Previous Page Table of Contents Next Page


Some interaction issues in the fisheries for tunas and tuna-like fishes of the Indian Ocean

Michel Bertignac and David Ardill
Indo-Pacific Tuna Programme
P.O. Box 2004
Colombo, Sri Lanka

ABSTRACT

Increases in the catch of tunas and tuna-like fishes of the Indian Ocean have raised concerns about interactions between major fisheries exploiting those species. After a short presentation of the fisheries and their recent evolution, the paper reviews the main interactions as they currently appear. Each interaction issue is classified according to the type of competition which exists: fisheries catching the fish at the same stage of their life cycle in the same general area, fisheries exploiting fish at two different stages of their life cycle and fisheries exploiting the same stock in two different areas. Past and current work conducted to assess interactions are described and proposals are made for future studies. These include the development of models incorporating movement, the implementation of tagging experiments and increase in the collection of statistical data from coastal fisheries.

1. INTRODUCTION

Tuna fisheries of the Indian Ocean range from subsistence and commercial fisheries (pole-and-line, drift gillnets, purse seining, trolling, handline, etc.) in coastal countries to large industrial fisheries (purse seine and longline) on the high seas and in the Exclusive Economic Zones (EEZs) of several countries. The main particularity of Indian Ocean (compared with other oceans) is the relative importance of coastal fisheries, which have accounted for more than half of the catch of tunas and tuna-like species (Figure 1).

Figure 1. Relative importance of coastal and non-coastal fisheries in the total catch of tuna and tuna-like species in Indian Ocean (FAO Areas 51 and 57) from 1970 to 1992 (source: IPTP, 1994a).

During recent years, trends have been towards an increase in the activity of both artisanal and longline fisheries and the development of a new component, the industrial purse seine fishery, at the beginning of the 1980s. The evolution of both coastal and high-seas fisheries has resulted in a tremendous increase in total catch of tunas and tuna-like species. This growth has been particularly important for yellowfin, whose catches even exceeded those of skipjack in 1992. This difference constitutes another specificity of Indian Ocean compared to other oceans. It has also raised a growing concern about the impact that some fisheries may have on others. Some of those concerns1 can be summarised as:

1 The albacore drift gillnet fishery, which exploited small fish, may have had an impact on the longline fishery. As driftnetting has now stopped, this effect will disappear within a few years.

· what impact can the distant water fishing nation (DWFN) fisheries (industrial purse seine and longline) have on coastal country fisheries (subsistence and commercial),

· what impact can the development of the purse seine fisheries have on longline fisheries,

· what impact can the DWFN longline fisheries have on locally based longline fisheries, and

· what impact can the development of fisheries for small tunas and seerfish have on the catch of neighbouring countries?

This paper presents some of the major interaction issues that can, with present knowledge, be listed in the Indian Ocean. After a short presentation of the main fisheries and their evolution, it reviews some work already done on interactions, presents some studies currently being carried out and gives recommendations for future work.

2. GENERAL PRESENTATION OF THE FISHERIES FOR TUNAS AND TUNA-LIKE SPECIES

2.1 Evolution and Present Status of the Main Fisheries

2.1.1 Artisanal fisheries

In countries bordering the Indian Ocean, artisanal troll, handline and drift gillnet fisheries have traditionally exploited tunas and tuna-like species, in particular the neritic species such as seerfish, longtail tuna and kawakawa (Figure 2).

In the Maldives, a pole-and-line fishery and, in Pakistan and Sri Lanka, drift gillnets produce significant catches of tropical oceanic tunas. Typically, troll and pole-and-line fisheries exploit all sizes of skipjack and small yellowfin and bigeye tunas. Drift gillnet fisheries in Sri Lanka exploit the same sizes, but, in the Arabian Sea, these gears tend to catch intermediate-sized yellowfin and are the main gears exploiting seerfish and longtail tuna. Handline fisheries exploit large yellowfin, some bigeye and billfish. Artisanal fleets have tended in the past to remain within day-range of their bases. This tendency is changing rapidly, with drift-gillnetters from Pakistan, for example, taking trips of over one-month duration and ranging from the border with India to Yemen and Somalia. In Sri Lanka also, 10-m vessels are ranging off the west coast as far as Maldivian and Lakshwadeep waters.

Figure 2. Evolution of tuna and tuna-like catches in Indian Ocean (FAO areas 51 and 57) from 1970 to 1992, by species (a) and gear (b) (source IPTP, 1994a).

Figure (a)

Figure (b)

Neritic tunas and seerfishes are caught almost exclusively by artisanal fisheries. Longtail tuna, whose distribution is limited to the Indian Ocean and Indo-Pacific, is exploited increasingly by drift gillnet fisheries, particularly in the Arabian Sea and gulfs. The catch of this species was 46,000 mt in 1992 (Table 1); this catch exceeded the Indian Ocean bigeye catch, and is four times the current albacore catch. Seerfish are important throughout the shelf areas surrounding the Indian Ocean basin, and catches peaked at about 110,000 mt, equivalent to the artisanal skipjack catch and one third more than that of yellowfin. This resource is particularly valuable, as seerfish are high-value species in most coastal countries. Important catches of kawakawa are also reported, mainly in countries bordering the Arabian Sea, with more than 50,000 mt caught in 1992 (33,000 mt in India). The other small tunas are of minor importance in the Indian Ocean.

Table 1. Recent catch of small tunas and seerfishes in Indian Ocean.
Source: IPTP (1994a).

Unit: mt

Species

1991

1992

Seerfishes

110,609

105,658

Longtail

42,224

45,953

Kawakawa

51,348

51,707

Frigate and bullet

17,346

17,909

Other tunas (mostly small species)

50,609

50,715


2.1.2 Industrial fisheries

Industrial fishing is more recent in the Indian Ocean. Following trials in the mid-1950s, Japanese longliners entered the Indian Ocean, followed shortly after by the Taiwanese and in 1971 by the Korean fleet. In 1970, the industrial longline fishery was producing close to 100,000 mt, this figure climbing slowly to 150,000 mt in the succeeding twenty years, of which 90,000 mt were yellowfin, 30,000 mt bigeye and 10,000 mt albacore.

The most significant development in this fishery has been the entry into the eastern Indian Ocean and Arabian Sea of hundreds of longliners (mostly under 50 GT), mainly under Taiwanese and, more recently, under Chinese registry. Many of these vessels have re-flagged in coastal countries or are operating under charter or joint-venture arrangements which give them access to EEZs. At this stage, of the coastal countries, only Indonesia has a substantial domestic longline fleet working in the Indian Ocean.

Throughout the range of the longline fishery in tropical waters, yellowfin tuna are the main catch component. North of 10°N, the catch of other tuna species is negligible, but south of this, the main target species becomes bigeye tuna, except for the fishery to the south of Java, where southern bluefin are targeted. There is also a fishery targeting albacore which operates mainly south of the Tropic of Capricorn.

The major part of the production of the longline fleet is for the sashimi market. The large DWFN boats use low temperature freezing, while the coastal vessels conserve most of their catch on ice for air shipment. The albacore fishery produces fish for canning.

An industrial drift gillnet fishery for albacore has had a short life span. This fishery developed in the eastern Indian Ocean in 1985, moving into the western part of the ocean in 1987. The fishery stopped operating in 1992, following the global ban on the use of large-scale drift gillnets on the high-seas. Peak catches, attained in 1990, were 21,142 mt, pushing the total catch of this species beyond estimates of maximum sustainable yield (Hsu and Chang, 1994).

Purse seining started in the Indian Ocean in the early 1980s. The development was very rapid, as a consequence of movement of the French and Spanish fleets from the eastern Atlantic. Catches attained 200,000 mt in 1987 and 310,000 mt in 1992, 157,000 mt being skipjack and 110,000 mt yellowfin tuna. This fishery is conducted mostly by non-Indian Ocean countries, although a few local vessels operate under joint-venture arrangements.

The main purse seine fleets have two distinct fishery phases. From December to May when the thermocline is well established in the southern part of the Indian Ocean, the fishery targets free swimming schools of large yellowfin. Over the rest of the year, purse seining is conducted mainly on schools of skipjack and small yellowfin, with some juvenile bigeye, aggregated by flotsam or artificial drifting Fish Aggregating Devices (FADs). Between July and August, these fleets are mainly in the more-sheltered waters of the Mozambique Channel, catching mainly skipjack. The Japanese and Mauritian purse seiners fish on FADs throughout the year, and tend to centre on more northern waters. Some fishing is concentrated around seamounts where intermediate-sized yellowfin aggregate.

2.2 Statistical Data Available on Tuna Fisheries in the Indian Ocean

The most complete set of statistical data on tuna and tuna-like species in the Indian Ocean is the Indo-Pacific Tuna Programme (IPTP) nominal catch database. The data are aggregated by year, country, species and gear, and data sets generally go back to 1970. In many countries, however, statistics did not cover tunas adequately prior to the mid-1970s and the 1980s and, until that time, catches were probably greater than records suggest. In a number of cases, species and gear aggregations have been split into their constituent parts using sub-samples. The sub-samples have a limited temporal coverage, and their accuracy is, therefore, suspect.

For most coastal countries, the only information available on the location of catches comes from points of landing. This was adequate when fishing was essentially confined to day-range of the coast, but landings data became increasingly unsatisfactory as fleets range further afield. Equally, effort data are usually limited to fishing-craft statistics. Exceptions are the Maldives pole-and-line and Mauritian purse seine fisheries where detailed effort and length-frequency data are available. Sampling schemes in a number of countries have provided length-frequency data.

For the major industrial DWFN fisheries, catch and effort data are available, usually aggregated by month and 5° grid. For these fisheries, length-frequency data are also available. Non-aggregated data are available to the national institutions which monitor these fisheries, but are generally kept confidential.

2.3 Distribution and Stock Structure of Tunas and Tuna-Like Species in the Indian Ocean

Distribution of tunas and tuna-like species in the Indian Ocean is fairly well established through catch information. The tropical oceanic tunas and billfish range from the Arabian Sea and Bay of Bengal to between 30° and 40°S. Few bigeye, however, are caught north of 10°N. An important fishery for southern bluefin exploits a zone of concentration to the south of Java, but the juvenile migration routes pass around the south of Australia and the species is exploited between 40° and 50°S, from South Africa to Australia and New Zealand. The main longline fisheries for albacore extend slightly to the north of this zone, as far as 10°S in the first and fourth quarters of the year.

Neritic tunas are mainly caught on the shelf areas around the northern half of the Indian Ocean. Seasonal fisheries for both longtail and seerfish operate in the Arabian Sea and gulfs. The catch of seerfish on the east African coast suggests that these fish may undertake coastal migrations.

Very little is known on stock structures and the only study undertaken to date in the Indian Ocean was on yellowfin tuna (Nishida, 1992). This study hypothesised the existence of three stocks, covering the eastern, western and far-western Indian Ocean, with some degree of intermingling in adjacent areas. The study used analysis of variance for CPUE, age-specific CPUE and coefficient of variation for size (CVS) to distinguish among stocks. This approach needs to be confirmed by other methods. Mitochondrial DNA studies on juvenile yellowfin from various parts of the Ocean have not to date been reported on (S. Chow, IPTP, pers. comm.).

3. KINDS OF INTERACTIONS CONSIDERED

In this paper, interactions are restricted to what Kleiber (1994) called resource-mediated interactions. This excludes possible competition in marketing or interference between gears. Furthermore, some aspects of the harvest competition defined by Kleiber as secondary effects will not be considered. These secondary effects concern the exploitation by one fishery of the food of the target species of the other fishery or the effect of heavy exploitation of the spawning stock on recruitment. Those two aspects are, up to now, not thought to be significant in tropical tuna fisheries. The definition is therefore restricted to interactions which occur when a fish which could have been harvested by a fishery has been caught by another. If we follow Hampton (1994), the reasons for interactions can be categorised as:

Type A. Competition for fish at the same stage in their life cycle in the same general area by two or more fisheries.

Type B. The effect of fishing a stock at an early stage in its life cycle upon a fishery that exploits the stock at a later stage, typically with another gear.

Type C. The effect of fishing a stock in one area upon a fishery that exploits the stock elsewhere.

A summary of the main tuna fishery interactions that are of current concern in the Indian Ocean is presented in Table 2. The column labelled “Type” gives the category of interaction according to the above classification. Gears used in subsistence and commercial fisheries in coastal countries (pole-and-line, drift gillnets, purse seining, trolling, handline, etc.) have been aggregated under “small-scale fisheries”.

Table 2. Main tuna fishery interaction that are of current concern in the Indian Ocean.

Description

Species

Affecting FisheryArea/Countries involved

Gear

Affected FisheryArea/Countries involved

Gear

Type

Interaction between DWFN purse seine fishery and longline fisheries (DWFN and locally based)

yellowfin bigeye

WIO (Seychelles, Mozambique Channel, Chagos, moving to north Arabian sea)France, Japan, Spain, USSR

purse seine

All Indian OceanTaiwan, Japan, KoreaIndonesia, Mauritius, India, Oman, Pakistan, Sri Lanka, Yemen, Maldives

longline

mainly B or C, (some A)

Interaction between DWFN purse seine fishery and small scale fisheries

skipjack yellowfin bigeye (?)

WIO (Seychelles, Mozambique Channel, Chagos, moving to north Arabian sea)France, Japan, Spain, USSR

purse seine

MaldivesIran, India, Oman, Pakistan, Sri LankaComoros, India, Indonesia, Sri Lanka, Pakistan

pole-and-line
drift gillnet
handline
trolling

C
B
A, B or C

Interaction between DWFN longline fishery and locally based longline fishery

yellowfin bigeye

All Indian OceanTaiwan, Japan, Korea

longline

Indonesia, Mauritius, India, Oman, Pakistan, Sri Lanka, Yemen

longline

A or C

Interaction between fisheries for small tunas

seerfish longtail kawakawa

Arabian SeaYemen, Oman, UAE, Iran, Pakistan, India, Sri Lanka

drift gillnet trolling hand-line

Arabian SeaYemen, Oman, UAE, Iran, Pakistan, India, Sri Lanka

drift gillnet trolling handline

A or C


4. REVIEW OF MAIN INTERACTIONS IN THE INDIAN OCEAN

4.1 Interactions Between Purse Seine and Longline Fisheries

4.1.1 Description of the potential interactions

The longline fisheries exploit virtually the whole of the Indian Ocean, whereas the purse seine fishery has developed only in the western and central part. Potential interactions are of the three types, as the purse seine fishery exploits both juvenile yellowfin and bigeye associated with flotsam and large yellowfin in free-swimming schools. It must be noted that another possible type B interaction may exist here between purse seine fisheries catching different sizes of yellowfin. This however depends on how we define the term fishery, as many purse seiners are involved in both fishing techniques.

Yellowfin is probably the more affected species, as purse seine catches of bigeye amount to only about 8% of yellowfin catches. The increasing tendency to exploit flotsam and FAD-associated schools with deep nets, however, will add to fishing pressure on bigeye. Bigeye are commonly found at depths below the yellowfin and skipjack aggregations. This trend is confirmed by the recent increase in the purse seine catches of bigeye which accounted for half of the catch in weight of this species in 1991. As purse seiners are catching mainly small sizes of bigeye, this could represent a large number of fish and raise concern of type B interaction between purse seines and longlines.

4.1.2 Assessment of interactions

Analysis of catch and effort data, using a General Linear Model (GLM) has not been conclusive. Despite the sharp increase in the yellowfin catch of the purse seine fishery in the early 1980s (from 20,000 mt in 1983 to more than 100,000 mt in 1989), the CPUE of longliners has not appeared to be affected (Figure 3).

Figure 3. Trends in the longline CPUE and purse seine catch from 1970 to 1992. Longline CPUE have been standardised from Japanese, Taiwanese and Korean CPUE using GLM method. Source: IPTP, 1994b.

On the other hand, simulations conducted using fishing mortality by gear obtained after the use of catch-at-age data and sequential population analysis, but with no spatial pattern introduced in the model (IPTP, 1991, and Bertignac, 1994a), showed that interaction between purse seines and longlines should occur if the two fisheries are exploiting the same stock. Several hypotheses can explain this discrepancy:

i) the stocks available to the two fisheries are not (at least partially) the same,
ii) the stock is huge, or
iii) the measure of fishing effort is biased.
Concerning the first hypothesis, a way of getting complementary information (e.g., movement rates between fisheries, possibility of exchange between surface and sub-surface stocks) would be to conduct tagging experiments. Proposals were made at the last workshop on Indian Ocean Yellowfin stocks (IPTP, 1991) and reiterated at the last Expert Consultation on Indian Ocean Tunas (IPTP, 1994b) for the implementation of such a programme. Where tagging has been conducted in the past, recoveries in the longline fishery of fish tagged in surface fisheries have been less than expected. However, as mentioned by Hampton (1994), several explanations can be given to such discrepancy (lower reporting rates in the longline fishery, only partial recruitment to the longline fishery of the population of which the tagged fish are representative, existence of two different population implying that fish available to the longline fishery have never been available to the purse seine fishery) and “conventional” tagging does not provide enough information to discriminate between the fisheries. Tagging large yellowfin in both the surface and longline fisheries would help to answer questions on the availability of fish to both fisheries. Technically, such a tagging is possible as the South Pacific Commission Regional Tuna Tagging Project has tagged large yellowfin at the surface using short handlines. Some large line-caught bigeye and billfish have also been tagged successfully (Boggs, 1992). The use of sonic and archival tags might also help to establish behavioural differences which could explain varying vulnerability to surface and deep-water gears.

In regard to point ii), exploitation rates that can be obtained by tagging experiments (as described above) would also be useful. If we simply look at the FAO catch statistics, we can see however that compared to the yellowfin production of the western Pacific in 1992 (~400,000 mt), the Indian Ocean which is of equivalent size to the western Pacific has produced around 220,000 mt during the same period. Assuming an equivalent productivity (which might be a little bit simplistic), this would suggest that the Indian Ocean is capable of supporting larger catches of yellowfin.

To address point iii), it is obviously necessary to improve the standardisation of the longline CPUE data. This could be done by using a model incorporating biological and oceanographic data such as the one currently developed on Pacific billfishes (Bayliff, 1994, page 5).

For bigeye, very few studies have been conducted to date. Analyses have been made using a surplus production model which did not incorporate the purse seine fishery (Hsu and Chang, 1994). No interaction analysis has been carried out. As with yellowfin, the development of an age-structured model would be recommended if the objective is to study interactions (purse seine vs. longline). We saw above that seiners are catching large numbers of young bigeye. However, availability of data is much lower than for yellowfin and we know little of the biology (e.g., growth, movement). Furthermore, the quantity of bigeye caught by coastal fisheries is still unknown and is probably underestimated, due to the difficulty of distinguishing small bigeye from yellowfin.

4.1.3 Data requirements

IPTP holds catch and effort and length-frequency data for both purse seine and longline fisheries. These data would be satisfactory for age-structured models. Tagging is expected to improve our knowledge of growth for yellowfin (and maybe bigeye) as well as give some insight on movement needed for models incorporating a spatial component.

4.2 Interactions Between Purse Seine and Small-Scale Fisheries

4.2.1 Description of the potential interactions

In the Maldives, an important fishery using pole-and-line as the main gear is presently catching more than 60,000 mt of skipjack and 10,000 mt of yellowfin. This fishery is of prime importance for the country, and concerns exist about the increases in the catch of purse seiners. FADs deployed by purse seiners on the high seas have drifted ashore on the islands under the effect of the Equatorial Countercurrent. This has caused concern among small-scale fishermen in the Maldives that schools contributing to the influx of tunas could be caught prior to their arrival in coastal waters. Sri Lanka is also dependent on skipjack and small yellowfin for its drift gillnet fishery and could be similarly affected. The problem could be compounded by the fact that the purse seine fishery is moving further north and east year by year. The interactions here would be of types A or C.

Other locally-based commercial fisheries which could be affected, at long range, by the development of the purse seine fishery are the drift gillnet fisheries of the northern part of the Arabian Sea. The target species is yellowfin of an intermediate size, so that type B interactions are possible.

Similar type A, B or C interactions may occur between purse seine fisheries with countries using traditional gears such as troll lines (type A or C) and handlines (type B). While the effect on total catches might be minor, the negative effect on the livelihood of fishers could be significant.

4.2.2 Assessment of interactions

Few studies have been conducted to assess the effect of the purse seine fishery on the pole-and-line and drift gillnet fisheries. The simulations on the yellowfin fisheries mentioned above (4.1.2) indicated a low level of interaction, although no spatial pattern was introduced in the model. To get information on the migration patterns and status of stocks in the Maldives, a tagging experiment was conducted in 1990. Of the 8,033 skipjack and 1,908 yellowfin tagged, 1,407 and 128 have been recovered, respectively. This tagging experiment has given information on movement of fish inside and outside the area, and estimates of natural and fishing mortality (Yesaki and Waheed, 1992; Bertignac et al., 1994; Bertignac, 1994b). The diffusion rate of skipjack inside the Maldives was found to be low compared with other regions (e.g., the western Pacific), which seems to indicate that some sort of a local population exists in the Maldives. At the same time, exploitation rates obtained were moderate, indicating a low level of fishing mortality compared with natural mortality and emigration from the area.

To date, no estimate has been made of the recruitment to the fishery, as fish were tagged only in the Maldives, but there is no reason to believe that recruitment has not been sufficient to compensate the losses due to fishing, natural mortality and emigration. This should reduce concerns about the status of the Maldivian skipjack fishery.

Some tag recoveries were made outside the Maldives, showing that at least some skipjack and yellowfin migrate from Maldives to Sri Lanka and the western Indian Ocean, where other important fisheries exist. Those recoveries were, however, not sufficient to quantify movement rates outside the Maldives. A second tagging programme is currently being carried out. Its objectives are to put more emphasis on yellowfin and offshore skipjack tagging. Some tetracycline injection is being conducted for growth studies, as well as double tagging to more accurately estimate tag-shedding parameters.

As mentioned above, there is not much concern about the skipjack stock to date. This may not be the case for yellowfin where, although no signs of interaction are visible, the stock has a higher probability for over-exploitation. Where we are dealing with type C interactions, rates of movement of fish between fisheries and local recruitment are the key elements. The regional tagging project mentioned above could address those aspects. Tagging results would be then used in simulation models incorporating movement between fisheries, permitting the estimation of interaction. This type of model requires good-quality data.

4.2.3 Data requirements

Accurate and complete data are available on the industrial purse seine fisheries. For coastal fisheries, catch and effort data are of relatively good quality in the Maldives, but this may be not the case for other coastal fisheries, e.g., those bordering the Arabian Sea. Some improvements are thus required in the collection and compilation of statistics from those countries. The approach recommended by IPTP is to establish sampling schemes for representative fisheries in the countries concerned, using the data to split aggregated figures available from national statistical systems. The IPTP has assisted by designing sampling schemes, developing software for data entry and analysis and training staff in Sri Lanka, Pakistan and Iran and has, in certain cases, provided equipment and funded staff for limited periods.

4.3 Interactions Between Longline Fisheries

4.3.1 Description of the potential interactions

Interactions can include type A (local) but are more likely to be of type C (spatial), as most of the DWFN longline effort is in high-seas areas. Yellowfin and bigeye are the main species involved in this interaction issue. Large fishes are targeted. Both DWFN and coastal longline fisheries are of importance for the region, as they are a source of income and social benefits for many countries through license agreements, chartering, and induced activities of landings.

4.3.2 Assessment of interactions

No assessments have been conducted on this interaction issue. Although data are still not available, reports indicate a drop in the longline CPUE in the Arabian Sea in 1993 (G. Carrara, pers. comm.). This may be due to a number of factors, including recruitment and vulnerability of the fish to this gear, but could also be an indication of type A or C interactions between longline fisheries. If the growth of the fishery was to continue at the present rate, some acute type A interactions could emerge. Future studies should concentrate on yellowfin and bigeye, which are the main species caught. In the case of type C interactions, the main problem will concern the rates of movement of large yellowfin and bigeye, which are unknown. Here also, tagging with fish released in both fisheries will give some information useful for the development of spatial models.

4.3.3 Data requirements

The IPTP has not to date obtained catch and effort or length-frequency data from coastal longline operations, except for an extensive length-frequency data set covering one year for landings at Penang. Contacts are being established with authorities in India, Pakistan, Iran, Oman and possibly Yemen for establishing logbook, observer or shore-sampler programmes to obtain the desired data. Observers may be able to obtain data on discards, which are important in longline fisheries producing tuna for the sashimi market.

4.4 Interactions Between Fisheries for Small Tunas and Seerfishes

4.4.1 Description of the potential interactions

The recent drop in the catches of seerfish and longtail tuna in Oman (Figure 4) has raised concerns about the status of these resources and possible interaction between fisheries in the region.

Figure 4. Seerfish and longtail tuna catches in Oman.

The main gears used are drift gillnets, with some catch made by handline or trolling. Little is known of the stock structure and age distribution of the catch. It is thus difficult to say which type of interaction we are dealing with. We can however assume that it is mainly a type C interaction with some type A interaction. Unknown quantities of small seerfish are also caught by trawlers, particularly in India. In that case, a type B interaction might exist.

4.4.2 Assessment of interactions

A preliminary assessment of interaction in a fishery for small tunas has been conducted in the Gulf of Thailand. Simulation studies were carried out using a length-based model aggregated in space (Bertignac and Morón, 1994). The study underlined the impact that a fishery harvesting large quantity of juveniles of small tunas (mainly kawakawa) can have on other fisheries targeting larger fish (type B interactions). It shows, however, the limitations of this type of model when the whole distribution of the stock is not covered by the analysis.

No assessments have yet been undertaken of interaction between fisheries for small tunas and seerfishes in Indian Ocean sensu stricto. Following recommendations made at the last Expert Consultation on Indian Ocean tunas in Mahé, Seychelles, 1993, a project to assess seerfish and longtail tuna fisheries in the Arabian Sea and gulfs has been made by IPTP. The preliminary intention is to use multi-gear yield-per-recruit models aggregated in space. Such a simulation model, with only one gear (drift gillnet), was used to analyse the seerfish fishery in Oman and has been promising (Dudley et al., 1992; Bertignac and Yesaki, 1994). The inclusion of gear components in this model would help to assess whether type B interaction can potentially be important (see above the remarks on seerfish catches by trawlers). If type C interactions also prove to be important, the inclusion of spatial dimensions will be necessary and involve the estimation of movement rates between fisheries. Tagging of longtail does not appear to be an insurmountable problem, but to our knowledge, no tagging of seerfish has been conducted to date.

4.4.3 Data requirements

In the project mentioned above, proposals are made for the collection of data on catch and effort in each country involved in the fishery and on length-frequency distribution of the catch. Compilation of all information available on the fisheries and collection of biological data are also requested.

5. CONCLUSIONS

This paper represents an attempt to review the main fishery interaction issues in the Indian Ocean as they appear with the present knowledge we have of the fisheries for tuna and tuna-like species. Several potential interactions currently considered as minor have not been mentioned. These include, for example, the potential impact of longliners on local sport fisheries for billfish. Even if they are not of prime importance for Indian Ocean as a whole, such interactions can be important locally, and should not be ignored. As stated above, interactions which may have occurred between drift gillnet and longline fisheries for albacore have not been examined as, with the ban which has taken effect on drift gillnet fishing, no further research or management is justified. Similarly, interactions in southern bluefin fisheries are not examined as they are being tackled by the newly established Convention for the Conservation of Southern Bluefin Tuna.

Whatever types of interaction we have to address, having reliable statistics is an important element. Up to now, we can assume that there has not been a major problem with industrial fisheries (purse seiners and longliners) where, generally speaking, data are of good quality (catch and effort statistics and length-frequency distributions) and are sent to IPTP with acceptable delays (IPTP has recently faced problems to get catch and effort and length-frequency distribution from Korean longliners and Japanese data are submitted with considerable delay. The latter problem is expected to disappear with the development by Japan of a real-time catch reporting system).

On the other hand, data from the recently-developed coastal longline operations have not yet reported their catches to IPTP. A sampling programme was implemented in Malaysia, but was not maintained at the termination of IPTP support. Contacts have also been established with local authorities (in India, Pakistan, Iran, Oman and Yemen) to establish logbook or observer data collection systems. This is expected to produce information in the future on catch and effort and length distribution of catches. Observers could also obtain information on discards which are known to be important in the longline fisheries producing tuna for the sashimi market.

Another major problem concerning statistics is the availability of data from coastal fisheries. Although the Maldives are reporting good quality catch and effort statistics, statistical systems in bordering countries are often producing aggregated data. These data are difficult to use in any assessment studies. Furthermore, few length-frequency distributions of the catch have been collected to date. In some countries, statistics produced now are even of poorer quality than in the past. However, it seems hard in the present situation to improve significantly on that aspect due to the characteristics of all artisanal fisheries (small quantities landed in a large number of landing sites), the cost of the sampling (money and time) and the difficulty of monitoring enumerators. Tuna and tuna-like species may also not support priority fisheries in many of the countries concerned, justifying the necessary detailed statistical treatment. The IPTP has tried to mitigate the lack of data by implementing sampling programmes (Malaysia, Sri Lanka, Pakistan, Iran and Indonesia), but these are limited in space and time.

In addition to the quality of data, the type of data to collect is an important issue. We have seen that, for yellowfin and bigeye, potential interaction issues were mainly a combination of type B and type C (with some type A in areas where both purse seiners and longliners operate). In such cases, assessment of interactions could involve the use of simulation models incorporating both movements and size/age distribution of the catch. Such model should allow the estimate of the effect of variations in the activity of one fishery on the performance of the others under various conditions. Some progress is still to be made in the study of the growth of yellowfin and bigeye and the collection of length-frequency distributions for bigeye. No information is available on movement of either species. The large-scale tagging programme proposed for the Western Indian Ocean should provide such information and should try to investigate questions related to growth of yellowfin or/and its availability to surface and longline gears. Bigeye will presumably be tagged on an opportunistic basis, but it may be difficult to capture sufficient juvenile fish to study interactions in the small sizes. Tag recoveries could be analysed using tag attrition models incorporating movements (Sibert and Fournier, 1994). The objectives and implementation of the tagging programme are still to be finalised. Some simulations are currently being conducted by IPTP to investigate different patterns in the tagging process. In parallel to these tagging experiments, IPTP should try to improve data collected from drift gillnet fisheries in the Arabian Sea (catch and effort and length frequency distribution).

The main interaction issue for the stock of skipjack is a type C interaction between the purse seiners in the western Indian Ocean and the pole-and-line fishery in the Maldives. Tagging conducted in the latter fishery has shown that, locally, the fishing pressure is still low. For a complete picture of the stock dynamics, however, estimation of movement rates between the fisheries, population size and fishing and natural mortality could be made, using spatial models (Sibert and Fournier, 1994). Estimates of local recruitment are also important. As this is not a type B interaction, the size or age of the fish caught is not a major issue in a preliminary approach. All this would imply the implementation of a large scale tagging programme similar to the South Pacific Commission Regional Tuna Tagging Project. As the Indian Ocean skipjack stock is probably less heavily exploited than those of yellowfin and bigeye (as it is the case in other oceans), such a tagging programme, thus, may not be a priority in the medium term.

The project proposed by IPTP for the creation of a working group on seerfish and longtail tuna fishery in the Arabian Sea should permit the conduct of preliminary assessments of those stocks. Emphasis will be put on seerfish, which is economically more important for the region and for which the use of methods separating modal size groups, e.g., the Bhattacharya method described in Bhattacharya (1967) or Multifan (Fournier et al., 1990) seem to be particularly appropriate. Assessing the impact each gear/country might have on the others will necessitate the collection of catch and effort data and length-frequency distributions by fishery component. A spatial dimension could also be incorporated into the model. Tagging experiments may provide some insight into these matters.

Although statistical analyses of fishery data (CPUE and catches) have not yet been conclusive and are in any case difficult to interpret from an interaction point of view, they still need to be investigated further. If GLM techniques are to be used, very precise catch and effort data are needed (possibly on a daily basis), together with operational data, boat and gear characteristics, environmental data associated with the catch, etc. to permit the separation of major factors playing a role in catch rates. Although the present data compiled by IPTP covers the needs partially, some progress is still to be made, depending on the species considered.

6. REFERENCES CITED

Bayliff, W.H. (ed.). 1994. Quarterly Report of the Inter-American Tropical Tuna Commission, October to December 1994. Quarterly Report I-ATTC. 50 p.

Bertignac, M. 1994a. Preliminary assessment of interactions between Indian Ocean yellowfin tuna fisheries. In: Ardill, J.D. (ed.). Proceedings of the Expert Consultation on Indian Ocean Tunas, 5th Session, Mahé, Seychelles, 4-8 October, 1993. IPTP Collective Volumes No. 8: 209-211.

Bertignac, M. 1994b. Analysis of skipjack (Katsuwonus pelamis) tagging data in the Maldives Islands using a spatial tag attrition model. In: Ardill, J.D. (ed.). Proceedings of the Expert Consultation on Indian Ocean Tunas, 5th Session, Mahé, Seychelles, 4-8 October, 1993. IPTP Collective Volumes No. 8: 231-238.

Bertignac, M., P. Kleiber and A. Waheed. 1994. Analysis of Maldives Island tuna tagging data with a spatially aggregated attrition model. In: Ardill, J.D. (ed.). Proceedings of the Expert Consultation on Indian Ocean Tunas, 5th Session, Mahé, Seychelles, 4-8 October, 1993. IPTP Collective Volumes No. 8: 226-231.

Bertignac, M., and J. Morón. 1994. Preliminary assessment of interactions in the fishery for small tunas off the South China Sea coast of Thailand and Malaysia. In: Ardill, J.D. (ed.). Proceedings of the Expert Consultation on Indian Ocean Tunas, 5th Session, Mahé, Seychelles, 4-8 October, 1993. IPTP Collective Volumes No. 8: 203-209.

Bertignac, M., and M. Yesaki. 1994. Preliminary assessment of the narrow-barred Spanish mackerel stock off Oman using catch-at-age data from length-frequency distributions by the Bhattacharya method. In: Ardill, J.D. (ed.). Proceedings of the Expert Consultation on Indian Ocean Tunas, 5th Session, Mahé, Seychelles, 4-8 October, 1993. IPTP Collective Volumes No. 8: 88-95.

Bhattacharya, C.G. 1967. A simple method of resolution of a distribution into Gaussian components. Biometrics 23: 115-135.

Boggs, C.H. 1992. Depth, capture time, and hooked longevity of longline-caught pelagic fish: Timing bites of fish with chips. Fish Bull., U.S. 90(4): 642-658.

Dudley, R.G., P.A. Arundhati and E.B. Brothers. 1992. Management of the Indo-Pacific Spanish mackerel (Scomberomorus commersoni) in Oman. Fish. Res. 15: 17-43.

Fournier D.A., J.R. Sibert, J. Majkowski and J. Hampton. 1990. MULTIFAN, a likelihood-based method for estimating growth parameters and age composition from multiple length frequency data sets illustrated using data for southern bluefin tuna (Thunnus maccoyii). Can. J. Fish. Aquat. Sci. 47: 301-317.

Hampton, J. 1994. A review of tuna fishery interaction issues in the western and central Pacific Ocean. In: Shomura, R.S., J. Majkowski and S. Langi (eds.). Interactions of Pacific tuna fisheries. Proceedings of the First FAO Expert Consultation on Pacific Tuna Fisheries, 3-11 December 1991, Noumea, New Caledonia. Vol. 1: Summary report and papers on interaction. FAO Fish. Tech. Pap. (336/1): 138-151.

Hsu, C.C., and H.C. Chang. 1994. An ASPIC analysis of Indian Ocean bigeye tuna stock. In: Ardill, J.D. (ed.). Proceedings of the Expert Consultation on Indian Ocean Tunas, 5th Session, Mahé, Seychelles, 4-8 October, 1993. IPTP Collective Volumes No. 8: 124-128.

IPTP. 1991. Report of the Workshop on stock assessment of yellowfin tuna in the Indian Ocean, Colombo, Sri Lanka, 7-12 October 1991. Indo-Pacific Tuna Programme IPTP/91/GEN/20: 90 p.

IPTP. 1994a. Indian Ocean and Southeast Asian Tuna Fisheries Data Summary for 1992. Indo-Pacific Tuna Programme Data Summary No. 14: 147 p.

IPTP. 1994b. Report of the Expert Consultation on Indian Ocean Tunas, 5th Session, Mahé, Seychelles, 4 - 8 October 1993. Indo-Pacific Tuna Programme IPTP/94/GEN/22: 32 p.

Kleiber, P. 1994. Type of tuna fishery interactions in the Pacific Ocean and methods of assessing interaction. In: Shomura, R.S., J. Majkowski and S. Langi (eds.). Interactions of Pacific tuna fisheries. Proceedings of the First FAO Expert Consultation on Pacific Tuna Fisheries, 3-11 December 1991, Noumea, New Caledonia. Vol. 1: Summary report and papers on interaction. FAO Fish. Tech. Pap. (336/1): 61-73.

Nishida, T. 1992. Considerations of stock structure of yellowfin tuna (Thunnus albacares) in the Indian Ocean based on fishery data. Fish. Oceanogr. 1(1): 10 p.

Sibert, J., and D. Fournier, 1994, Evaluation of diffusion-advection equations for the estimation of movement patterns from tag recapture data. In: Shomura, R.S., J. Majkowski and S. Langi (eds.). Interactions of Pacific tuna fisheries. Proceedings of the First FAO Expert Consultation on Pacific Tuna Fisheries, 3-11 December 1991, Noumea, New Caledonia. Vol. 1: Summary report and papers on interaction. FAO Fish. Tech. Pap. (336/1): 108-121.

Yesaki, M., and A. Waheed. 1992. Results of the tuna tagging programme conducted in the Maldives during 1990. Indo-Pacific Tuna Programme IPTP/92/WP/24: 23 p.


Previous Page Top of Page Next Page