITION FOR BAITING AREAS FISHED BY THE SKIPJACK PROGRAMME IN THE VICINITY OF THE SLOT, SANTA CRUZ ISLANDS AND ONTONG JAVA ATOLL

Bait Species, Average Kg Caught Bait Species, Average Kg Caught Baiting Localities per Haul (% Total) Baiting Localities per Haul (% Total) (number of hauls) (number of hauls)

NEW GEORGIA GROUP SANTA YSABEL

Herklotsichthvs punctatus 56 (34) Stolephorus devisi 47 (26) Spratelloides gracilis 36 (22) Stolephorus heterolobus 41 (23) Spratelloides (new species) 24 (15) Rastrellieer kanaeurta 27 (15) Stolephorus heterolobus 16 (10) Stolephorus buccaneeri 11 ( 6) Stolephorus bataviensis 15 ( 9) Sardinella sirm 10 ( 6) Sardinella melanura 6 ( 4) Herklotsichthvs punctatus 10 ( 6) Spratelloides delicatulus 2 ( 1) Rastrellieer fauehni 1 (<1) Gvmnocaesio evmnopterus 1 «1) Sardinella melanura <1 (<1)

Total 165 (95)* Total 181

Ghizo (1) Kolokofa Harbour (5) Marovo Lagoon (2) Thousand Ships Bay(15) Sandfly Bay (6) Austria Sound (4)

RUSSELL AND FLORIDA ISLANDS GUADALCANAL ISLAND

Stolephorus heterolobus 23 (48) Gvmnocaesio evmnopterus 63 (48) Spratelloides gracilis 11 (23) Herklotsichthvs punctatus 36 (27) Stolephorus buccaneeri 4 ( 8) Dipterveonotus leucoerammicus 8 ( 6) Herklotsichthvs punctatus 4 ( 8) Spratelloides (new species) 7 ( 5) Gvmnocaesio evmnopterus 1 ( 2)

Total 48 (89) Total 132 (86)

Macquitti Bay (1) Haoto Bay (1) Gavutu Harbour (2) Danae Bay (4)

CHOISEUL ISLAND MALAITA ISLAND

Rastrellieer kanaeurta 313 (66) Sardinella sirm 59 (43) Scomberoides tol 66 (14) Rastrellieer fauehni 36 (26) Sardinella sirm 42 ( 9) Decapterus maruadsi 16 (12)

Total 473 (89) Total 136 (81)

Labaka Bay (2) Are Are Lagoon (2)

SANTA CRUZ ISLANDS ONTONG JAVA

Spratelloides (new species) 13 (29) Spratelloides gracilis 233 (98) Herklotsichthvs punctatus 10 (21) Stolephorus devisi 6 (13) Thrissina baclama 3 ( 7) Hvpoatherina ovalaua 2 ( 4) Spratelloides gracilis 2 ( 4) Franesus pinguis 1 ( 2) Stolephorus bataviensis 1 ( 2)

Total 45 (84) Total 237 (98) 539

Mangadai Bay (4) Languia Island (1) Byron Bay (6) Graciosa Bay (2)

* The above species accounted for 95 p =r cent of the average total catch of 165 kg per haul of all bait species. 22 of other species. In contrast, catches at three localities in the New Georgia Group were dominated by gold-spot herring, sardines (Sardinella melanura) and sprats (Spratelloides spp.), with lesser catches of stolephorid anchovies. Two bouki-ami hauls at Wickham Harbour on Vangunu Island were aborted due to strong currents, although the echo-sounder trace indicated presence of considerable bait. Stolephorus dey i s i , S to 1ephor us heterolobus and Spratelloides gracilis were present in a sample from one of the aborted hauls.

In the opinion of the Programme's scientists the two baiting localities in Marau Sound, Guadalcanal Island had good habitat for bait (weed beds, shallow areas with mud bottom [see Hallier and Gillett 1982 and papers in Shomura 1977 for discussion of baitfish habitat]). The relatively low catches and the absence of any quantity of anchovies in the catch was disappointing, but not surprising since the Programme's baiting in this area took place during full moon conditions. Similar moon conditions and poor catches were encountered while baiting for two nights in Gavutu Harbour, Florida Island. A distinctive sounder trace in Macquitti Bay, Russell Island, indicated the presence of considerable quantities of ^_ buccaneeri. which unfortunately failed to rise when the bait lights were dimmed. Good catches at one locality on Choiseul Island and one locality on Malaita Island were dominated by large individuals (124 to 167 mm) in the families Clupeidae, Carangidae and Scombridae, a somewhat disappointing result since in the case of Are Are Lagoon, Malaita Island, the habitat again looked suitable for good catches of anchovies. Anchovies were present on the sounder trace and S. devisi occurred in the catch from one haul. Suitable habitat and presence of anchovies suggest that good catches would be possible from Are Are Lagoon.

6.1.3 Other baiting localities

Twelve hauls of the bouki-ami net were made at three localities in the Santa Cruz Islands (Tables 3 and 9). The average catch per haul was only 47 kg; however, this value is somewhat influenced by poor catches at one location, Graciosa Bay, and by two occasions at Byron Bay, Utupua Island, when the above-water light malfunctioned, thus reducing catches of sprats (Spratelloides spp.) and hardyheads (Pranesus pinguis. Hypoatherina ovalaua). It should be noted that hardyheads are not particularly attractive to skipjack, although they survive well in bait wells (Skipjack Programme 1981d).

Stolephorid anchovies were present in catches from Byron Bay and Mangadai Bay, Vanikoro Island, which was an encouraging result. Gold-spot herring were also present in most of the hauls, and a school of approximately 150 kg was sighted in Byron Bay on 2 December 1977 while the crew was experimenting with operation of the beach seine. However, this species has been shown to undergo significant natural fluctuations in abundance in the Marshall Islands (Hida and Uchiyama 1977) and in Kiribati (Kleiber and Kearney MS); and in Palau there was a marked decline in abundance of this species in apparent response to fishing pressure (Johannes 1981).

Since none of the Santa Cruz localities are particularly large, and total catches only once exceeded 100 kg per haul, it would be unreasonable to expect the Santa Cruz Islands to support sufficient stocks of baitfish for more than a few vessels operating on a periodic basis. The lagoons at some of the Duff and Reef Islands may contain fishable quantities of bait, and this possibility should be examined if pole-and-line fishing is contemplated amongst the Santa Cruz Islands. TABLE 11. TUNA SCHOOL SIGHTINGS PER HOUR PER FISHING DAY AND TOTAL CATCHES PER FISHIN SURVEYS BY THE SKIPJACK PROGRAMME RESEARCH VESSEL IN THE WATERS OF SOLOMON Slot and vicinity includes fishing time on the south coast of Guadalcanal I Shortland Islands, in Indispensible Strait and between Malaita Island and Island.

Days Hours SCHOOLS SIGHTED PER HOUR CATCH (KG) PER FI Fishing Sighting on Fishing Days Skipjack Yellowf in Total Skipjack Yellowf

1 9 7 7

The Slot and vicinity 21 212 .26 .17 .45 338 27 Ontong Java 3 34 .12 .18 .47 2 11 Santa Cruz Islands 4 34 .15 .12 .32 475 5 Total for Solomon 28 280 .23 .16 .44 322 22 Islands

1 9 8 0

The Slot and vicinity 18 140 .23 .11 .61 555 134 Santa Cruz Islands 5 42 .26 .19 .90 148 166 Total for Solomon 23 182 .24 .13 .68 467 141 Islands

BOTH YEARS

TOTAL FOR SOLOMON 51 462 .23 .15 .53 387 76 ISLANDS 23

Only one haul was made at Ontong Java, under bright moon conditions, but this haul produced an exceptional catch of sprats (Spratelloides gracilis). Atoll populations of sprats of this genus are thought to undergo high natural variability in recruitment, and may be vulnerable to overfishing. Although Ontong Java is a large atoll with possibly a number of good baiting sites, it would be premature to suggest on the basis of the limited data available, that it could support substantial continuous fishing.

6.2 Tuna Fishing Results

The Solomon Islands Ministry of Natural Resources has an excellent catch and sampling data base for the commercial fishing area by one-half degree square going back to the commencement of the pole-and-line fishery in 1971. There is little that the Programme's fishing results could add to this data base; consequently, detailed comment on daily fishing results is limited to the 17 survey days (12 days fishing, 5 days spotting) spent in the vicinity of Ontong Java and the Santa Cruz Islands (Section 6.2.2), two unfished locations that the Programme was specifically asked to survey by the Government of Solomon Islands.

6.2.1 General observations

During the 1977 survey period, catches by the Hatsutori Maru No.l« along with those of the Solomon Taiyo pole-and-line fleet, were poor (Table 10). In the preliminary Solomon Islands country report, Kearney and Lewis (1978) noted that skipjack availability appeared to decrease markedly in the vicinity of the Shortland Islands over a short period (31 October to 4 November), and that low apparent abundance of skipjack was common to both the Papua New Guinea and Solomon Islands fisheries throughout November. Poor commercial catches continued well into 1978. Commercial catch per unit effort CPUE was also very low in June 1980 by comparison with the same period in previous years (Table 10); however, CPUE for subsequent months was generally above average.

TABLE 10. CATCH (TONNES) PER FISHING DAY FOR THE SOLOMON ISLANDS POLE-AND-LINE FLEET IN NOVEMBER AND JUNE, 1976 TO 1980, AND FOR THE SKIPJACK PROGRAMME TAGGING VESSEL IN NOVEMBER 197 7 AND JUNE 1980 IN THE COMMERCIAL FISHING AREA

Year Commercial Pole -and-Line Estimated Commercial Catch by Research Vessels

June November June November

1980 2.10 5.95 2.59 1979 5.61 5.15 1978 3.03 5.70 1977 2.41 2.41 1.27 1976 6.17 4.96

Average 3.86 4.83

Within the commercial fishing area in 1977 the Hatsutori Maru No.l caught an average of only 367 kg per day (Table 11), which converts to an estimated commercial catch per day of 1.3 tonnes using the 3.47:1 research vessel: commercial vessel conversion ratio developed during the Programme's 25

survey in Fiji (Kearney 1978). This was lower than the 2.1 tonnes per day average by the commercial fleet. In 1980, the larger Hatsutori Maru No.5 averaged 2.6 tonnes per day (747 kg x 3.47) in the commercial fishing area, compared to 2.1 tonnes per day for the commercial fleet. If allowance is made for catches of small skipjack (<1.5 kg) by the Hatsutori Maru No. 5. which are generally avoided by the commercial fleet, the research vessel's CPUE in 1980 was approximately the same as that for the commercial fleet.

The crew on the Hatsutori Maru No.l sighted only 0.44 schools per hour in 1977 (Table 11), well below the Programme average of 0.77 schools per hour for all countries and territories. Total schools sighted per hour increased during the second Solomon Islands survey to 0.68, but many of these schools were unidentified (52% unidentified in 1980 versus only 22% unidentified in 1977). Sighting rates of schools in which skipjack were identified were approximately the same for the two Programme visits, as were sighting rates for schools in which yellowfin were identified. Over both visits, yellowfin were identified in an average of 21 per cent of the schools sighted.

The difference in the Programme's catch rates between the two survey visits was probably due to a number of factors. Total school sightings suggest that abundance was considerably higher in 1980, but because of the large number of unidentified schools this probably overstates the difference between years. The amount of bait carried would not appear to be a factor, since during both years the vessel carried more than enough good quality bait to enable equivalent commercial catches of over 3.5 tonnes per day. On the other hand, skipjack response to bait was apparently poorer in 19771 The crew chummed approximately the same number of schools each visit, yet chumming success was considerably lower in 1977 than in 1980 (Table 12). It follows that weight of fish caught per school chummed was less in 1977, compared to 1980. Schools that were enticed to bite produced lower catches during the first visit (compare catches per positive school in Table 12). Possibly poor chumming success and CPUE in 1977 were influenced by fishing tactics at this time. During the first visit the crew, skipper and scientists were only two months into the Programme and were still experimenting with fishing techniques that would optimise catches for the purpose of tagging and scientific data collection. On the other hand, many of the schools in 1977 were "breezing" or "boiling" while apparently feeding on natural food (Stolephorus buccaneeri). and could not be enticed to bite. As well, skipjack gonad indices were significantly higher in November 1977 compared to June 1980 (Section 6.3.1). On this latter point, Lewis (1981) presents an interesting inverse relationship between CPUE and maturity index in the Papua New Guinea pole-and-line fishery. Lowest CPUE occurred in that fishery between October and March when gonad indices were highest, suggesting that skipjack catchability may be less during periods when spawning is more prevalent, which is another partial explanation for poor catches in November 1977.

6.2.2 Ontong Java

On 21 November the research vessel left The Slot via Manning Strait for Ontong Java atoll (see Figure 1) carrying 194 kg of bait, predominantly stolephorid anchovies and sprats. A number of schools were sighted on the way to and around Ontong Java; however, none of the 13 schools that were chummed responded positively, and only a few fish were trolled. Four of the schools sighted in the vicinity of Ontong Java were identified as yellowfin schools. Skipjack may have been mixed with these schools; however, poor response to chum prevented confirmation of skipjack presence. On 24 November the research vessel spent 12 hours fishing in the vicinity of Roncador Reef, 26

approximately 40 nautical miles due south from Ontong Java. Three schools were sighted, one of which was a very large school of skipjack that was rippling at the surface. Again the crew encountered a negative response to chum. During a total of 34 hours' fishing the crew sighted an average of 0.47 tuna schools per hour near Ontong Java and Roncador Reef (Table 11), which is a similar rate to that achieved in The Slot. Bait was not a limiting factor during this brief survey, for, as well as carrying a good quantity of bait to the atoll, the vessel made large bouki-ami catches of sprats near Languia Island within the perimeter of the atoll.

TABLE 12. INDICES OF FISHING PERFORMANCE FOR TUNA SCHOOLS FISHED BY THE SKIPJACK PROGRAMME RESEARCH VESSELS DURING 1977 AND 1980 SURVEYS IN THE WATERS OF SOLOMON ISLANDS

Number of Positive Chumming Catch(kg) Catch(kg) Schools Schools* Success per School per Positive Chummed Chummed School

19 7 7

The Slot and vicinity 57 26 45.6% 135 299 Santa Cruz Islands and Ontong Java 16 4 25.0% 123 491 Total for Solomon 73 30 41.1% 133 323 Islands 19 8 0

The Slot and vicinity 54 33 61.1% 249 408 Santa Cruz Islands 17 6 35.3% 94 268 Total for Solomon 71 39 54.9% 213 387 Islands

BOTH YEARS

TOTAL FOR SOLOMON 144 69 47.9% 172 359 ISLANDS

* A positive school is defined as a school from which at least one fish was poled on board the research vessel.

There is no reason to believe that skipjack are any more or less abundant near Ontong Java than in The Slot. In past years Japanese distant-water pole-and-line vessels have made excellent skipjack catches averaging 13 tonnes per day in the vicinity and to the north of Ontong Java (Tuna Programme, South Pacific Commission, unpublished data). On the other hand, Ontong Java is at least two days return travel from the main commercial fishing and baiting areas in Solomon Islands. Thus, sustained fishing there by small Solomon Islands vessels would in all likelihood require some local baiting, which, as previously mentioned, may not give consistent results.

6.2.3 Santa Cruz Islands

The Skipjack Programme vessel completed nine fishing days amongst the Santa Cruz Islands during the visits to the waters of Solomon Islands. In early December 1977 only 0.32 schools were sighted per hour (Table 11), but total tuna catches were somewhat higher than they were in and near The Slot. 27

Sixteen schools were chummed; four responded positively (Table 12). Two of the schools were encountered approximately 20 nautical miles south-east of Nendo Island; each produced catches equivalent to a commercial harvest of over three tonnes.

In late May 1980 the crew of the research vessel spotted 38 unidentified tuna schools and one school with yellowfin during 28 hours of spotting (Table 2) while steaming in rough seas without bait from Futuna Island (Wallis and Futuna), past Tikopia Island and through the Santa Cruz Islands to Honiara. On 1 June the research vessel, on its way to the Santa Cruz Islands, carried 228 kg of bait (Rastrelliger kanagur ta, Stolephorus heterolobus, Sardinella sirm) caught at Are Are Lagoon on the south-east coast of Malaita Island and at Gavutu Harbour in the Florida Islands. Over the next five fishing days, generally poor catches were experienced in the vicinity of Nendo, Vanikoro and Utupua Islands, albeit under rough sea conditions. There was no lack of tuna schools as shown by sightings data (Tables 2 and 11), and from 2 to 6 June the vessel carried sufficient bait, more than during the 1977 Santa Cruz survey. It was the opinion of scientists and crew that the low catches reflected weather conditions and behaviour of the schools, rather than any lack of skipjack.

While leaving Solomon Islands at the end of the 1980 survey the crew spotted only three schools during 15 hours of sighting while steaming through the Duff Islands towards Funafuti atoll in Tuvalu (Table 2). At the end of a survey, particularly when steaming to the next country without bait, and in uncomfortable seas, the crew's enthusiasm for sighting wanes; thus searching efficiency was probably below average through the Duff Islands.

Skipjack abundance amongst the Santa Cruz Islands would not appear to be a limiting factor to commercial operations. However, approximately twice as much travel time would be required between the present commercial fishing area and the Santa Cruz Islands, as compared to between Ontong Java and the commercial grounds. Thus, for the smaller pole-and-line vessels currently fishing in Solomon Islands, operations amongst the Santa Cruz Islands would more than likely be dependent on basing vessels in this area.

6.3 Skipjack Population Biology

Several aspects of skipjack biology were considered by the Skipjack Programme. These included sexual maturity, juvenile ecology, feeding, growth and population structure. Mortality, production and migration are discussed later in Section 6.4.

6.3.1 Maturity and juvenile recruitment

Figure 4 presents combined maturity data for both visits to Solomon Islands. The upper graph in this figure shows that 45 cm is a reasonable separation point between female skipjack with immature gonads (stages 1 and 2) and fish with maturing (stage 3) or mature/ripe/spent and recovering (stages 4/5/6,7) gonads. Stage 3 skipjack dominated the sample, as they do in most samples from pole-and-line catches in the tropical western Pacific. As shown below, there were also significantly more CX2=30.1, p<0.01) immature female skipjack in the June 1980 sample from Solomon Islands. Presence of skipjack females with mature and spent gonads implies that some spawning took place during both visits. 28

IMMATURE MATURE SPENT SAMPLE SIZE

November 1977 9% 89% 2% 128

June 1980 37% 53% 10% 139

Two authors (Naganuma 1979, Lewis 1981) present monthly trends in maturity indices that suggest that skipjack spawning is less common from April through September in tropical waters south of the Equator. For Solomon Islands data, gonad indices(l) were calculated for two size ranges of female skipjack (40 to 49.9 cm and 50 to 59.9 cm), since gonad indices for female skipjack tend to increase with increased fish size. Females caught during June had, in both cases, average gonad indices (17.7 and 43.2) that were significantly lower (t7Wf=8.51, t157df =5.91, p<0.01) than average values (52.1 and 63.8) for skipjack caught during November. These results are consistent with the more continuous data series of Naganuma and Lewis and suggest that skipjack spawning is least intense during winter months in tropical waters south of the Equator.

An index of spawning activity and recruitment is given by the incidence of skipjack juveniles observed in the stomachs of their predators (Table 6). An average of 9.4 skipjack juveniles per 100 skipjack predator stomachs was observed in Solomon Islands. This level is intermediate between the high values of 25-50 juveniles per 100 stomachs found at Vanuatu, Wallis and Futuna, and the Marquesas Islands, and the low levels of 0-4 juveniles per 100 stomachs found at the northern and southern extremes of the Programme study area and in the Society Islands.

Argue, Conand and Whyman (MS) discuss a more detailed analysis of the tuna juvenile data, taking into account size selective predation by adults, and the time of day, distance from land and season that adults were sampled. They suggest that occurrence of juvenile skipjack within the SPC region was highest in two areas during the period between the 1977 and 1980 surveys, one roughly bounded by Solomon Islands, Papua New Guinea and Vanuatu, and the other including the Marquesas and Tuamotu Islands. Skipjack juveniles also occurred most frequently in the stomachs of skipjack predators between October and March south of the Equator, which is roughly the period of maximum gonad development in skipjack. These results, along with the maturity data above, suggest that spawning was more intense during the Programme's first survey to Solomon Islands, and that Solomon Islands is in a region which has relatively high abundance of juvenile skipjack. However, as virtually nothing is known about the movements of juvenile skipjack, the extent to which local production contributes to recruitment in Solomon Islands and elsewhere cannot be established with confidence.

6.3.2 Skipjack diet

Common diet items of skipjack in Solomon Islands, other than chum and fish remains, were squid (cephalopoda), the alima stage of stomatopods,

1 Gonad index=107(gonad weight gm/(fish length mm) ). High index values, particularly over 50, are associated with skipjack whose gonads have a high percentage of eggs that are ready to be spawned (Raju 1964). 29 juvenile anchovy (probably Stolephorus buccaneeri), adult j^. buccaneeri and surgeonfish (Acanthuridae) (Table 5). S. buccaneeri was relatively common in skipjack stomachs from most countries west of 170°E longitude. Each of the above items occurred in over ten per cent of the stomachs examined. Species of tuna juveniles occurred in seven per cent of stomachs of adult skipjack examined for diet composition in Solomon Islands.

The wide variety of diet items observed indicates that skipjack are highly opportunistic feeders. Community groups of skipjack prey species are thought to vary across the SPC region, and identification of groups is the subject of ongoing analyses.

6.3.3 Skipjack growth

Growth of skipjack in the SPC region has been analysed by Lawson and Kearney (MS). Data on growth of tagged skipjack from Solomon Islands were an important part of their study. Table 13 presents growth rate estimates, by size at release and time-at-large, for tagged skipjack from Solomon Islands and several other countries. Growth rates, measured in centimetres per year, varied among areas, with smaller fish from Fiji growing fastest and larger fish from Kiribati and the Solomon Islands growing slowest. Slow growth of 40-49 cm skipjack from Solomon Islands, as contrasted with more rapid growth of fish of similar size from Papua New Guinea and Fiji, suggests that this size of skipjack would be vulnerable to the Solomon Islands fishery for a longer period of time; assuming of course that natural mortality and emigration are the same in all areas.

TABLE 13. ESTIMATES OF SKIPJACK GROWTH RATES FOR SEVERAL COUNTRIES IN THE SPC REGION, BY SIZE AT RELEASE AND TIME-AT-LARGE. Average growth rates with standard errors greater than 3 cm or for samples of less than six skipjack are considered unreliable, and are given in brackets.

Area of Release Size at Days at Sample Growth Standard Release Large Size Rate Deviation (cm) (cm/yr) (cm)

Fiji 40-49 31-180 38 17.23 14.89 Fiji 50-59 31-180 12 (11.95) 20.79 Fiji 40-49 181-450 20 16.6 3.91 Fiji 50-59 181-450 10 7.01 6.10

Kiribati (Gilbert Is) 40-49 31-180 180 9.46 9.96 Kiribati (Gilbert Is) 50-59 31-180 39 1.42 12.78 Kiribati (Gilbert Is) 40-49 181-450 1 (5.43) - Kiribati (Gilbert Is) 50-59 181-450 0 - -

Papua New Guinea 40-49 31-180 16 (20.85) 14.47 Papua New Guinea 50-59 31-180 292 5.40 11.75 Papua New Guinea 40-49 181-450 3 (19.38) 7.70 Papua New Guinea 50-59 181-450 15 8.23 2.45

Solomon Islands 40-49 31-180 87 12.72 11.23 Solomon Islands 50-59 31-180 42 5.75 18.43 Solomon Islands 40-49 181-450 77 11.37 7.90 Solomon Islands 50-59 181-450 50 4.08 6.35 30

Temporal variation was observed in growth rates estimated from data for several years of tagging in Papua New Guinea both by the Commission (Lawson and Kearney MS) and the Department of Primary Industry, Papua New Guinea (Josse et al. 1979). Geographic and temporal variability of skipjack growth is thought to reflect variation in environmental conditions, but as yet, neither the degree of environmental heterogeneity nor the precise effects of the environment on skipjack growth is well understood.

6.3.4 Population structure

Blood Genetics and Tagging Results

There is movement of some skipjack adults over much of the western and central Pacific (see Figure B), and such movement suggests that genetic exchange is possible among all countries within the Programme's study area. On the other hand, the fishery interaction analysis (Section 6.4.4 below) suggests that the actual level of exchange, for skipjack of the size caught by pole-and-line gear, was generally low, at least among the locally based fisheries.

Results from electrophoretic analysis of skipjack blood samples show a gradient in esterase gene frequency, a genetic marker used to infer population structure, from west to east across the Pacific between approximately 120°E and 120°W (Figure 5). The esterase gene frequency for the single sample taken in the waters of Solomon Islands was somewhat above the regression line in the Figure, which depicts the average gene frequency one would expect at any particular longitude between 120°E and 120°W, but was within the 95 per cent prediction limits for the regression line. There was considerable variation in individual esterase gene frequency values along this average line, although the cause of this variability was unclear (Anon 1981b).

Several population structure models are consistent with the tagging and blood genetics data (Anon 1981b). One such model, called the clinal population structure model, has as a basic premise that the probability of breeding between skipjack is inversely proportional to the distance between them. Acceptance of this model implies that there are no genetically isolated skipjack subpopulations in the study area, which is contrary to previous hypotheses of Fujino (1970, 1976) and Sharp (1978).

The gradient in esterase gene frequency is consistent with a relatively even distribution of skipjack spawning in tropical waters across the study area. One could also view the gradient as a region of "overlap" of skipjack from two or more centres of higher spawner density at the approximate extremes of the study area or beyond. The similarity between eastern Pacific esterase gene frequencies (to right of the dotted line in Figure 5) and those from French Polynesia suggests that eastern Pacific skipjack have the same genetic origin as skipjack in French Polynesia, and thus would collectively represent the group at one extreme. Occurrence of skipjack juveniles (Argue et al. MS) also appeared highest at the longitudinal extremes of the Programme study area, thus lending some support to this latter view of the distribution of skipjack spawning.

After two workshops hosted by the Skipjack Programme to examine the question of skipjack population structure, it was concluded that due to limitations of the extant blood genetics, tagging and ancillary data, it is difficult to choose between the various population structure hypotheses (Anon 1981b; Skipjack Programme 1981e). However, the genetics data supported the FIGURE 5. SKIPJACK SCHOOL SERUM ESTERASE GENE FREQUENCY VERSUS LONGITUDE OF THE SAMPLE L The circle represents the esterase gene frequency for the single sample from Islands. The regression line on the left of the dotted line includes 145 collected between Palau and the Marquesas Islands (correlation coefficient Esterase gene frequencies for 18 eastern Pacific samples are shown to the righ dotted line.

100E 120E 140E 160E 180 1 60W 1 40W 120W 10O LONGITUDE 32 conclusions that there should be minimum short-term interactions between fisheries at the extremes of the Programme's study area, and that the potential for interactions should increase as the distance between fisheries decreases. The exploitation rates of adjacent fisheries must increase substantially over present levels, and the distance between centres of exploitation must decrease substantially before interactions will be of overriding importance.

Parasite Results

Samples of skipjack gills and viscera were collected commencing in December 1979, during the Programme's last survey year. Samples were taken over a wide range of tropical waters, and also from subtropical waters of New Zealand and Norfolk Island. Preliminary results presented by Lester (1981) suggest that skipjack from New Zealand have a different overall parasite fauna to that of skipjack from tropical waters to the north, and that the parasite fauna of tropical samples from widely separated areas are quite similar.

Analyses of parasite data are continuing; however, preliminary results do not hold out much hope for clarifying regional fishery interactions in tropical waters based on parasite fauna, nor is it likely that definition of skipjack population structure will be greatly improved based on further analysis of the existing parasite data.

6.4 Skipjack Tagging Results

This section first presents general results from skipjack tagging in Solomon Islands, followed by comment on international recoveries of Solomon Islands tagged skipjack, and recoveries by Solomon Islands fisheries of skipjack tagged in other countries. The section concludes with an analysis of tagging data to estimate skipjack mortality, population size, recruitment and fishery interactions. Readers are referred to Skipjack Programme (1981c) for a more detailed presentation of the Programme's approach to analysis of tagging data.

The cut-off date for tag recoveries presented in this report was 1 March 1982. The Programme had received less than 50 tags from all releases in the six-month period preceding March, and considerably fewer tags are expected in the future. Thus this report includes virtually all recoveries that are likely to accrue from tag releases in the waters of Solomon Islands.

6.4.1 Solomon Islands tagging

Skipjack were tagged during two time periods, 25 October to 3 December 1977 and 2 to 23 June 1980. Hereafter, 1977 tagging is referred to as having taken place in November, since nearly three-quarters of the tags were released in this month. Table 14 gives the numbers of tags that were released each year and the number of recoveries from these releases, divided into local recoveries, international recoveries (outside the Solomon Islands fishery zone) and recoveries for which recovery location could not be determined.

Eighty-eight per cent of the tag recoveries from both years were from local fisheries, three per cent were from international fisheries and nine per cent were from fisheries of unknown location. There were 53 tag returns in this latter category. Most of these (49) were recovered from transhipped fish that were handled at processing plants in Puerto Rico, California and 33

American Samoa. Efforts to determine accurate recapture information for these fish proved fruitless. However, in the process of checking bills-of-lading and other transhipment and company records, it became apparent that the majority of these recoveries were from skipjack that must have been caught by the Solomon Islands pole-and-line fleet, but had not been noticed by the fishermen "t time of capture or transhipment.

TABLE 14. RELEASES OF TAGGED SKIPJACK DURING 1977 AND 1980 IN THE WATERS OF SOLOMON ISLANDS, NUMBERS OF RECOVERIES MADE INSIDE AND OUTSIDE THE SOLOMON ISLANDS FISHERY ZONE, AND NUMBERS OF RECOVERIES FROM UNKNOWN RECOVERY LOCATIONS. Percentage of tags recovered in brackets.

1977 1980 TOTAL

Numbers Tagged 2490 3731 6221 Local Recoveries 82(3.3) 439(11.8) 521(8.4) International Recoveries 5(0.2) 13(0.4) 18(0.3) Unknown Location* 3(0.1) 50(1.3) 53(0.9) Total Recoveries 90(3.6) 502(13.5) 592(9.5)

* 49 of these recoveries wer e from process ing plants.

Most tagged skipjack recovered in Solomon Islands were from local catches (Table 14), although the percentage of tags that were recovered differed greatly between the November 1977 and June 1980 releases (3.3% versus 11.8%). This resulted from several factors. First and probably most important, the November releases occurred at the end of the poor 1977-early 1978 fishing period. From January through March 1978 less than 50 tonnes were harvested; and catches did not exceed 1,000 tonnes per month until June. In contrast, tagging in June 1980 occurred just before local pole-and-line catches reached a peak, and the 1980 fishery continued at a high level (more than 2,000 tonnes per month) through December. Thus skipjack tagged in June were exposed to more fishing effort, particularly within the first few months after tagging when most recoveries normally occur.

A second contributing factor relates to the size of the tagged skipjack. There was a marked difference in the size composition of tagged skipjack between the two surveys. Skipjack tagged in November were from a unimodal size distribution, much the same as that for the commercial catch (Figure 6a). In contrast, skipjack tagged in June were from a broad size distribution with 36 per cent less than 45 cm in length (Figure 6b). The decrease in tag recovery rate, by two centimetre interval (crosses in Figures 6c and 6d) for fish over approximately 43 cm in June, suggests that natural mortality may increase with increased skipjack size or that vulnerability to capture may decrease with increased size. In either case, the smaller skipjack tagged in June would have been available in relatively greater numbers for the months after tagging than would the larger skipjack tagged in November, hence contributing to the higher tag recovery rate for tags released in June.

Tagging in the waters of Solomon Islands took place over a wide area (see Figure 1), in some cases well removed from the main commercial fishery. Table 15 shows the tagging data for subdivisions of the tagging area. The Slot includes tag releases just south of Vangunu and Russell Islands. Within the commercial fishing area (The Slot, Shortland Islands, Indispensable Strait), tag releases were distributed approximately in proportion to the FIGURE 6. LENGTH FREQUENCY DISTRIBUTIONS FOR SAMPLED SKIPJACK FROM THE POLE-AND-LINE CATCH DURING THE MONTH OF TAGGING, CONTRASTED WI DISTRIBUTIONS OF TAGGED SKIPJACK THAT WERE RELEASED IN THE COMMERCIA AND 6b). In 6c and 6d the Xs denote the tag recovery percentages fo within two centimetre length intervals. Mean lengths, standard devi (SD) and sample sizes (N) are indicated.

6a 20.00 NOVEMBER 1977 20.00- JUNE 1980

Commercial SPC 52.3 51.8 3.96 3.42 15.00- 15.00- 5566 1612 ui

5.00 5.00-

1 0.00 0.00- "T 1- 30 65 70 30 35 40 45 50

6c 6d 20.00- 20.00- X

X

x x

U l b o 15.00- RECOVERE D en o

U. 10.00- 10.00- x o X X

* X

U l b o XX X 5.00- X PERCENTAG E

X X

0.00- 0.00- 1 1 1 1 30 35 40 45 50 55 60 65 70 30 35 40 45 50 FORK LENGTH IN CENTIMETRES FORK LENGTH IN C 35 distribution of fishing effort. The obvious feature in Table 15 is the higher recovery rate for The Slot, which is to be expected since this is the area of highest commercial fishing pressure. Smaller numbers of tags released in Indispensable Strait and Shortland Islands, towards the south-eastern and north-western extremes of the fishing area, have corresponding lower recovery rates. Two skipjack were recovered from Santa Cruz tagging, seven and twelve months after tagging, by Solomon Taiyo vessels fishing in Indispensable Strait approximately 400 nautical miles to the west of the Santa Cruz Islands.

TABLE 15. NUMBERS OF SKIPJACK TAGGED, TOTAL RECOVERED AND PERCENTAGE OF TAGS RECOVERED FOR SUB-AREAS IN THE WATERS OF SOLOMON ISLANDS. Tag recoveries to 1 March 1982 include those with unknown date and location of capture.

Shortland The Slot South Coast Ind ispensable Santa Total Islands and Guadalcanal Strait Cruz vicinity Island Islands

1 9 7 7

Numbers Tagged 279 1698 96 _ 417 2490 Numbers Recovered 2 84 - - 4 90 Percentage Recovered .72 4.95 .96 3.62

1 9 8 0

Numbers Tagged 183 2558 - 804 186 3731 Numbers Recovered 13 413 - 73 3 502 Percentage Recovered 7.10 16.15 — 9.08 1.61 13.46

6.4.2 International migrations from Solomon Islands

Figure 7 depicts the straight line migration trajectories for the 18 Solomon Islands tagged skipjack that made international migrations. Table 16 shows the numbers and percentage of tag releases that were recovered outside Solomon Islands waters. Complete tagging and recovery information for all 18 fish are listed at the beginning of Appendix C.

TABLE 16. INTERNATIONAL RECOVERIES OF TAGGED SKIPJACK FOR SUBDIVISIONS OF THE SOLOMON ISLANDS TAGGING AREA. For explanation of country abbreviations see Appendix C.

Tagging Location Total International Percentage of Tagged Recoveries International Recoveries

Shortland Islands 462 2 PNG .43 The Slot 4256 10 PNG .24 S. Guadalcanal Island 96 - - Indispensable Strait 804 3 PNG .49 1 IND Santa Cruz Islands 603 1 KIR .33 1 FIJ

TOTALS 6221 18 .29 36

FIGURE 7. MIGRATION ARROWS FOR 18 OF 23 SOLOMON ISLANDS TAGGED SKIPJACK THAT MADE INTERNATIONAL MIGRATIONS. Tick marks denote intervals of 30 days between release and recapture dates. In the absence of accepted zones for all countries, 200-mile boundaries were estimated by staff of the Tuna Programme. The boundaries shown should not prejudice any boundaries that may be derived in the future.

135E 140E 145E 150E 1S5E 160E 165E 170E 175E 180 175W 37

There were international recoveries from tagged skipjack released at four of the five tagging locations, with no evidence that any one location produced significantly more international migrants (column 3, Table 16); most of these migrants moved in a northwesterly direction. Of interest, the only evidence of long distance movement to the east was from the Santa Cruz area where one tagged skipjack was recovered five months later in Fiji, a great circle distance of 819 nautical miles to the east; and another was recovered two months later in Kiribati, a distance of 604 nautical miles to the northeast. From Indispensable Strait one skipjack moved 1,370 miles to the northwest in six months (a net movement of eight nautical miles per day) to the waters of Irian Jaya, Indonesia, near its border with Papua New Guinea. A second skipjack from this tagging location moved even faster (18 nautical miles per day) in the same general direction to the Papua New Guinea side of the Indonesia border. Skipjack tagged in The Slot that went to the Papua New Guinea pole-and-line fishing area had a faster rate of net movement in 1977 (four fish, range 2.9-3.0 nautical miles per day) than in 1980 (nine fish, range 1.1 to 1.7 nautical miles per day). This may be an effect of seasonal variation in movement patterns. It would not appear to reflect fishing patterns, since there was similar effort in Papua New Guinea in the twelve months after the June 1980 tagging as there was in the twelve months after November 1977 tagging. Most Solomon Islands to Papua New Guinea migrants were recovered in the Bismarck Sea commercial fishing area, between Cape Lambert and New Hanover, six to thirteen months after tagging.

6.4.3 International migrations to Solomon Islands

Figure 8 shows straight line trajectories for 41 of 79 tagged skipjack that migrated into the waters of Solomon Islands. In this case recoveries have been selected to show no more than two examples of movement in each direction between any pair of five degree squares. Appendix C details tagging and recovery information for all 79 fish. Table 17 arrays these tag recoveries by country of tagging and month of recapture. Most recoveries (70) were well distributed throughout the Solomon Islands commercial fishing area and season. Seven recoveries were by Japanese distant-water pole-and-line vessels in extreme southern and eastern portions of the Solomon Islands fishery zone. These seven fish were tagged in Australia and New Caledonia.

From Table 17 it can be seen that "immigrant" skipjack tag recoveries were dominated by fish from Australia (32) and Papua New Guinea (26). Six of the Papua New Guinea fish were released in one school that was fished in May 1979 in the Solomon Sea, approximately 120 nautical miles southeast from Vitiaz Strait. Three Papua New Guinea tagged skipjack (October 1977) moved from the Solomon Sea, near the southern tip of Bougainville Island, just 240 nautical miles to their eventual recapture site in The Slot area. The longest migration into the waters of Solomon Islands was by a skipjack tagged 1,752 nautical miles to the west near Helen Reef in the waters of Palau. Six recoveries of skipjack from a number of Queensland schools composed of large fish (average length 62 cm) occurred in the last two months of 1979, which coincided with the only significant occurrence of exceptionally large skipjack (~5% of catch sample >65 cm) in the 1979/1980 Solomon Islands catch samples (Solomon Islands Ministry of Natural Resources, unpublished data).

Migrations into Solomon Islands waters were inversely related to distance from Solomon Islands. Table 18 shows the percentage of Papua New Guinea tagged skipjack that were recovered by the Solomon Islands pole-and-line fishery for several Papua New Guinea tagging areas that were varying distances from the Solomon Islands fishery. Recovery rates were 38

FIGURE 8. MIGRATION ARROWS FOR 41 OF 7 7 TAGGED SKIPJACK THAT MADE INTERNATIONAL MIGRATIONS TO THE WATERS OF SOLOMON ISLANDS. Recoveries have been selected to show no more than two examples of movement in each direction between any pair of five degree squares. Tick marks denote intervals of 30 days between release and recapture dates. In the absence of accepted zones for all countries, 200-mile boundaries were estimated by staff of the Tuna Programme. The boundaries shown should not prejudice any boundaries that may be derived in the future.

13SE 140E 14SE 150E 155E 160E 165E 170E 175E 180 175 W 39

TABLE 17. MONTHLY SKIPJACK TAG RECOVERIES BY ALL FISHING GEARS IN THE WATERS OF SOLOMON ISLANDS FROM TAG RELEASES IN OTHER COUNTRIES. Year/month of tagging and numbers of tags released are at the top of each column. Monthly Solomon Islands pole-and-line catch of skipjack and catch per day fishing (CPUE) are in columns 2 and 3.

YEAR/ SOLOMON ISLANDS PAPUA NEW GUINEA NEW CALEDONIA VANUATU TUVALU WESTERN NEW SOUTH QUEENSLAND PONAPE NORFOLK PALAU MONTH Pole-anc -line SAMOA WALES TOTAL Catch CPUE 77/10 79/05 79/06 77/12 78/01 78/01 78/06 78/06 79/04 79/05 79/11 80/03 80/08 TAG .tonnes) 900 3227 4423 6572 3622 1203 1766 1767 4322 2651 936 1113 6515 RECOVERIES (874)* (504)

77/11 1190 2.4 77/12 365 .9 X 78/01 15 .2 X X 78/02 - 78/03 15 .4 78/04 941 2.1 1 1 78/05 888 1.9 78/06 1526 3.2 1 XX 1 78/07 1973 3.6 1 1 2 78/08 1587 3.1 1 78/09 2304 4.5 (2) 4 78/10 2317 4.7 1 3 4 78/11 2915 5.7 1 78/12 2723 5.3 (1) 1 79/01 1106 5.0 79/02 - 79/03 - 79/04 1438 4.7 X 79/05 1788 3.7 X X 79/06 2955 5.6 X 79/07 3195 5.7 (1) 1 1 2 5 79/08 2150 3.8 1 1 79/09 2279 3.8 79/10 3330 5.4 2(2) 3 7 79/11 2944 5.0 1 1 3 4,3° X 12 79/12 2340 3.8 1 5,1° 7 80/01 1614 2.9 2 1° 3 80/02 64 1.7 1° 1° 2 80/03 - X 80/04 210 2.4 1 1 80/05 1029 2.8 1 1 2 80/06 1061 2.0 1 1 80/07 2550 4.6 1(2) 1 4 80/08 2778 5.0 (1) 1 X 2 80/09 2770 4.8 1 1 80/10 3244 5.7 1 1 1 3 80/11 3313 5.9 1 1 2 80/12 2774 4.7 • 2 1 2 1 6 81/01** 1531 3.3 1 1 81/02 - 81/03 - 81/04 1210 4.7 81/05 1881 3.4 1 81/06 2934 4.8 1 81/07 2796 4.5 81/08 3474 5.4 81/09 2631 4.1 81/10 2087 3.3 81/11 2131 3.4 81/12 1231 3.0 Unknown - 1 1 2 month

TOTALS (3) 10(6) 7 2 8 4 2 1 8 24 1 1 2 79

X Denotes tagging month. * Solomo n Sea releases anc recoveries in brackets. ** 1981 catch statistics are preliminary estimates from Solomon Islands Ministry of Natural Resources. ° Recovered by gear other than Solomon Islands pole-and-line. 40

highest for releases closest to Solomon Islands, i.e. Solomon Sea, and were lowest for releases in the Manus, New Hanover and Sepik areas. A similar inverse relationship between distance from Solomon Islands and percentage of tag recoveries in Solomon Islands is apparent in Figure 9 for countries from which tags were recovered in the waters of Solomon Islands. The percentage of recoveries in Solomon Islands of fish released elsewhere was very low (~0.1%) and roughly constant for tag releases at distances greater than 800 nautical miles from the Solomon Islands fishing area.

TABLE 18. PERCENTAGE OF TAG RECOVERIES IN THE WATERS OF SOLOMON ISLANDS FROM RELEASES IN PAPUA NEW GUINEA AT LOCATIONS INCREASINGLY DISTANT FROM THE SOLOMON ISLANDS FISHERY *

Release Area Number of Tag Percentage Releases Recovered

Solomon Sea 1378 1.96 St. George Channel 1492 .46 South Bismarck Sea 3807 .18 Manus 978 .10 New Hanover 759 0 Sepik 135 0

FIGURE 9. RELATIONSHIP BETWEEN TAG RECOVERY RATE BY SOLOMON ISLANDS POLE-AND-LINE VESSELS AND AVERAGE DISTANCE TRAVELLED BY TAGGED SKIPJACK ORIGINATING FROM COUNTRIES (AREAS) EXTERNAL TO THE SOLOMON ISLANDS FISHING AREA. Country abbreviations are explained in Appendix C except for SSI and SS2 - Papua New Guinea releases in the Solomon Sea in 1977 and 1979 respectively; and SCI and SC2 - Solomon Island releases in the Santa Cruz Islands in 1977 and 1980 respectively. Line fitted to the data points by inspection.

1.40n

(/)Q 1.20 SS2

1.00

UJ .80 QLD

QO .60-

°o .20- o0- NSW PON. TUV CAL ' NOR WES. 0.00 PAL 200 400 600 800 1000 1200 1400 1600 1800 AVERAGE STRAIGHT LINE MIGRATION DISTANCE (nautical miles) 41

Several additional observations deserve brief comment. First, there were only three recoveries in Solomon Islands from over 15,000 skipjack tag releases in the region of, and just above, the north equatorial counter current (Palau to Marshall Islands). In contrast, there were 46 Solomon Islands returns from approximately the same number of releases (19,500) in waters to the south of Solomon Islands (Australia, Norfolk Island, New Caledonia, Vanuatu). This difference in part reflects geographical distribution of tag releases in relation to seasonal movement of skipjack to and from higher latitudes. A more striking feature was the almost complete absence of tag recoveries (three returns) from over 61,000 releases in various countries and territories a similar distance to the east of Solomon Islands (170°E to 170°W, including Fiji, Wallis and Futuna, Kiribati, New Zealand). The second point is more general, and bears directly on the discussion of fishery interactions in the next section.

Tagging in areas surrounding local fisheries was expected to clarify sources of recruitment to the local fishery. Of course, during the time it takes fish to travel between areas one would expect them to continue to grow. Table 19 presents results from a comparison of tagging lengths and recovery lengths, for international migrants to Solomon Islands, with average size of skipjack in monthly samples from the Solomon Islands pole-and-line catch. At the time fish were tagged in the waters of other countries they were generally between 2 and 5 cm less than the average length of skipjack in the Solomon Islands catch for the same time period. However, on recovery, their reported lengths averaged between 0.7 and 5 cm above the average size of skipjack in the catch during the recovery month. The conclusion from these results is that even the comparatively small skipjack represented by 40 to 50 cm skipjack tagged at varying distances from Solomon Islands (approximately 250 to 1,500 nautical miles), were not contributing significantly to the small skipjack that were first recruited to the Solomon Islands pole-and-line fishery.

TABLE 19. A COMPARISON OF AVERAGE LENGTH OF SKIPJACK IN THE SOLOMON ISLANDS POLE-AND-LINE CATCH WITH LENGTH OF TAGGED SKIPJACK (RECOVERED IN SOLOMON ISLANDS) AT TIME OF TAGGING AND AT TIME OF RECOVERY. Differences in length were calculated by subtracting the average length of skipjack in a sample from the Solomon Islands catch, for the month of tag release or the month of tag recapture, from the length of individual tagged skipjack at release or at recapture. Average differences and sample sizes (N) are presented.

Average Difference in Length Average Le agth (cm) (cm)

Country of Time of Time of All Tag Recovered Recovered Release Tagging (N) Recovery (N) Releases Tags at Tags at Tagging Recovery

Papua New Guinea .2 (24) 5.0 (21) 54 53 59 New Caledonia -5.5 (10) 1.9 (9) 52 48 55 Vanuatu -4.3 (4) 2.8 (4) 50 50 57 Tuvalu -2.9 (2) 1.9 (2) 51 50 56 Western Samoa -4.9 (1) 2.8 (1) 48 48 58 New South Wales -4.4 (1) 1.8 (1) 49 47 56 Queensland _< 52 cm -4.6 (14) 0.8 (14) 48 48 55 > 52 cm 8.2 (8) 13.6 (8) 61 61 68 Ponape -22.5 (1) -1.5 (1) 34 32 52 Norfolk -9.0 (1) -0.1 (1) 57 45 54 Palau -15.2 (2) NA* (2) 41 38 NA*

* Average size of skipjack sample d from Solomon Islands c atches not available. 42

6.4.4 Mortality, production and fishery interactions

In August 1981 the Programme presented estimates of skipjack population size, mortality and migration for the region as a whole (Skipjack Programme 1981b). As tag returns and accompanying catch statistics were incomplete for several countries at this time, individual country estimates of mortality and population size were based on a simple graphical analysis of monthly tag returns, corrected for catch, for the available data sets. Since then, further tag and catch data have become available for some countries, notably Solomon Islands, and the data and method used to calculate country estimates have been refined.

For Solomon Islands, the following refinements were made to the tagging data base used in these analyses. First, release data were adjusted to include only those releases (and subsequent returns) within a one month period that were in the area of the commercial fishery. These fish should more closely represent the dynamics of the stock under exploitation. Thus for November 1977, releases off the south coast of Guadalcanal Island (96) and in the Santa Cruz Islands (417) were excluded (Table 2); 186 Santa Cruz releases were excluded from the June 1980 analysis (Table 2). As well, in June 1980, the size frequency distribution of the tagged population was considerably different from the size frequency distribution of the June catch (Figure 6). By disregarding 1,533 tag releases (and subsequent returns) of small skipjack from one school, which was encountered just north of Florida Islands on 18 June, the resulting size frequency distribution for tagged skipjack was much closer to the size distribution of the June commercial catch (Figure 10). In both years the tag releases were widely distributed over the fished area. Analyses were then carried out on the adjusted numbers of tag releases, the monthly recoveries by the Solomon Islands pole-and-line fleet from these releases, and monthly catch and effort by the fleet (data in Table 20).

FIGURE 10. LENGTH FREQUENCY DISTRIBUTIONS FOR SKIPJACK SAMPLED FROM THE JUNE 1980 COMMERCIAL CATCH AND FOR TAGGED SKIPJACK (ONE SCHOOL EXCLUDED) FROM THE SPC VESSEL CATCH, JUNE 1980. Mean lengths, standard deviations of the means (SD) and sample sizes (N) are indicated.

15.00- JUNE 1980

Commercial SPC

Mean 53.3 52.0 SD 4.96 7.33 u 10.00- o N 2001 1986 z oL±J a: UJ

5.00- SPC— /

O.OO- ! 1 1 1 — j —'f-—- 30 35 40 45 50 55 60 65 70 FORK LENGTH IN CENTIMETRES 43

TABLE 20. MONTHLY TAG RETURNS, CATCH AND EFFORT FOR THE SOLOMON ISLANDS POLE-AND-LINE FISHERY. The Table does not include 13 recoveries by the Skipjack Programme survey vessel. Catch statistics were supplied by the Solomon Islands Ministry of Natural Resources. The skipjack catch prior to 1981 is the product of total tuna catch by pole-and-line gear and the fraction of the catch estimated to have been skipjack; skipjack catch was provided for 1981.

Year/Month Tag Returns Skipjack Fishing Effort Catch (Days Fishing) (tonnes)

November 1977 - Tags Released 1709

77/11 17 1190(754)* 500(317)* 77/12 10 365 397 78/01 0 15 73 78/02 0 0 0 78/03 0 15 34 78/04 3 941 460 78/05 5 888 465 78/06 1 1526 515 78/07 5 1973 551 78/08 3 1587 512 78/09 4 2304 517 78/10 7 2317 497 78/11 5 2915 514 78/12 5 27 23 518 79/01 1 1106 223 79/02 0 0 0 79/03 0 0 0 79/04 2 1438 305 79/05 0 1788 485 Total 68 23091 6566 Unknown Date/ 9 Location** Grand Total 77

June 1980 - Tag s Released 2012

80/06 9 1061(354) 536(179) 80/07 28 2550 558 80/08 25 2778 554 80/09 9 27 70 574 80/10 8 3244 566 80/11 20 3313 566 80/12 16 2774 594 81/01 18 1531 463 81/02 0 0 0 81/03 0 0 0 81/04 5 1210 258 81/05 5 1881 560 81/06 11 2934 614 81/07 4 27 96 628 81/08 6 3474 640 81/09 2 2631 639 81/10 0 2087 630 81/11 0 2131 632 81/12 1 1231 415 Total 167 40396 9427 Unknown Date/ 51 Location** Grand Total 218

* Brackets enc .ose catch and e ffort prorated for portion of the month during which tags were at large. Prorated catch and effort was used in the fitting procedure.

** Date information on tag returns was considered imprecise if the range of possible values was more than half the span from the release late to the midp oint of the range of possible recovery dat as. If the range was less than half 3f this span, the return was inc uded and the date of recovery was taken to be the mi Ipoint of the range. 44

Mortality and Production

A non-linear, least squares fitting technique (Conway, Glass and Wilcox 1970) was used to fit a tag recapture and attrition model to the refined sets of Solomon Islands data (Kleiber MS). From the number of tag releases, N0, the model predicts the number of tag returns per time period, *l , as a function of fishing mortality per time period, Fi , and attrition of tags at large, A , due to natural mortality, emigration, declining vulnerability, and continuous tag mortality and tag shedding; p takes into account other forms of tag loss as described below. The mathematical formula for the model is:

iA A ri = PNoFie- (e -l)

Fi can be approximated by

Fi » ^ « QEi

where Ci is the catch in time period i , P is the standing stock, Q is the catchability coefficient, and Ei is the fishing effort in time period i .

Figure 11 shows monthly tag returns and predicted monthly tag returns from the model. Using monthly catches and tag returns in Table 20, the fitted model provided estimates, and 95 per cent confidence intervals for average population size (tonnes), average monthly tag attrition rate, average monthly throughput (tonnes) and, when effort was substituted for catch, the average catchability coefficient. This latter parameter is simply the fraction, usually very small (<.0001), of the vulnerable skipjack population that is harvested by one pole-and-line vessel fishing for one day.

It is assumed that the population of skipjack vulnerable to pole-and-line gear is in a steady state. In other words, recruits to the fishable stock, through immigration and growth of small fish, on average equals losses from the fishable stock due to fishing, natural death, decrease in vulnerability with increasing size and emigration. Thus, one could view the attrition rate, less a small fraction (<0.01 per month due to continuous tag shedding with time [Skipjack Programme 1981c]), as the turnover rate, or the fraction of the population that is renewed each month. The product of turnover rate and population size is an estimate of throughput, the tonnes of skipjack entering and leaving the fishable population each month. Throughput is equivalent to recruitment to the fishable stock.

Table 21 presents parameter estimates and confidence intervals from the analysis of tag returns from the two sets of Solomon Islands releases. The fitting procedure gives more weight to months with greater tag returns; hence population size and catchability coefficient estimates (Table 21) are most representative of the fished stock in the months immediately following tagging. In the table, p is a coefficient, ideally close to one, that corrects for the combined effects of short-term tagging mortality, short-term tag shedding, and non-reporting of recaptured tags. 45

FIGURE 11. SOLOMON ISLANDS MONTHLY SKIPJACK TAG RETURNS FROM TABLE 20 (STARS), AND ESTIMATED MONTHLY TAG RETURNS FROM THE FITTED TAG ATTRITION MODEL (CIRCLES). The ordinate is in log scale.

NOVEMBER 1977

10.0

O O A O O O O O

1.0 o on. —i—i—i—i—i—i—i—i—i—i—i—i—i—i i—i—i—r-*"—i ND'JFMAMJJASOND'JFMAM 1977 1978 1979

JUNE 1980 • $ $ O o o * *

10.0 o • * :* # o o i ii o O * o - O O - if °o

1.0- it - O

1 1 • 1 ' 1 1 1 • i A S o N D J F M A M J J A S J J O N D ' 1980 1981 46

TABLE 21. ESTIMATES OF SKIPJACK POPULATION SIZE, TURNOVER RATE, THROUGHPUT AND CATCHABILITY COEFFICIENTS FOR THE SOLOMON ISLANDS POLE-AND-LINE FISHERY.

Year/Month of Tagging Population Monthly Throughput Catchability Size Turnover per Month Coefficient (tonnes) (tonnes)

77/11 77,000p* .24 18,000p 1.7 x 10"5/p 95% confidence limit 38,000p .13 12,000p .5 x 10-5/p 95% confidence limit 196,000p .34 32,000p 4.2 x 10~5/p

80/06 164,000p .16 26,000p 3.0 x 10-5/p 95% confidence limit 85,000p .07 17,000p 1.5 x 10-5/p 95% confidence limit 354,000p .28 43,000p 5.3 x 10-5/p

* p is a coefficient (.<1.0) that corrects for short-term tag £sheddin g and tagging mortality, non-report ing of recaptured tags, and tags for which there is inadequate information.

Skipjack are clearly abundant in the fished area of Solomon Islands. Estimates of population size, turnover and throughput are all high. Estimated monthly throughput shows less statistical variability than either population size or turnover rate, and is considered a more meaningful resource measure. The two estimates in Table 21, 18,000p tonnes per month from November tag releases and 26,000p tonnes per month from June tag releases, are, respectively, 14p and 12p times the average monthly catch for the period of tag returns (the throughput:catch ratios would be 8 and 7 if p was 0.58 [see next page]). Thus there would appear to be considerable room for increased catches from the present fished area, perhaps four fold, before the fishery might be expected to reduce long-term yield, assuming that most recruitment originates locally. However, one must be careful in the use of these estimates as there are wide confidence limits; for example, population size (38,000p to 196,000p tonnes for 1977 and 85,000p to 354,000p tonnes for 1980 respectively), which is not unexpected considering the statistical variability encountered when tagged fish represent such a small fraction of the estimated population at large, and when tag recoveries are also a small fraction of tag releases. The estimates of population size or average standing stock, 77,000p tonnes from 1977 tagging, and 164,000p tonnes from 1980 tagging, are more than two per cent of the estimated standing stock for the whole Skipjack Programme study area, and this from a fished area in Solomon Islands that is less than one half of one per cent of the total study area. This suggests that skipjack are not uniformly dense within the SPC region. However, this interpretation could be biased because the boundaries defining the geographical limits of the stocks may be greater than the fished zones.

It is noteworthy that the lower population estimate based on tag returns from releases in November 1977 is consistent with the poor fishing conditions experienced by the commercial fleet at this time, and for several months thereafter.

Estimates of monthly" turnover rates of 0.24 and 0.16 (Table 21) have overlapped confidence intervals. Even so, it is interesting that the lower rate for June 1980 tag releases is consistent with the release of smaller tagged skipjack at this time (see Figures 6 and 10). It was previously 47

hypothesised (Section 6.4.1) that a combination of population processes, lower mortality and greater vulnerability for smaller skipjack could in concert lower the calculated turnover rate. Both estimates of Solomon Islands turnover (attrition) rates are similar to the value of 0.19 estimated from all Skipjack Programme tag returns (Kleiber MS), and are similar to the estimate of 0.23 monthly attrition from Joseph and Calkins (1969) for eastern Pacific skipjack in the northern fishr'ug area.

Catchability coefficients, multiplied by average monthly effort, provide estimates of the average monthly instantaneous fishing mortality rate. For the Solomon Islands pole-and-line fishery this rate was 0.006/p based on November 1977!tag releases and 0.015/p based on June 1980 tag releases. These fishing mortality rates represent only a small fraction (less than 10 per cent) of the monthly attrition rates, which, of course, follows from the previously discussed relationship between catch and throughput. Of interest, higher catchability for skipjack calculated from June 1980 releases is consistent with the hypothesised greater vulnerability of skipjack to pole-and-line gear at that time (see Section 6.2.1).

Skipjack attrition rates are high. Based on the value of 0.19 per month for the study area as a whole, approximately 90 per cent of catchable skipjack have left the fished population, in various ways, after one year. A preliminary analysis of change in biomass of a newly recruited cohort, which combined mortality and growth parameters to determine the skipjack size at which maximum biomass occurs, showed that at present levels of exploitation maximum yield per recruit occurred before the size at first recruitment to pole-and-line gear (Skipjack Programme, unpublished information). Hence it is not necessary, from a biological point of view, to adopt fishing strategies in any portion of the SPC region that would delay the harvest to skipjack of larger size.

Some comment on the correction factor p is warranted. In Table 20 it can be seen that approximately 20 per cent of the tag returns were unuseable in the analysis, leading to a factor of 0.8 in p. Results from a double tagging experiment carried out by the Programme during the second visit to Fiji gave an estimate of 0.997 for the combination of immediate tag retention and return rate of recaptured tags (Skipjack Programme, unpublished information). However, since the lower 95 per cent confidence limit of this estimate is 0.8, and since there is reason to suspect that the return rate was particularly favourable during the double tag experiment, a conservative value of 0.8 has been used for this factor in Solomon Islands. The immediate tagging mortality has not been measured. However, high tag return rates observed in some eastern Pacific tagging experiments strongly suggest that immediate tagging mortality and shedding are low. Return rates greater than 50 per cent have been observed (W.H. Bayliff, personal communication). This, coupled with the high natural mortality, implies that substantially more than 50 per cent of the tagged fish must have survived the tagging procedure. In the absence of more quantitative information on immediate tagging mortality, a best guess of 10 per cent has been provisionally assumed, implying a further factor of 0.9 in p. Combining all factors (0.8x0.8x0.9) leads to a value of 0.58 for p. Further work to directly measure tagging mortality, tag slippage and non-reporting of tags is planned.

Fishery Interactions

Many types of interactions are possible among fisheries; for example, changes in catch in one fishery from increased catches in other fisheries, within a generation or between generations; changes in recruitment (or 48 standing stock) that arise from immigration from neighbouring fished areas; and changes in yield per recruit resulting from different fishing strategies. The absence of any relationship between catch per unit effort and effort for intense skipjack fisheries (e.g., Japanese coastal fishery, eastern tropical Pacific fishery), suggests that between-generation fishery interactions would be negligible for present or expanded fisheries in the western Pacific. Therefore, the Programme concentrated on evaluation of fishery interactions within a single skipjack generation.

The approach followed in Skipjack Programme (1981b) was to use tagging data plus catch statistics to estimate coefficients of migration from particular fisheries. The product of population size in the donor fishery and migration coefficient gave an estimate of the tonnes of skipjack migrating between fishing areas. Comparison of these estimates with estimates of population size in the recipient country, or in the donor country, illustrated the fishery or stock interactions within one skipjack generation, since it measured the fraction of the standing stock that migrated to or from a particular area. Results demonstrated a generally low level of stock interaction for existing locally based fisheries.

A simpler expression of interaction is the percentage of throughput (recruitment) in the destination country that is due to immigration from the donor country (Skipjack Programme, manuscript in preparation). It is independent of p, assuming that p is the same in the donor and destination countries.

Table 22 presents results for Solomon Islands - Papua New Guinea fishery interactions. From column 6 it can be seen that skipjack moving from Papua New Guinea to Solomon Islands, measured by May and June 1979 tagging in Papua New Guinea (excluding Solomon Sea releases), were estimated to contribute only 1.9 to 2.2 per cent of Solomon Islands throughput. From column 6 it can also be seen that migrants from the Solomon Islands fishing area, represented by November 1977 tag releases, were estimated to have contributed only 1.3 per cent of the Papua New Guinea throughput. A somewhat higher percentage (5.2%) of recruits to Papua New Guinea's fishery arose from migrant skipjack represented by June 1980 tagging in Solomon Islands. Although the contribution of Solomon Islands skipjack to Papua New Guinea throughput is not large, the difference between 1977 and 1980 is of interest. One inference is that migration between the two fishing areas is variable, perhaps with a seasonal component. Another more important inference is that smaller skipjack from Solomon Islands are relatively more important to Papua New Guinea recruitment than are larger fish from Solomon Islands.

In combination, the above results suggest that even greatly expanded fisheries within each country's present fished areas would reduce monthly recruitment to each fishery by at most only a few per cent. Such a low level effect would be next to impossible to detect in the catch.

Several questions remain unanswered. Where do the majority of recruits to the Solomon Islands fishery come from? Are they predominantly from local production (skipjack do spawn in this area and skipjack juveniles were reasonably abundant), and/or is a large component due to immigration from surrounding areas? Recoveries in Solomon Islands of tags released in other countries (areas) shed's some light on the question of immigration from outside. Table 23 presents estimated contribution to throughput in Solomon Islands. In estimating the percentage of Solomon Islands recruits that originated from these countries (areas), it is assumed that throughput in each area is the same as the average for the two tagging periods in Solomon 49

Islands, that is, 22,000p tonnes per month ( [ 18000p + 26000p]/2 , from Table 22). The computation of percentage contribution to Solomon Islands throughput is illustrated in the footnote to Table 23. Since the assumed throughput of 22,000p tonnes is one of the highest observed for regional fisheries, the contribution estimates should be considered to be maximum estimates. The percentages in Table 23 < are not necessarily additive since skipjack may have passed through one or more of the locations to reach Solomon Islands. For instance, skipjack from New South Wales may first traverse the waters of Queensland and then the Solomon Sea before reaching the Solomon Islands; Tuvalu and Western Samoa skipjack may have passed through the Santa Cruz Islands.

TABLE 22. FISHERY INTERACTION RESULTS BETWEEN SOLOMON ISLANDS AND PAPUA NEW GUINEA. Column 4 gives the number of skipjack tag returns in the destination country's pole-and-line fishery. Papua New Guinea data from Ellway and Kearney (MS).

Year/Month No. of Tags Estimated No. of Tags Average Estimated of Released Monthly Re covered Monthly Percentage Tagging Throughput Catch of Throughput (Donor Country) (tonnes) from Migrants (1) (2) (3) (4) (5)** (6)*

PAPUA NEW GUINEA TO SOL 0 M 0 N I S L A N D S

79/05 2717 12000p 9 1772 2

79/06 3291 21000p 6 2062 2

S 0 L 0 M ON ISLANDS TO PAP U A NEW GUINEA

77/11 1709 18000p 4 3340 1

80/06 2012 26000p 9 2240 5

* Column, 6 = [(column 3 x column 4) / (co lumn 2 x colum n 5)]100

** Catches averaged over a 12-month period bracketing the period of tag returns, or averaged from one month before the first return to one month after the last return when tag returns spanned a perio d of more than one year. Solomon Islands catche s from Table 17; Papua New Guinea catches averaged from monthly catch st atistics su pplied by the Papua New Guinea Department of Primary Inc ustr y, Port Moresby (Papua New Guinea catch estimated for 80/06 taggi ng).

From column 4 in Table 23 it can be seen that contributions from outside are generally small, but not insignificant in the case of Queensland, Solomon Sea and Santa Cruz Islands, the closest locations to the Solomon Islands fishery in which tags were released. And secondly, the percentage contribution to recruitment generally falls as the distance from Solomon Islands increases. These results are consistent with the previous simpler treatment of migrant tag returns versus distance from Solomon Islands (see Figure 9), and with conclusions arising from population structure analyses (Section 6.3k4). They suggest that migrants of the size that were tagged and released within an 800-nautical mile radius to the south and southwest of 50

Solomon Islands, could be contributing as much as 10 to 15 per cent of throughput to the present Solomon Islands fishery. These results also highlight the fact that fishery interactions, within a skipjack generation, decrease rapidly as the distance between fisheries increases. It should be remembered, however, that seasonal effects could result in a somewhat different pattern. TABLE 23. ESTIMATED CONTRIBUTION TO SOLOMON ISLANDS THROUGHPUT BY IMMIGRANTS FROM TAGGING LOCATIONS OUTSIDE THE SOLOMON ISLANDS FISHING AREA

Tagging Location Solomon Islands Pole-and-Line Fishery and Date No. of Tags Tag Average Estimated Releas ed Recoveries Catch Percentage of (tonnes) Throughput from Migrants (1) (2) (3)* (4)**

Papua New Guinea 77/10 874 3 1565 4.8 (Solomon Sea) Papua New Guinea 79/05 504 6 1954 13.4 (Solomon Sea) New Caledonia 77/12 6572 1 1434 0.2 New Caledonia 78/01 3622 8 1525 3.2 Vanuatu 78/01 1203 4 2112 3.5 Tuvalu 78/06 1766 2 2008 1.2 Western Samoa 78/06 1767 1748 0.7 New South Wales 79/04 4322 2101 1.7 Queensland 79/05 2651 19 2107 7.5 Ponape 79/11 936 1772 1.3 Norfolk 80/03 1113 1926 1.0 Palau 80/08 6515 1826 0.4 Santa Cruz Islands 77/12 417 1640 3.2 Santa Cruz Islands 80/06 186 2082 5.7

* Catches averaged over a 12-month period bracketing the p eriod of tag returns, or averaged from one month before the first r eturn to one month after the last return when tag returns spanned a period of more than one year. Catches from Tabl e 17.

** Column 4 =[(22000 x column 2) / (colum n 1 x column 3)]100

Perhaps of more importance, the Programme tagged only a few skipjack less than 45 cm, and obviously such small fish must make a large contribution to recruitment into the Solomon Islands fished stock. Although it is unlikely that small skipjack from the region of the north equatorial counter current are a major source (see Section 6.4.3), this does not rule out recruitment of skipjack smaller than were tagged in waters immediately to the north and west of Solomon Islands, nor, of course, does it rule out a large contribution from production within the Solomon Islands fishery zone.

6.5 Yellowfin Biology and Tagging Results

Sixteen per cent of the Programme's catch in the waters of Solomon Islands were yellowfin; schools with yellowfin accounted for 21 per cent of all schools sighted. 51

Yellowfin averaged 56.1 cm fork length, about 6 cm longer than skipjack (Figure 3). There was a tendency to tag a narrower size range of yellowfin than actually were caught, since, as noted by Gillett and Kearney (1982), small yellowfin are difficult to tag while in the tagging cradles and large yellowfin were often too big to be safely tagged, even under relatively calm conditions. Consequently, a greater proportion of small and large yellowfin were included in the biological sample.

Virtually all of the sampled yellowfin were immature; and sex could seldom be determined for yellowfin less than 60 cm. Yellowfin diet (Appendix D) was similar to that of skipjack, although yellowfin had a greater variety of items per stomach; 0.51 items per yellowfin stomach versus 0.23 items per skipjack stomach. The incidence of skipjack tuna juveniles in yellowfin stomachs was less than one per cent, considerably below the 4.7 per cent level for skipjack juveniles in skipjack stomachs (Table 6). Similar species differences in the incidence of tuna juveniles occurred throughout the Programme's study area.

In both 1977 and 1980, most tagged yellowfin were released near Shortland Islands and in the Santa Cruz Islands; lesser numbers were released in The Slot and in Indispensible Strait (Table 24). There were three distinct size modes in 1977 (38 cm, 51 cm, 64 cm); and only one clear mode in 1980 (54 cm) (Figure 12). From the total of 858 releases, only 15 had been recovered as of 1 March 1982. Thirteen of these recoveries were by the local pole-and-line fleet. The yellowfin recovery rate of 1.8 per cent was well below that for skipjack from Solomon Islands releases (9.5 per cent for both surveys). This difference in recovery rate reflects, in part, the disproportionate number of yellowfin releases outside of the fished area and differences in growth rate between the species. Wankowski (1981) estimated that yellowfin of the size exploited by pole-and-line gear grew approximately twice as fast as skipjack; hence, they would be exposed to pole-and-line gear for considerably less time.

TABLE 24. NUMBERS OF YELLOWFIN TAGGED, TOTAL RECOVERED AND PERCENTAGE OF TAGS RECOVERED FOR SUB-AREAS IN THE WATERS OF SOLOMON ISLANDS. Tag recoveries to 1 March 1982.

Shortland The Slot South Coast Indispensable Santa Total Islands and Guadacanal Strait Cruz vicinity Island Islands

1 9 7 7

Numbers Tagged 81 37 - - - 118 Numbers Recovered 1* 1 Percentage Recovered 1.24 .85

19 8 0

Numbers Tagged 343 84 - 118 195 740 Numbers Recovered 11* 2 1 14 Percentage Recovered 3.21 2.38 - .85 - 1.89

* Two recoveries from processing plants (one from 1977 tagging, one from 1980 tagging) with no information on position and date of recapture. 52

FIGURE 12. LENGTH FREQUENCY DISTRIBUTIONS FOR TAGGED YELLOWFIN RELEASED DURING 1977 AND 1980 IN THE WATERS OF SOLOMON ISLANDS. Mean lengths, standard deviations of the means (SD) and sample sizes (N) are indicated.

15.00- 1977 1980

Mean 49.9 56.5 SD 10.96 5.70 N 114 529

10.00- UoJ <

o x hi a. 5.00- / I /Ll977 / I : "A A i

0.00- —- , : ,-* , ,—1—(•—, , 1 • •; . - • 30 35 40 45 50 55 60 65 70 FORK LENGTH IN CENTIMETRES International recoveries from yellowfin tagged in the waters of Vanuatu and Papua New Guinea (Skipjack Programme 1981b) suggest a relatively short and predominantly northwestward movement pattern, and this pattern may apply to yellowfin from the waters of Solomon Islands. One Solomon Islands yellowfin, tagged in Indispensible Strait in June 1980, was recovered in international waters just to the south of Yap, Federated States of Micronesia, six months later. Further recoveries by surface and longline fisheries and adjustment for recovery effort are necessary to clarify yellowfin migration patterns. These and other analyses of the yellowfin data are continuing under the South Pacific Commission's Tuna and Billfish Programme.

7.0 CONCLUSIONS

7.1 Baitfish

The Programme's bait catches in the vicinity of the Slot were excellent. The vessel seldom carried less than 100 kg of bait, and often carried in excess of 400 kg at the commencement of each fishing day. At most localities average catches exceeded 100 kg per haul, of which at least two-thirds were considered to be highly effective bait for pole-and-line fishing. In general, baitfish were found to be very abundant in the waters of Solomon Islands, more so than in any other country or territory in the South Pacific Commission region, with the possible exception of Papua New Guinea.

Bait resources adjacent to the main fishing area are characterised by an abundance of stolephorid anchovies, which are amongst the most desirable tropical bait species, and a mixture of other effective bait species. Baiting areas here are both numerous and extensive; many are not yet fished to any great degree by the local fleet. Although localised short-term declines in catch rates might be expected to occur at times in some of the 53 heavily fished baiting grounds, the presence of many alternative sites should give an expanded pole-and-line fleet sufficient options to compensate for local depletions. It would be advisable to monitor certain aspects of the baitfish fishery, particularly changes in species composition, size and CPUE at major baiting grounds, since there are examples in Papau New Guinea of apparent fishery-induced declines of Stolephorus spp. and Spratelloides spp. bait fisheries (Dalzell and Wankowski 1980), although these are thought to have occurred under considerably greater and more localised fishing pressure than presently takes place in Solomon Islands.

The bait resources of the three major Santa Cruz Islands appear quite limited as shown by the Programme's low average catch per set. As well, there are only a few suitable baiting localities, although there may be a few additional localities amongst the Duff and Reef Islands. Considering these limitations, a cautious approach is suggested for developing the commercial potential of this area by small pole-and-line vessels.

7.2 Skipjack

The present pole-and-line fishery in Solomon Islands is only harvesting a fraction of the available resource in the fished area. Even though skipjack may be densest in the fished area, this area is less than ten per cent of the total area of the Solomon Islands fishery zone. Clearly there is potential for greatly increased skipjack catches from the waters of Solomon Islands.

Sizeable locally based skipjack fisheries in countries near Solomon Islands are at present having little impact on abundance and recruitment in Solomon Islands of skipjack of the same generation. Low numbers of recoveries in Solomon Islands of tagged skipjack released near large distant-water pole-and-line and seine fisheries to the north and west of Solomon Islands suggest that these fisheries are also not yet greatly affecting the Solomon Islands fishery. The complete absence of tag recoveries of Solomon Islands origin from these distant-water fisheries is evidence that they, in turn, are not affected by the Solomon Islands fishery. Nor does the Solomon Islands fishery appear to be having any impact on abundance and recruitment to nearby locally based fisheries.

Between-generation effects are also felt to be minimal. At present levels of exploitation, recruitment into the population of Pacific skipjack would appear to be independent of catches. In the absence of a demonstrable relationship between catches and subsequent recruitment for skipjack anywhere, there would appear no need for anxiety over reduction of the spawning stock.

Mixing of skipjack stocks is believed to depend strongly on the distance separating them. On the basis of tagging analyses, interactions between skipjack in Solomon Islands and in areas currently supporting skipjack fisheries, though present, are minor. Nevertheless, the evidence from tagging and blood genetics analyses does not support the hypothesis that skipjack in Solomon Islands belong to any identifiable subpopulation that can be considered as genetically isolated from the rest of the central and western Pacific resource.

Eventually, as fisheries in Solomon Islands and neighbouring countries increase their levels of exploitation, and as areas of exploitation expand and begin to overlap, the importance of fishery interactions within a skipjack generation can be expected to increase. These interactions could be 54 sizeable if a large fishery for skipjack was to develop in waters immediately adjacent to Solomon Islands, in particular to the west and southwest.

Purse-seining has the potential to harvest large amounts of skipjack over a wide range of sizes. Recent progress in adapting seining techniques to conditions in the central and western Pacific suggests that seining in the Commission region will increase substantially in the near future. Already the area of the distant-water purse-seine fishery has expanded into the waters around the Solomon Islands 200-mile zone, and one Solomon Taiyo seiner is operating successfully in local waters. As well, Solomon Islands has plans for a sizeable increase in its local seine catch of skipjack. These operations may well produce significant revenue for Solomon Islands, but it must not be overlooked that seining at this level will undoubtedly result in increased fishery interactions, both between local pole-and-line and seine fisheries, and among other fisheries of the region. Clearly a regional approach to monitoring trends in these fisheries is essential for rational fisheries development. 55

REFERENCES

ANON (Manuscript, no date). Baitfishing. Solomon Islands Ministry of Natural Resources, Honiara. (Unpublished report).

ANON (1975). The pole-and-line fishery of the Solomon Islands. Regional Technical Meeting on Fisheries 1976. 8 : Country Statement, South Pacific Commission, Noumea, New Caledonia. 6pp.

ANON (1980). Solomon Islands 1980 Statistical Yearbook. Statistics Office Bulletin No.32/80. Ministry of Finance, Honiara, Solomon Islands. 203pp.

ANON (1980a). Review of preliminary results from genetic analysis of skipjack blood samples collected by the Skipjack Survey and Assessment Programme. Skipjack Survey and Assessment Programme Technical Report No.l, South Pacific Commission, Noumea, New Caledonia. 23pp.

ANON (1981). Solomon Islands External Trade January to June 1981. Statistical Bulletin No. 18/81. Statistics Office, Ministry of Finance, Honiara, Solomon Islands. 2pp with tables.

ANON (1981a). Report of the second Skipjack Survey and Assessment Programme workshop to review results from genetic analysis of skipjack blood samples. Skipjack Survey and Assessment Programme Technical Report No.6, South Pacific Commission, Noumea, New Caledonia. 45pp.

ARGUE, A.W. (1982). Methods used by the Skipjack Survey and Assessment Programme for collecting biological, tuna school and ancillary data from a pole-and-line fishing vessel. _In Kearney, R.E. (Ed). Methods used by the South Pacific Commission for the survey and assessment of skipjack and baitfish resources. Tuna and Billfish Assessment Programme Technical Report No.7, South Pacific Commission, Noumea, New Caledonia.

ARGUE, A.W., F.C. CONAND and D. WHYMAN (Manuscript). Observations on juvenile tunas from stomachs of skipjack and other surface tunas caught by pole-and-line gear in the Pacific Ocean. Tuna and Billfish Assessment Programme Technical Report. South Pacific Commission, Noumea, New Caledonia.

ARGUE, A.W. and J.-P. HALLIER (Manuscript). Observations on fishing performance of baitfish species used by tuna pole-and-line gear in the central and western Pacific Ocean. Tuna and Billfish Asses sment Programme Technical Report. South Pacific Commission, Noumea, New Caledonia.

BALDWIN, W.J. (1977). A review on the use of live baitfishes to capture skipjack tuna, Katsuwonus pelamis. in the tropical Pacific Ocean with emphasis on their behaviour, survival and availability. In Shomura, R.S. (Ed). Collection of tuna baitfish papers. United States Department of Commerce, National Oceanic and Atmospheric Administration Technical Report, National Marine Fisheries Service Circular 408:8-35.

BROOKFIELD, H.C. (1969). Some notes on the climate of the British Solomon Islands. Philosophical Transactions of the Royal Society of London, Series B. Biological Sciences 255(800): 207-210. 56

CONWAY, G.R., N.R. GLASS and J.C. WILCOX (1970). Fitting nonlinear models to biological data by Marquardts algorithm. Ecology. 51(3):503-507. DALZELL, P.J. and J.W.J. WANKOWSKI (1980). The biology, population dynamics and fisheries dynamics of exploited stocks of three baitfish species, Stolephorus heterolobus. S. devisi and Spratelloides gracilis in Ysabel Passage, New Ireland Province. Papua New Guinea, Department of Primary Industry Research Bulletin 22. 124pp.

DONGUY, J.R. and C. HENIN (1978). Hydroclimatic anomalies in the South Pacific. Oceanologica Acta 1(1):25-30.

DONGUY, J.R. and C. HENIN (1981). Two types of hydroclimatic conditions in the south-western Pacific. Oceanologica Acta 4(l):57-62.

ELLWAY, C.P. and R.E. KEARNEY (Manuscipt). An assessment of the skipjack and baitfish resources of Papua New Guinea. Skip jack Survey and Assessment Programme Final Country Report. South Pacific Commission, Noumea, New Caledonia.

FUJINO, K. (1966). Instructions for collecting blood and serum samples from tuna fishes. FAQ Circular 26, Food and Agriculture Organization, Rome, Italy. 5pp.

FUJINO, K. (1970). Skipjack tuna subpopulation identified by genetic characteristics in the western Pacific. Jin Marr, J.C. (Ed). The Kuroshio: symposium on the Japan current. East-West Center Press, Honolulu, 1968. pp 385-393.

FUJINO, K. (1976). Subpopulation identification of skipjack tuna specimens from the southwestern Pacific Ocean. Japanese Society of Scientific Fisheries Bulletin 42(11):1229-1235. GILLETT, R.D. and R.E. KEARNEY (1982). Methods used by the Skipjack Survey and Assessment Programme for tagging skipjack and other tuna. _I_n Kearney, R.E. (Ed). Methods used by the South Pacific Commission for the survey and assessment of skipjack baitfish resources. Tuna and Billfish Assessment Programme Technical Report No.7, South Pacific Commission, Noumea, New Caledonia.

HALLIER, J.-P. and R.D. GILLETT (1982). Baitfishing methods used by the Skipjack Survey and Assessment Programme and their comparison with those used by commercial Japanese pole-and-line vessels. Jja Kearney, R.E. (Ed), Methods used by the South Pacific Commission for the survey and assessment of skipjack and baitfish resources. Tuna and BilIfish Assessment Programme Technical Report No.7, South Pacific Commission, Noumea, New Caledonia. HIDA, T.S. and J.H. UCHIYAMA (1977). Biology of baitfishes Herklotsichthvs punctatus and Pranesus pinguis in Majuro, Marshall Islands. In Shomura, R.S. (Ed). Collection of tuna baitfish papers. United States Department of Commerce, National Oceanic and Atmospheric Administration Technical Report, National Marine Fisheries Service Circular 408:63-68. 57

JOHANNES, R.E. (1981). Words of the lagoon; fishing and marine lore in the Palau district of Micronesia. University of California Press, Berkeley. 245pp.

JOSEPH, J. and T.P. CALKINS (1969). Population dynamics of the skipjack tuna (Katsuwonus pelamis) of the eastern Pacific Ocean. Inter-American Tropical Tuna Commission Bulletin 13(l):l-273.

JOSSE, E., J.C. LE GUEN, R. KEARNEY, A. LEWIS, B. SMITH, L. MAREC and P.K. TOMLINSON (1979). Growth of skipjack. South Pacific Commission Occasional Paper No.11. South Pacific Commission, Noumea, New Caledonia.

KEARNEY, R.E. (1978). Interim report of the activities of the Skipjack Survey and Assessment Programme in the waters of Fiji (26 January - 18 February, 28 March - 10 April 1978) . Skipjack Survey and Assessment Programme Preliminary Country Report No.5, South Pacific Commission, Noumea, New Caledonia. 13pp.

KEARNEY, R.E. (1979). An overview of recent changes in the fisheries for highly migratory species in the western Pacific Ocean and projections for future developments. South Pacific Bureau for E conomi c Co-operation, SPEC(79) 17. 99pp.

KEARNEY, R.E. (Ed) (1982). Methods used by the South Pacific Commission for the survey and assessment of skipjack and baitfish resources. Tuna and Billfish Assessment Programme Technical Report No.7, South Pacific Commission, Noumea, New Caledonia.

KEARNEY, R.E. and A.D. LEWIS (1978). Interim report of the activities of the Skipjack Survey and Assessment Programme in the waters of the Solomon Islands (1 November - 4 December 1977). Skipjack Survey and Assessment Programme Preliminary Country Report No.2, South Pacific Commission, Noumea, New Caledonia. 14pp.

KLAWE, W.L. (1978). Estimates of catches of tunas and billfishes by the Japanese, Korean, and Taiwanese longliners from within the 200 mile economic zone of the member countr ies of the South Pacific Commission. South Pacific Commission Occas ional Paper No.10, South Pacific Commission, Noumea, New Caledonia. 41pp.

KLEIBER, P.M. (Manuscript). Estimation of production, mortality and migration of skipjack (Katsuwonus pelamis) in the central and western Pacific Ocean. Tuna and Billfish Assessment Programme Technical Report, South Pacific Commission, Noumea, New Caledonia.

KLEIBER, P.M. and R.E. KEARNEY (Manuscript). An assessment of the skipjack and baitfish resources of Kiribati. Skipjack Survey and Assessment Programme Final Country Report. South Pacific Commission, Noumea, New Caledonia.

KLEIBER, P.M. and C.A. MAYNARD (1982). Data processing procedures of the Skipjack Survey and Assessment Programme. Ta Kearney, R.E. (Ed). Methods used by the South Pacific Commission for the survey and assessment of skipjack and baitfish resources. Tuna and Billfish Assessment Programme Technical Report No.7. South Pacific Commission, Noumea, New Caledonia. (In press). 58

LAWSON, T.A. and R.E. KEARNEY (Manuscript). Estimates of growth rates and length measurement errors for skipjack (Katsuwonus pelamis) from the western Pacific Ocean based on tagging data. Tuna and Billfish Assessment Programme Technical Report. South Pacific Commission, Noumea, New Caledonia. LESTER, R.J.G. (1981). Preliminary results of observations on the parasites of Pacific skipjack tuna. Regional Technical Meeting on Fisheries 1981. 13 : Background Paper, South Pacific Commission, Noumea, New Caledonia. 10pp.

LEWIS, A.D. (1981). Population genetics, ecology and systematics of Indo-Australian Scombrid fishes, with particular reference to skipjack tuna (Katsuwonus pelamis). Unpublished Ph.D thesis, Australian National University, Canberra. 314pp.

NAGANUMA, A. (1979). On spawning activities of skipjack tuna in the western Pacific Ocean. Tohoku Regional Fisheries Research Laboratory Bulletin 40:1-13.

RAJU, G. (1964). Studies on the spawning of the oceanic skipjack Katsuwonus pelamis (Linnaeus) in Minicoy waters. Proceedings Symposium on Scombroid Marine Biological Association of India. Part 11:744-768.

RICHARDSON, B.J. (Manuscript). The distribution of protein variation in skipjack tuna (Katsuwonus pelamis) from the central and south-western Pacific. Department of Population Biology, Research School of Biological Sciences, Australian National University, Canberra. (Unpublished manuscript).

SEVELE, F.V. and A. BOLLARD (1979). South Pacific economies: statistical summary. South Pacific Commission Occasional Paper No.15. South Pacific Commission, Noumea, New Caledonia. 19pp. SHARP, G.D. (1969). Electrophoretic study of tuna hemoglobins. Comparative Biochemistry and Physiology 31:249.

SHARP, G.D. (1978). Behavioural and physiological properties of tunas and their effects on vulnerability to fishing gear. JEn. Sharp, G.D. and A.E. Dizon (Eds). The physiological ecology of tunas. Academic Press, New York, pp 397-450.

SHOMURA, R.S. (Ed) (1977). Collection of tuna baitfish papers. United States Department of Commerce, National Oceanic and Atmospheric Administration Technical Report, National Marine Fisheries Service Circular 408. 167pp.

SKIPJACK PROGRAMME (1980). Skipjack fishing effort and catch, 1972-1978, by the Japanese pole-and-line fleet within 200 miles of the countries in the area of the South Pacific Commission. Skip]ack Survey and Assessment Programme Technical Report No.2. South Pacific Commission, Noumea, New Caledonia. 91pp. SKIPJACK PROGRAMME (1981). Fishing effort and catch by the longline fleets of Japan (1962-77) and Taiwan (1967-77) within 200 miles of the countries in the area of the South Pacific Commission. Skipjack Survey and Assessment Programme Technical Report No.3, South Pacific Commission, Noumea, New Caledonia. 257pp. 59

SKIPJACK PROGRAMME (1981a). Skipjack migration, mortality and fishery interactions. Regional Technical Meeting on Fisheries 1981. 13 : Working Paper No.9, South Pacific Commission, Noumea, New Caledonia. 35pp.

SKIPJACK PROGRAMME (1981b). Effects of skipjack tagging procedures on subsequent tag recoveries. Regional Technical Meeting on Fisheries 1981 13 : Working Paper No.8, South Pacific Commission, Noumea, New Caledonia. 14pp.

SKIPJACK PROGRAMME (1981c). Further observations on fishing performance of baitfish species in the South Pacific Commision area. Regional Technical Meeting on Fisheries 1981. 13 : Working Paper No.13, South Pacific Commission, Noumea, New Caledonia. 15pp.

SKIPJACK PROGRAMME (1981d). An appraisal of the genetic analysis of skipjack blood samples. Regional Technical Meeting on Fisheries 1981 , 13 : Working Paper No.10, South Pacific Commission, Noumea, New Caledonia. 8pp.

SMITH, B.R. (1977). Appraisal of the live-bait potential and handling characteristics of the common tuna bait species in Papua New Guinea. In Shomura, R.S. (Ed). Collection of tuna baitfish papers. United States Department of Commerce, National Oceanic and Atmospheric Administration Technical Report, National Marine Fisheries Service Circular 408:95-103.

WANKOWSKI, J.W.J. (1981). Estimated growth of surface schooling skipjack tuna, Katsuwonus pel amis, and yellowfin tuna, Thunnus albacares, from the Papua New Guinea region. National Oceanic and Atmospheric Administration, United States National Marine Fisheries Service Fishery Bulletin 79(3):517-532. 61

APPENDIX A. SCIENTISTS, OBSERVERS AND CREW ON BOARD THE RESEARCH VESSELS

South Pacific Commission Scientists

Jean-Pierre Hallier November 1 - December 12, 1977 May 25 - June 27, 1980 Robert Gillett November 1 - December 12, 1977 Antony Lewis November 1 - December 12, 1977, A.W. Argue May 25 - June 27, 1980 James Ianelli May 25 - June 27, 1980

Observers

Masakazu Yao November 1 - December 12, 1977 Fisheries Biologist Far Seas Fisheries Division Fishery Agency, Japan

Masao Hashizume November 1 - December 12, 1977 Fisheries Biologist Far Seas Fisheries Division Fishery Agency, Japan

Steven Kemp November 1 - December 12, 1977 Fisheries Officer Ministry of Natural Resources Solomon Islands

Jacob Rake November 1 - December 12, 1977 Fisheries Officer Ministry of Natural Resources Solomon Islands

Simon Wale May 30 - June 8, 1980 Fisheries Officer Ministry of Natural Resources Solomon Islands

Lionel Laka June 3 - June 6, 1980 Fisheries Officer Ministry of Natural Resources Solomon Islands

John Molia June 8 - June 23, 1980 Fisheries Officer Ministry of Natural Resources Solomon Islands

Gabriel Raywin June 8 June 23, 1980 Fisheries Officer Ministry of Natural Resources Solomon Islands 62

Japanese Crew Japanese Crew Cruise One Cruise Two

Masahiro Matsumoto, Capta:' Mitsutoyo Kaneda, Captain Ryoichi Eda Kenji Arima Sakae Hyuga Seima Kobayashi Yoshihiro Kondoh Yoshihiro Kondoh Yoshio Kozuka Yoshikatsu Oikawa Yoshikatsu Oikawa Tsunetaka Ono Akio Okumura Yukio Sasaya Kohji Wakasaki Kohji Wakasaki Mikio Yamashita Mikio Yamashita

Fijian Crew Fijian Crew Cruise One Cruise Two

Eroni Marawa Eroni Marawa Lui Andrews Lui Andrews Vonitiese Bainimoli Jovesa Buarua Mosese Cakau Samuela Delana Kitione Koroi Eroni Dolodai Jone Manuku Luke Kaidrukiya Isoa Rodan Metuisela Koroi Jeke Savirio Aminiasi Kuruyawa Ravaele Tikovakaca Sovita Lequeta Samuela Ue Josua Raguru Jona Ravasakula Napolioni Ravitu Tuimasi Tuilekutu Samuela Ue 63

APPENDIX B. BAIT SPECIES COMPOSITION, PERCENTAGE OF BOUKI-AMI HAULS CONTAINING A PARTICULAR SPECIES, AND ESTIMATED TOTAL CATCH (KILOGRAMS), FOR THE 1977 AND 1980 SKIPJACK PROGRAMME SURVEYS IN THE WATERS OF SOLOMON ISLANDS

1 9 8 0

Species Percentage Estimated Species Percentage Estimated Occurrence Catch Occurrence Catch (kg) (kg)

Stolephorus devisi 81 676 Herklotsichthvs punctatus 56 1140 Spratelloides gracilis 71 476 Stolephorus heterolobus 62 1008 Gymnocaesio gymnopterus 57 471 Rastrelliger kanagurta 31 945 Stolephorus beterolobus 71 305 Stolephorus devisi 72 769 Spratelloides (new species) 38 286 Stolephorus buccaneeri 44 498 Herklotsichthys punctatus 90 241 Sardinella sirm 33 322 Sardinella airm 57 39 Stolephorus bataviensis 82 279 Dipterygonotus leucogrammicus 67 38 Scomberoides tol 3 130 Thrissina baelama 48 38 Sardinella melanura 31 125 Pterocaesio sp. 43 19 Rastrelliger faughni 51 114 Pterocaesio pisang 57 19 Rastrelliger brachvsoma 26 88 BregmaceroB sp. 43 13 Spratelloides (new species) 49 85 Pterocaesio tile 24 10 Thrissina baelama 33 58 Dussumieria sp. 52 7 Spratelloides gracilis 38 57 Hypoatherina ovalaua 90 7 Hypoatherina ovalaua 33 41 Hypoatherina temmincki 10 7 Decapterus russelli Pranesus pinguis 10 31 48 7 Gymnocaesio gvmnopterus 18 22 Spratelloides delicatulus 62 Sphyraena sp. 4 Spratelloides delicatulus 38 18 Pterocaesio diagramma 10 0 Pranesus pinguis 33 12 Apogon(Rhabdamia) cypselurus 5 0 Gazza minuta 49 8 Sp. of Siganidae 67 0 Leiognathus bindus 21 7 Sp. of Holocentridae 38 0 Pranesus endrachtensis 21 4 Sp. of Synodontidae 52 0 Stolephorus indicus 18 4 Sp. of Aluteridae 19 0 Selar boops 26 3 Pseudamia sp. 19 0 Selar crumenonhthalmus 49 3 Pseudamia sp. 14 0 Atule sp. 8 1 Priacanthus crueatatus 5 0 Xiphasia setifer 3 0 Mullodichthys auriflamma 10 0 Zanclus canescens 3 0 Selar crumenophtbalmus 5 0 Sp. of Balistidae 3 0 Leiognathus bindu 33 0 Sp. of Tetrodontidae 3 0 Cheilodipterus macrodon 48 0 Scomberoides sp. 8 0 Secutor sp. 10 0 Sp. of Centriscidae 3 0 Archamia zosterophora 5 0 Pterocaesio pisang 3 0 Pellona ditchela 43 0 Sp. of Sphyraenidae 36 0 Trichiuris haumela 19 0 Caranx sp. 8 0 Gazza minuta 10 0 Decapterus macrosoma 5 0 Saurida undosquamis 43 0 Dussumieria sp. 69 0 Sardinella sp. 5 0 Sp. of Leiognathidae 5 0 Stolephorus bataviensis 5 0 Apogon(Rhabdamia) cvnselurus 15 0 Caranx sexfasciatus 33 0 Pellona ditchela 33 0 Benthosema sp. 19 0 Sp. of Squid 15 0 Leiognathus berbis 10 0 Pterocaesio sp. 8 0 Rastrelliger kanagurta 5 0 Sp. of Megalopidae 3 0 Stolephorus indicus 67 0 Scomberomorus commerson 8 0 Stenatherina panatela 43 0 Sp. of Lutjanidae 3 0 Hypoatherina sp. 5 0 Sp. of Holocentridae 5 0 Apogpn fragilis 5 0 Thrissocles setirostris 3 0 Pseudamia polvstigma 14 0 Sp. of Mullidae 8 0 Parapriacanthus dispar 14 0 Sp. of Chaetodontidae 5 0 Sp. of Mugilidae 10 0 Stenatherina panatela 3 0 Sp. of Balistidae 10 0 Sp. of Hemirhamphidae 3 0 Decapterus macrosoma 10 0 Sp. of Apogonidae 3 0 Caesio coerulaureus 10 0 Mene maculata Sp. of Lutjanidae 3 0 19 0 Saurida gracilis 8 0 Scomberomorus commerson 10 0 Sp. of Siganidae 5 0 Sp. of Bothidae 5 0 Sp. of Myctophidae 3 0 Pranesus duodecimalis 10 0 Archamia lineolata Atule mate 8 0 Gastrophysus sp. Lutjanus malabaricus 10 0 3 0 10 0 Secutor insidiator 3 0 Fiatularia sp. 10 0 Caranx sp. Sp. of Exocoetidae 5 0 5 0 Sp. of Acanthuridae 10 0 5 0 Sp. of Crustacea 3 0 Leiognathus elongatus 10 0 Sp. of Anguillidae (j) 3 0 Parupeneus sp. 10 0 Bregmaceros sp. 8 0 Sp. of Mullidae 10 0 Megalasuis cordvla 3 0 Benthosema fibulaturn 5 0 HemiramDhus far 3 0 Apofton(Rhabdamia) gracilis 14 0 Sp. of Trichiuridae 33 0 Sp. of Myctophidae 5 0 Priacanthus sp. 8 0 Sp. of Platycephalidae 5 0 Manta birostris 3 0 Sp. of Chaetodontidae 10 0 Apogon(Rhabdamia) gracilis 8 0 Sp. of Syngnathidae 10 0 Secutor sp. 13 0 Sp. of Canthigasteridae 5 0 Sp. of Carangidae 8 0 Sp. of Anguillidae (j) 14 0 Sp. of Synodontidae 8 0 Archamia lineolata 19 0 Cheilodipterus macrodon 5 0 Sp. of Crustacea 5 0 Archamia zosterophora 10 0 Sp. of Sphyraenidae 14 0 Sp. of Belonidae 3 0 Sp. of Acanthuridae 48 0 Fiatularia villosa 5 0 Sp. of Crustacea (j) 5 0 Sp. of Squid 10 0 Benthosema sp. 10 0 Selar boons 5 0 Mullodichthys samoensis 5 0 Decapterus sp. 14 0 Dianhus sp. 5 0 Upeneus sp. 5 0 65

APPENDIX C. LISTING OF TAG AND RECOVERY DATA FOR EACH TAGGED SKIPJACK AND YELLOWFIN THAT MADE AN INTERNATIONAL MIGRATION OUT OF OR INTO THE WATERS OF SOLOMON ISLANDS. Release data are presented for the school from which one or more tagged fish were recovered. The inset lines present release data as follows: country abbreviation; school number; year/month/day of release; time of release; latitude of release; longitude of release; numbers of tagged skipjack released; numbers of tagged yellowfin released; numbers of species other than skipjack and yellowfin that were tagged and released. Line(s) following that for release data present the following data for each tag recovery: species, S for skipjack, Y for yellowfin; recovery country abbreviation (see list); year/month/day of recovery; days at large; recovery latitude; recovery longitude; straight line (great circle) distance in nautical miles between release and recovery location; fork length in millimetres at time of tagging and length credibility code (see list); fork length at recovery and credibility code (see list); tag number; tag recovery gear (see list). The Appendix does not include recoveries for which country of recovery was unknown. Date or position of recovery was excluded if the range of possible values was more than half the span from the release date or release position to the midpoint of the range of possible recovery dates or positions. If the range was less than half of this span, the information was included and the date or position of recovery was taken to be the midpoint of the range.

Migrants from the waters of Solomon Islands to other countries

SOL 30 771109 1715 0857S 15846E 482 0 0 S PNG 780604 207 0230S 15045E 0615 503B 580W AY02276 PNGPOL

SOL 35 771114 1415 0854S 15830E 301 13 0 SPNG 780603 201 0230S 15045E 0601 500M U AY03041 PNGPOL

SOL 36 771114 1530 0859S 15830E 183 24 0 S PNG 780529 196 0330S 15050E 0563 525M 580W AY03381 PNGPOL

SOL 43 771125 1054 0841S 15821E 132 0 0 S PNG 780615 202 0300S 15000E 0604 530M 580W AY03939 PNGPOL

SOL 46 771203 1510 1105S 16536E 213 0 0 S KIR 780203 062 0125S 16826E 0604 550M 590J AY04409 JAPPOL

SOL 877 800604 0945 1131S 16644E 68 1 0 S FIJ 801114 163 1545S 17955W 0819 550M 450E 1E10638 FIJPOL

SOL 882 800611 1115 0744S 15811E 288 18 0 S PNG 810226 260 0153S 15632E 0365 510M 544W 1E11080 JAPSEN S PNG 810314 276 0342S 15126E 0470 510M 570W 1E11031 PNGPOL S PNG 810323 285 0359S 15147E 0443 510M 580W 1E10789 PNGPOL S PNG 810709 393 0359S 15147E 0443 525M 510W 1E11042 PNGPOL

SOL 886 800614 1030 0737S 15616E 120 146 0 S PNG 810102 202 0420S 15124E 0351 450M 500B 1E11596 PNGPOL 66

SOL 888 800615 1030 0740S 15618E 26 51 0 S PNG 801018 125 0459S 14910E 0455 570M 600B 1E12151 PNGPOL

SOL 891 800617 1100 0831S 15928E 246 0 0 S PNG 810611 359 0355S 15157E 0526 550M 600W 1E11788 PNGPOL

SOL 893 800618 1030 0840S 15936E 1533 26 0 S PNG 810618 365 350S 15130E 0563 460M 510W 1E12482 PNGPOL

SOL 897 800620 1335 0809S 16016E 125 0 0 S IND 801211 174 0055N 13915E 1370 600M 480E 1E14205 PALPOL

SOL 901 800622 1150 0807S 16020E 204 0 0 S PNG 800817 056 158S 15038E 0687 580M 560B 1E14595 PNGPOL S PNG 800825 064 0350S 15110E 0604 580M 620B 1E14982 PNGPOL S PNG 800829 068 0038N 14150E 1225 600M 602W 1E14819 JAPSEN

SOL 884 800613 1510 0747S 15621E 24 139 0 Y INT 801202 172 0405N 14025E 1190 700M 784W 1E11475 JAPSEN

Migrants into the waters of Solomon Islands

PNG 10 771026 1400 0707S 15530E 134 0 0 S SOL 781219 419 0830S 15900E 0224 550M 590B AB05755 SOLPOL

PNG 11 771026 1500 0706S 15525E 122 0 0 S SOL 780928 337 0910S 15910E 0255 541B 560W AB05783 SOLPOL

PNG 13 771026 1600 0705S 15522E 64 0 0 S SOL 780914 323 0900S 15858E 0243 520M 560W AB06825 SOLPOL

PNG 547 790520 1700 0736S 14947E 482 0 0 s SOL 790724 065 0835S 15700E 0433 490M 540W SK35668 SOLPOL s SOL 490M U SK35657 SOLPOL s SOL 791016 149 0730S 15630E 0400 550M u SK35736 SOLPOL s SOL 791018 151 0930S 15930E 0588 520M 530W SK35698 SOLPOL s SOL 800712 419 0830S 15930E 0580 511B 620W SK36067 SOLPOL s SOL 800715 422 0830S 15630E 0403 490M 620W SK35646 SOLPOL s SOL 800825 463 0800S 15600E 0370 511B 620W SK35824 SOLPOL PNG 551 790527 1600 0330S 15105E 95 0 0 S SOL 801203 556 0952S 15935E 0634 564B 600B SK36854 SOLPOL PNG 553 790528 0640 0325S 15051E 550 8 0 S SOL 800711 410 0830S 15630E 0455 530M 600W SK36507 SOLPOL

PNG 554 790528 1745 0328S 15053E 135 6 0 S SOL 801202 554 0840S 16032E 0655 520M 630B SK37700 SOLPOL

PNG 559 790530 1525 0415S 15236E 163 3 0 S SOL 800106 221 0945S 16025E 0570 530M 540W SK39243 SOLPOL S SOL 801102 522 0940S 16120E 0613 510M 610B SK37958 SOLPOL

PNG 562 790531 0805 0416S 15234E 143 1 0 S SOL 791018 140 0930S 15930E 0519 540M 510B SK39629 SOLPOL S SOL 791023 145 0830S 16030E 0537 530M 600W SK39345 SOLPOL 67

PNG 563 790531 0855 0408S 15234E 121 9 0 S SOL 800109 223 0730S 15630E 0310 480M 520W SK39816 SOLPOL

PNG 567 790531 1110 0418S 15233E 174 4 0 S SOL 790819 080 0730S 15700E 0328 560M 680B SK40142 SOLPOL

PNG 575 790603 1500 0404S 15101E 904 12 0 S SOL 800918 473 0830S 15800E 0494 550M 630W SK40397 SOLPOL

PNG 587 790608 1120 0345S 15048E 231 0 0 S SOL 800715 403 0830S 15630E 0444 550M 630W SK41618 SOLPOL

PNG 591 790609 1230 0406S 15233E 187 4 0 S SOL 800508 334 0730S 15830E 0410 540M 570W SK44614 SOLPOL

PNG 609 790623 0945 0320S 15102E 93 3 0 S SOL 800829 433 0850S 15840E 0562 540M 590W SK38548 SOLPOL

PNG 610 790623 1015 0319S 15058E 47 18 0 S SOL 801215 541 0910S 15810E 0554 570M 560B SK42772 SOLPOL

PNG 614 790626 0915 0221S 15026E 39 0 0 S SOL 650M SK42830

PNG 618 790627 1115 0153S 14950E 133 11 0 S SOL 810103 556 0900S 15930E 0718 440M 560B SK43037 SOLPOL

CAL 56 771216 1235 2100S 16545E 184 0 0 S SOL 780725 221 0900S 15830E 0833 540M 610B SA03036 SOLPOL

CAL 80 771228 1325 2212S 16603E 152 0 0 S SOL 800127 760 1258S 15902E 0684 430M 581W SA05113 JAPPOL

CAL 88 780103 1745 2035S 16609E 260 0 0 S SOL 780902 242 0905S 15945E 0783 480M 520B SA06311 SOLPOL

CAL 89 780104 1210 2125S 16614E 235 1 0 S SOL 780616 163 0845S 16030E 0829 526B 570E SA06473 SOLSUB

CAL 94 780107 1810 2127S 16618E 231 8 0 S SOL 780828 233 0950S 15830E 0829 440M 510W SA06918 SOLPOL

CAL 95 780108 1230 2142S 16640E 115 0 0 S SOL 781011 276 0725S 15715E 1016 410M 520W SA07123 SOLPOL

CAL 97 780109 1635 2237S 16737E 71 0 0 S SOL 780930 264 0910S 15910E 0942 500M 560W SA07548 SOLPOL

CAL 108 780113 1505 2148S 16528E 228 0 0 S SOL 780420 097 0850S 16030E 0829 498B U SA08209 SOLPOL

CAL 110 780115 0840 2058S 16424E 842 0 0 S SOL 780721 187 0845S 15852E 0800 510M 598W SA08479 SOLPOL S SOL 781123 312 0915S 15938E 0755 450M 520B SA08635 SOLPOL 68

VAN 117 780121 0710 1645S 16756E 225 0 0 S SOL 781008 260 0745S 15700E 0838 500M 550W SA09340 SOLPOL VAN 119 780121 1200 1615S 16751E 416 225 0 S SOL 781002 254 0900S 15900E 0676 500M 600W SA09760 SOLPOL S SOL 781017 269 0900S 16045E 0601 520M 570W SB00119 SOLPOL S SOL 791115 663 0900S 15920E 0661 470M 570W SB01080 SOLPOL

TUV 247 780625 1220 1023S 17848W 486 16 0 S SOL 790719 389 0900S 16035E 1222 490M 560B SF01788 SOLPOL S SOL 791125 518 0830S 15900E 1319 510M 550B SK21289 SOLPOL

WES 241 780614 1000 1342S 17145W 1637 56 0 S SOL 800514 700 0830S 16000E 1691 475M 580W SK20351 SOLPOL

NSW 496 790406 1410 3520S 15059E 157 0 0 S SOL 791228 266 0830S 15630E 1638 560M 650W SK06856 SOLPOL

NSW 503 790407 1605 3522S 15102E 47 0 0 S SOL 801006 548 0900S 15900E 1641 455M 520W SK06985 SOLPOL

NSW 506 790408 0840 3508S 15104E 194 0 0 S SOL 791119 225 0830S 15900E 1656 490M 580W SK07000 SOLPOL

NSW 507 790408 0920 3506S 15104E 745 0 0 S SOL 791125 231 0730S 15630E 1683 420M 550W SK07360 SOLPOL

NSW 513 790409 1100 3455S 15110E 119 0 0 S SOL 791125 230 0700S 15830E 1723 475M 520W SK08452 SOLPOL

NSW 515 790409 1605 3458S 15105E 764 0 0 S SOL 790724 106 0835S 15700E 1616 459B 540W SK08999 SOLPOL S SOL 800227 324 1028S 16905E 1769 459B 531W SK08929 JAPPOL S SOL 800428 385 0910S 15805E 1595 459B 570W SK09156 SOLPOL

QLD 523 790501 0800 1756S 14822E 457 0 0 S SOL 791126 209 1455S 16225E 0828 580M 700W SK09891 JAPPOL S SOL 791127 210 1450S 16227E 0831 639B 695W SK09659 JAPPOL

QLD 529 790502 1510 1736S 14809E 152 0 0 S SOL 790708 067 0845S 15705E 0744 604B 455B SK32617 SOLPOL S SOL 791227 239 0900S 15930E 0839 620M 690W SK32893 SOLPOL QLD 531 790503 1105 1622S 15012E 725 3 0 S SOL 790706 064 0800S 15730E 0659 480B 580B SK34843 SOLPOL S SOL 791011 161 0915S 15920E 0684 480B 530W SK34733 SOLPOL S SOL 791014 164 0930S 15930E 0682 480B 520W SK34231 SOLPOL S SOL 791019 169 0800S 15800E 0679 470M 550B SK34185 SOLPOL S SOL 791101 182 0900S 15700E 0595 490M 520B SK34529 SOLPOL S SOL 791129 210 1102S 15838E 0586 480B 500W SK32981 JAPPOL S SOL 791205 216 0825S 15930E 0724 480B 530B SK34721 SOLPOL S SOL 791219 230 0830S 15900E 0699 480M 520W SK34557 SOLPOL S SOL 791221 232 0830S 15900E 0699 480B 530W SK34759 SOLPOL S SOL 801018 534 0900S 15930E 0701 480B U SK34728 SOLPOL S SOL 801114 561 0920S 16037E 0741 480B 570B SK34844 SOLPOL S SOL 801203 580 0730S 15600E 0631 490M 620W SK34137 SOLPOL 69

QLD 532 790503 1200 1621S 15010E 71 0 0 S SOL 791111 192 0830S 15730E 0637 620M 650W SK34881 SOLPOL

QLD 533 790503 1250 1620S 15008E 135 0 0 S SOL 801202 579 0840S 16032E 0763 620M 640B SK35030 SOLPOL

QLD 534 790503 1310 1620S 15007E 171 0 0 S SOL 791124 205 0730S 15700E 0666 480M 560W SK35303 SOLPOL S SOL 791127 208 0835S 15940E 0727 500M 570B SK34693 SOLPOL S SOL 800625 419 0800S 15830E 0701 480M 580W SK34997 SOLPOL

QLD 539 790504 1722 1627S 15008E 212 0 0 S SOL 791209 219 1353S 15904E 0540 630M 670W SK35177 JAPPOL S SOL 791231 241 0900S 15930E 0707 590M 680W SK35242 SOLPOL S SOL 800227 299 1052S 15916E 0629 600M 710W SK35240 JAPPOL

PON 627 791111 1630 0739N 15520E 196 0 0 S SOL 801001 325 0920S 16040E 1068 320M 520B SC01704 SOLPOL

NOR 798 800327 1425 2837S 16757E 456 123 0 S SOL 801219 267 0830S 16020E 1281 450M 540W 2B22480 SOLPOL

PAL 929 800809 1600 0300N 13135E 85 16 0 S SOL 810528 292 0656S 15715E 1649 340M 540B 2E22153 SOLPOL

PAL 931 800811 0715 0257N 13140E 151 123 0 S SOL 810612 305 0730S 15900E 1752 422T 560W 2E22606 SOLPOL 70

CODES FOR LENGTH MEASUREMENTS, RECAPTURE GEARS AND COUNTRY ABBREVIATIONS

Release Length Credibility

M Measured B Estimated from Biological Data T Estimated from Tagging Data G Guessed U Unknown Q Length Questionable

Recapture Length Credibility

A Measured by Hatsutori Maru No.l SPC staff) B Measured by joint local ventures C Measured by Japanese long range boats, or long- liners of other nationalities D Measured by other supposedly reliable sources E Measured by unreliable sources W Measured length verified by weight J Estimated from weight K Estimated from other sources (string, etc.) U Unknown

Nationality of Recapture Vessel (Country Abbreviations)

AMS American Samoa CAL New Caledonia FIJ Fiji IND Indonesia INT International waters JAP Japan KIR Kiribati KOR Korea NOR Norfolk Island NSW New South Wales (Australia) PAL Palau PHL Philippines PNG Papua New Guinea POL French Polynesia PON Ponape (Federated States of Micronesia) QLD Queensland (Australia) SOC Society Islands (French Polynesia) SOL Solomon Islands TAW Taiwan TOK Tokelau TON Tonga TUV Tuvalu USA United States VAN Vanuatu WAL Wallis and Futuna WES Western Samoa ZEA New Zealand 71

Type of Recapture Vessel

SEN Purse-seine POL Pole-and-line LON Long-line SHE Pearl-shell trolling ART Artisanal GIL Gill net REC Recreational (sport fishing) SUB Subsistance (village) UUU Unknown 73

APPENDIX D. STOMACH CONTENTS OF YELLOWFIN SAMPLED BY THE SKIPJACK PROGRAMME IN THE WATERS OF SOLOMON ISLANDS

Item Diet Item Number of Percentage No. Stomachs Occurrence Fish and Invertebrates

1 Chum from Hatsutori Maru 52 56.52 2 Fish remains (not chum) 44 47.83 3 Alima stage (Stomatopoda) 43 46.74 4 Squid (Cephalopoda) 21 22.83 5 Shrimp (Decapoda) 21 22.83 6 Megalopa stage (Decapoda) 21 22.83 7 Stolephorus buccaneeri (Engraulidae) 17 18.48 8 Acanthuridae 15 16.30 9 Stomatopoda 11 11.96 10 Chaetodontidae 10 10.87 11 Phyllosoma stage (Decapoda) 7 7.61 12 7 7.61 13 Balistidae Unidentified fish 7 7.61 14 5 5.43 15 Leiognathidae Blue goatfish (Mullidae) 4 4.35 16 4 4.35 17 Decapterus sp. (Carangidae) Aluteridae 4 4.35 18 4 4.35 19 Trichiuridae 4 4.35 20 Anchovy juvenile (Engraulidae) 4 4.35 21 Argonauta (Cephalopoda) 3 3.26 22 Ostraciidae 23 Penaeid shrimp (Decapoda) 3 3.26 24 Dactylopterus orientalis (Dacylopteridae) 3 3.26 25 Empty stomach 3 3.26 26 Amphipoda 3 3.26 27 Pteropoda (Gasteropoda) 3 3.26 28 Tunicate (Urochordata) 2 2.17 29 Crustacean remains 2 2.17 30 Tetrodontidae 2 2.17 31 Tuna juvenile (Scombridae) 2 2.17 32 Bramidae 2 2.17 33 Priacanthidae 1.09 34 Holocentridae 1.09 35 Plant material 1.09 36 Siganidae 1.09 37 Carangidae 1.09 38 Oxystoma crab larva (Decapoda) 1.09 39 Zoaea stage (Crustacea) 1.09 40 Euphausiid (Euphausiacea) 1.09 41 Unidentified invertebrate 1.09 42 Gastrophvsus sp. (Lagocephalidae) 1.09 43 Exocoetidae 1.09 44 Octopus (Cephalopoda) 1.09 45 Anthiidae 1.09 46 Juvenile fish 1.09 47 Ammodytidae 1.09 Caesiodidae 1.09

Total Stomachs Examined 92