Previous Page Table of Contents Next Page


The Precautionary Approach creates an atmosphere in which RFBs, States, and participants in the fisheries for tunas and tuna-like fishes have a common interest in reducing uncertainty in fisheries data, and in establishing standards for quality control, methods of verification, confidentiality of proprietary information, transparency in data processing, and timely data sharing and exchange. The RFBs use two approaches to obtain data, acquisition of information directly from participants in fisheries and acquisition of aggregated data from national organizations.

Two articles of the UN Fish Stocks Agreement addressing directly the issues of responsibility of fishing nations in fisheries data and statistics are:

Article 8.4. Only those States that participate in the work of a subregional or regional fisheries management organization or arrangement, or that otherwise cooperate in the application of conservation and management measures established by that organization or arrangement, shall have access to the fishery to which those conservation and management measures apply, and

Article 17.1. A flag State whose vessels fish on the high seas shall take the necessary measures to ensure that vessels flying its flag comply with subregional and regional conservation and management measures.

All logbook, landings, processing, trade, and sales data should be verified, which is not an easy task. For example, even with the best of data systems, errors are made in data entry. They are often easily identified, and they can be verified and corrected if cost and time are not critical. However, misreporting, non-reporting, and intentional falsification of data are much more serious problems, and not so easily identified, validated, and corrected. They require that the data manager have a comprehensive knowledge of the fisheries, of monitoring and sampling techniques for fisheries, and of data base management, and an understanding of the principles and means by which cross checking and validation of data from different sources may be accomplished.

4.1 Data confidentiality

The RFBs must maintain the confidentiality of information for individual business enterprises. Information obtained from participants in the fisheries is intellectual capital, and, as with property of any individual, when in the hands of others it must be guarded against loss or devaluation resulting from the manner in which it is handled. The level at which confidentiality of information can be ensured determines in large part the ease with which information is obtained, its level of detail, and its accuracy. The RFBs must recognize that information remains the property of its source, and must not be released in formats or manners that compromise its value to the individuals from which it was obtained. The ability of the RFBs to achieve these levels of confidentiality is ensured through commission resolutions or statements of policy. Further, the nations in which the headquarters of these RFBs are located must not interfere with RFBs' need to maintain the confidentiality of their records. The secretariat of the CCSBT currently does not hold data, but confidentiality of individual data is ensured by the policies and practices of data-sharing by the national program participants.

The RFBs, recognizing that there is a need to provide data to the public in formats that as are as detailed as possible, but which do not compromise confidentiality, have established mechanisms for release of data to the public, including publishing data in various reports, presenting it at meetings, and posting it on their web sites. In many cases the data are aggregated by species, gear, 5-degree areas, and quarters of the year.

A number of mechanisms to protect confidential data have been adopted by the RFBs. These mechanisms provide that those conducting basic research may request access to confidential data by stating the nature of information sought and the purpose for which the information will be used. On consideration of the request, and possible consultation with the sources of the data, access to the information may be provided to the requesting party. Such access is conditional with restrictions placed against subsequent release to others and with a requirement that an account of the results obtained from the use of the data be presented prior to publication or release of the results.

While it is recognized that the detailed data required by the RFBs to meet their objectives and mandates must be available for analysis while being held in confidence, it is also recognized that there must be a transparency in the compilation and sharing of data which leads to mutual trust by the concerned parties that all relevant information are valid and adequate to meet the objectives and mandates of the RFBs. In various instances, nevertheless, the contracting and collaborating parties have denied access of the RFBs to the detailed and/or aggregated data required to meet their objectives and mandates.

4.2 Principal fisheries

4.2.1 Purse-seine fisheries

Purse-seine logbook data provide information on individual sets, including position, times of initiation and completion of the set, type of set, e.g. associated with flotsam or a FAD, associated with marine mammals, or unassociated, use of aircraft, estimated catch by species or size-based species aggregates, wells in which the fish were stored, etc. In general, the species composition of the catch and weights of fish caught recorded in the logbooks are verified by comparison to landings data. While the logbooks may request that the fishermen record information on discards, fishermen generally do not report these data, and they are generally available only from data recorded by observers.

The aggregated data provided to ICCAT and the IOTC are provided by 1-degree latitude by 1-degree longitude by half-month (or month) by the States in which the vessels are registered. There may be delays of up to one year in the provision of these data. The States often provide data for purse seiners owned by their nationals that are registered in other nations, sometimes without specifying the flag. Because it is the responsibility of the States in which the vessels are registered to provide these data, failure of national organizations to identify catches by vessels operating under different flags may lead to double counting.

The species and size compositions of the landings are estimated by port sampling, although some very limited data on size are collected by observers. In general, port-sampling programs obtain data for all tunas and tuna-like species, although the IATTC regularly samples only yellowfin, skipjack, bigeye, bluefin, and black skipjack, Euthynnus lineatus. It samples billfishes on an ad hoc basis, however. The data on catches and nominal fishing effort from the principal purse-seine fisheries are generally considered reliable. Also, as small tuna are often discarded at sea, this fraction can be sampled only by observers.

Another component of the purse-seine fishery, comprised generally of vessels of limited operational capabilities, can be identified as participating in fisheries for tunas and other species, such as sardines, anchovies, and mackerels. The vessels in this component of the purse-seine fishery are based in many widely-distributed ports, and, though logbook data are available for some of these, others may escape the monitoring efforts of the RFBs, and sometimes the national organizations as well.

4.2.2 Longline fisheries

Longline logbook data, which are generally available only for larger, high-seas longline vessels that use freezer systems for catch preservation, provide information by individual set, including positions at the start and end of the set, and retained catches by species, in weight and/or numbers of fish. They may also include data on species that are discarded at sea. Some logbooks may also include data on gear configuration, construction materials, bait, and species at which the fishery was directed. There may be considerable delays in the submission of logbooks to national authorities, and, except for the southern bluefin tuna fishery, wherein catches are also reported via telecommunications, this may result in delays of up to two years in provision of aggregated data to the RFBs. The data on the catches and nominal fishing effort for some of these longline fisheries are reliable, although delays in availability and, in some situations, deficiencies in coverages of the data are cause for concern.

Data from certain components of the longline fisheries, principally smaller, artisanal-type vessels and fleets that operate from widely-distributed ports and coastal locations are extremely difficult to obtain on a regular basis, and thus may remain essentially unknown. Many small longliners, with their operational capabilities limited by the use of ice or refrigerated seawater (RSW) for preservation of their catches, operate in the Indian and Pacific Oceans. These vessels are based in many widely-scattered ports, many, if not most, of which are not regularly visited by personnel of national organizations or RFBs. In many cases information on the activities of these vessels is not reported to either the nations in which the vessels are registered or the nations in which the owners or operators reside. Fortunately, however, the coastal States are beginning to require that vessels registered under their flags and report their catches to them. In general, estimates of the catch and nominal effort in historical data bases for this segment of the longline fisheries are considered to be underestimates.

Size-frequency data for catches made by larger, high-seas longline fisheries are generally problematic. Measurements are sometimes obtained by the crews of these vessels from small, non-random selections of the fish caught. Fortunately, however, reliable size-frequency data are collected aboard Japanese training vessels. Port-sampling programs are of limited value because these vessels make trips of long duration, and the catches are generally not identified as to individual sets or small area-time strata. Also, the catches are frequently transshipped, which adds further uncertainty to identification of the area-time strata for catches sampled at some stage during the transshipment process. The fish are normally dressed at sea, which, for tunas, usually includes removal of the gills, internal organs, and caudal fins. Dressing of small billfishes includes removal of the heads, as well, and large billfishes are filleted at sea. For sharks, in many cases, only the fins are retained. If weight data are used it is necessary that the condition of the fish at the time of weighing be known and that factors for conversion of dressed weights to undressed weights be available.

Size-frequency data for catches made by segments of the longline fishery with limited operational capabilities (artisanal and ice or RSW vessels) may also be problematic. Fisheries information on area-time strata in which individual fish were captured may be better determined for these, but the wide distribution of the ports in which these vessels are based has generally limited the acquisition of data. As a result, size-frequency data from catches made by this segment of the longline fisheries, with some exceptions, have been generated for ad hoc purposes, and may be limited in scope and coverage.

4.2.3 Pole-and-line and gillnet fisheries

Some of the pole-and-line and gillnet vessels make extended trips. The activities of these vessels are generally well documented by logbook programs, which provide information such as locations, dates, catches by species, types of schools, nominal fishing effort, and bait. The logbook records are generally verified for errors in estimates of weight by comparison to landing data, which in most cases indicate that they are reliable. The catches are often categorized by size rather than species, however, and in such cases port sampling must be conducted to establish species and size compositions. Aggregated pole-and-line data are provided to some RFBs at the level of 1-degree areas and half-months or months by national organizations. In some areas, chartered pole-and-line vessels fish under joint-venture agreements. In most of these joint ventures the catches are reported by the coastal States, rather than by the States in which the vessels are registered. The delays in provision of aggregated data may be as long as one year following the year of catch.

The activities of significant portions of the pole-and-line and gillnet fisheries are not monitored by logbook programs. The trips by vessels in this component of the fisheries are generally are of short duration, frequently lasting only one day. With trips of short duration, information on catch and nominal fishing effort by relatively small geographical area and time can be estimated from detailed landings data when these are available. However, some of the boats that participate in these fisheries are beginning to extend the durations and distances from port of their fishing trips, which will make estimations of the catches and nominal fishing effort by area more problematic. The total catches of this component of the fisheries are frequently estimated by market surveys, which do not record data on fishing effort and locations where fishing takes place. This method provides underestimates of the actual catch. There is thus a general lack of information adequate for stock assessment for these fisheries. Unmonitored fisheries are a particular concern in the Indian Ocean, where they account for nearly half the catch of tunas and tuna-like fishes. In contrast, it is estimated that these fisheries account for less than 5 percent of the catches in the eastern Pacific Ocean. Size-frequency data from these fisheries is frequently lacking, although some may be obtained by port-sampling programs at selected ports.

4.2.4 Artisanal and recreational fisheries

Smaller-scale fisheries catching tunas and tuna-like fishes include recreational and subsistence fisheries, and commercial troll and harpoon fisheries. Sample surveys and sociological data indicate that these catches are significant. The levels of catch and the information obtained from these fisheries are quite variable, and reliable estimates of the total landings are often lacking. Nearly all fish caught by artisanal fishermen are retained, but large portions of the billfishes caught by recreational fishermen are tagged and released. In general, no effort data or estimates of fishing power are available for either artisanal or recreation gear.

4.3 Other data

It should also be noted that biological and population parameters are subject to change over time, and, while lacking recent information may necessitate the use of data from past studies, analyses and studies of these parameters should be validated by periodic review and updating of the studies to develop recent information.

4.3.1 Biological data

Data on the biology of tunas and tuna-like fishes, and of species associated with and/or captured in fisheries which harvest tunas, are developed from ad hoc studies conducted by the RFBs, and national organizations. In general, the objectives of these experiments have been well considered, the experimental designs have been good, and the data obtained have been of good quality. These studies have included research on morphology, physiology, behaviour, feeding, reproduction, age and growth, mortality, movements, and stock structure.

Tagging has been conducted by most of the RFBs and some of the national organizations. Fairly extensive conventional tagging programs have been conducted for the major species of tunas in the Atlantic and Pacific Oceans, but there have been no large-scale conventional tagging programs in the Indian Ocean except for those on southern bluefin. The release data for conventional tagging studies are generally considered very good, but some of the recapture data may be unreliable. Fortunately, however, it usually fairly easy to tell which data are reliable and which are not, and the unreliable data can be excluded in the analyses of the data. Data from experiments with conventional tags yield information on movements, stock structure, growth, mortality, and schooling behaviour. Data from sonic tags yield information on physiology and behaviour, and data from archival and pop-up tags yield information on physiology, behaviour, movements, and stock structure.

4.3.2 Environmental data

Environmental data are generally not obtained directly by the RFBs except in conjunction with ad hoc research. A principal exception is observations of hydrographic conditions obtained from logbooks and/or coordinated programs to obtain expendable bathythermograph data from fishing vessels. Applications of results from fisheries oceanography and research that utilize hydrographic and environmental data obtained from research programs, in many cases unrelated to fisheries science, are playing an ever-increasing role in meeting the objectives and mandates of the RFBs.

4.3.3 Observer program data

Observer programs, conducted by RFBs and national organizations, have developed over the last two to three decades. In general, these observer programs were created to monitor activities such as compliance with licensing agreements and restrictions on incidental catches. In addition to providing information required for meeting those objectives, observer programs provide the highest quality and most detailed data on catches and fishing effort, and may also provide environmental data. They also provide essentially the only reliable, detailed information on catches discarded at sea. The coverage of these programs varies greatly from fishery to fishery, however.

4.3.4 Fleet data

Vessel characteristics and operations

The fishing industry continually seeks increased efficiency, incorporating developments in materials, technology, vessel and equipment design, fishing and fish-handling techniques, and fish transport. Most of the RFBs do not compile detailed information on the characteristics of individual vessels, such as fish-carrying capacity or speed, or of their gear, such as net dimensions, use of aircraft, or electronic equipment. Likewise, they do no often compile information on the relative skills of individual vessel captains. In some cases national organizations have limited information on vessels and their gear, but these are not generally supplied to the RFBs, which most frequently have only listings of the vessels and their main characteristics.

Vessel-monitoring and global-positioning systems

Vessel-monitoring systems (VMSs) are increasingly being employed to provide information with which to track the positions of fishing vessels. Most frequently vessels are required to carry VMS equipment as a condition for obtaining fishing permits for territorial waters. The nations in which vessels are registered can impose VMS reporting on vessels operating under its jurisdiction in international waters. A VMS provides a powerful mechanism for positive identification of vessels, fishing grounds, and fishing activity on a current basis. In the absence of VMSs, low-cost global-positioning system (GPS) receivers placed on board a vessel can record its positions at frequent intervals. When the vessel returns to port the records can be examined; the intervals between successive positions provide indications of when the boat had been traveling from one location to another and when it had been moving slowly in a "fishing mode."

4.4 Key problems in improving data quality and coverage

4.4.1 Data aggregation and sharing

In general, aggregated data are submitted to the RFBs by the national organizations, the exceptions being data obtained directly from the fisheries by the IATTC and the SPC. Current arrangements for provision are such that catch and effort data for longline fisheries are provided at the level of 5-degree areas by quarters or months, and data for the surface fisheries data are provided by 1-degree areas by months or half-months. Aggregated data at these levels are quite useful for some research, but data at finer scales are required for some purposes, such as standardization of effort, studies of the effects of local environment on effort or expected species composition in sets, and analyses of catches and effort on FADs. If the fisheries from which imprecise data are collected contribute relatively minor amounts of catch to the total, the problem is less serious than if it comprises a major portion of the total catches. A key point to note is that the lack of data and/or access to data at fine scales may significantly hamper the ability of the RFBs to reduce uncertainty in their findings.

Detailed data that exist should be made available to the RFBs. The status of data provision by parties should be reviewed by each RFB, and, when necessary, action should be taken to ensure that these data are provided in a timely manner.

4.4.2 Catch and landings data

Illegal, unregulated, and unreported (IUU) catches

Information on catches and landings collected by national organizations may be used for determining tax liabilities, as well as stock assessment, so the vessel owners and captains may falsify their records to reduce their taxes. Logbook records might be used as proof that a vessel was fishing in a closed area. This is not a problem when the data are collected by the RFBs, provided that they are not obliged to show the data to the national authorities.

Large-scale fisheries

The existence of IUU catches by vessels fishing for tunas and tuna-like fishes is well known. Few data are available on these catches and the associated fishing effort. It is believed that these catches are made principally by large, high-seas longline vessels, although such activities by purse-seine vessels have recently become a concern. Although the total number of vessels making IUU catches, and the details of their operations, are not known, some of these have been tracked moving from one ocean to another. It is important to not confuse vessels operating under flags of convenience with vessels making IUU catches, as IUU catches may be made by any vessel at any time. It is considered that the impacts of IUU catches are significant.

ICCAT has adopted trade measures against bluefin tuna and swordfish caught illegally. It identified over 300 longline vessels making IUU catches, and this appears to have had a strong impact on the operations of these vessels. Recently some of the other RFBs have begun collecting information on the activities and operations of vessels making IUU catches, and they are now estimating IUU catches from port-sampling data and trade and transshipment records. For example, ICCAT adopted the bluefin statistical document program, requiring that all bluefin imported into the territories of its Contracting Parties be accompanied with documents indicating the registries of the vessels which caught the fish and the areas and time periods of capture. The information in the documents must be validated by government officials. This system revealed significant amounts of catches which had previously not been reported. Trade data are also used to estimate the IUU catches of swordfish, bigeye, and yellowfin. The port-sampling data also include information on the size compositions of the IUU catches. The CCSBT has implemented a trade information scheme, similar to the ICCAT system, to document the catches of southern bluefin tuna.

The most direct solution for reducing the uncertainties created by IUU catches is to reduce the activities of vessels making these catches, as ICCAT has done. FAO is presently in the process of developing an International Plan of Action for reducing IUU fishing activities. Those combined efforts should be maintained, and similar programs should be put into place by all the RFBs to reduce what is one of the largest, if not the largest, source of error in data collection. Recognizing that the following will require legislative and/or administrative action by individual States, consideration should be given to requiring that then States close ports to vessels identified as having made IUU catches.

Small-scale fisheries

The fisheries of some developing countries are not covered by logbook reporting. In some cases the fishermen are illiterate, the boats too open to the elements to permit paperwork, and/or the boats are too numerous to permit collection and processing of logbooks. Several approaches, including stratified sampling schemes and market surveys, are used to estimate fishery statistics. In large countries, the statistics are usually compiled at the district or provincial level. This, together with manual processing and the fact that fishery statistics are generally not limited to tuna fisheries, imposes considerable levels of aggregation, with much of the detail present in the initial data, such as information on species composition, gear, and effort, being lost.

Samples collected for research purposes should be used to estimate species compositions from aggregated data, recognizing that many assumptions must be made in the process, with concurrent increases in uncertainty.

Most statistical systems used in the past for small-scale fisheries have been unable to cope with data collection for a prolonged period. This has led FAO to develop generic multi-lingual computerized solutions (ARTFISH) intended for sample-survey statistical estimation and reporting. The ARTFISH type of system produces estimates of the catch and effort, and provides estimates of variance in relation to the intensity of sampling chosen, but can collect data on the locations of the catches only through interviews with the fishermen. Data at the initial level are kept throughout the processing and archived, which permits the extraction of data that are of specialized interest. Several national and regional agencies are now collaborating in this development, either through complementary systems, or through training and implementation. These efforts should be continued and expanded where possible.

Other problems with catches

Lack of mechanisms for data collection for non-political reasons

Other important components of catch data that may not be reported to the RFBs result from the lack of mechanisms for data collection or from the data not being compiled at sufficient levels of detail.

The lack of data or the failure of national organizations to compile the data in sufficient detail is a serious problem, and the RFBs generally have to make estimates of the magnitude of these unreported or insufficiently-reported catches. These instances are generally associated with developing countries, and include the catches made by artisanal fisheries, for which the amount of unreported catch can be very significant, as is the case in the Indian Ocean. The recreational catches of billfishes, for which complete data are seldom available, may be significant for at least some species. The Code of Conduct for Responsible Fishing states that the nations have an obligation to monitor the fisheries and that, if necessary, assistance should be provided to them for developing the required statistical systems. Sometimes very basic data (e.g. logbook, cannery records, and sales records) which would allow such estimation, exist. In some cases trade data may also prove useful in developing estimates of these unreported catches.

It is strongly recommended that the sources for estimating these catches be identified and that estimates be developed by the RFBs and national organizations. The data collection programs should identify and/or develop sources for, and maintain data for, artisanal, recreational, and subsistence fisheries harvesting tunas and tuna-like fishes.

It is likely that at least some historical data exist within industry records, and that with effort and good communication among scientists of the RFBs and the national organizations and individuals from industry, these data may be made available and analyzed.

It is important that the estimates that are developed be well documented, and that they include information on reliability and variability. In addition, the RFBs should assist in establishing, operating, and coordinating monitoring systems for these fisheries and training scientists and technicians for this work. In instances where legislation obligating reporting of fisheries data to the nations in which the vessels are registered and/or the RFBs does not exist, the RFBs should recommend to the States that they adopt such legislation.

Lack of mechanisms for data collection for political reasons

The lack of mechanisms for data collection may be due to political issues. When this arises due to questions about the status of participants, the data are frequently available, but may not be included in the assessments at the international level due to difficulties in participation of representatives for these fisheries. It may be very difficult to solve the problem of RFBs not having access to fisheries data as a result of political problems.

If no political solutions are offered, a solution at the informal and scientific level must be sought. If this fails, assessments and statements of risk must be based on the estimates of missing data developed by RFBs. To the greatest extent possible these estimates should be based on trade statistics, and national agencies, e.g. customs, particularly those of parties to RFBs, should cooperate closely in these efforts.

Landings (retained captures)

Fish discarded at sea

Stock assessment requires data on the total removal of fish from a stock (retained catch plus fish released alive or discarded dead). In most instances, however, the so-called catch data are actually the retained catches. Attempts have been made to estimate discards from logbook records, which occasionally mention discards, but estimates based on logbook records are not good enough to be included in stock assessments. More recently, improved data required for estimates of releases and discards have been obtained from data collected by observers, and it has been demonstrated that the nature of the discards, such as species, size, and condition, varies significantly with area, season, gear type, etc.

Observer programs should be instituted and/or coverage increased, when necessary, to achieve the required accuracy and precision in the estimates of the total catches, the landings, and the releases and discards by category, as required to reduce the risks to acceptable levels under the Precautionary Approach.

Data on releases and discards that have not been processed or provided to RFBs may exist.

Records from observer programs should be checked for data of this nature, and, if found, the information should be included in future analyses.

Conversion to round weights

Landings are frequently not reported as round weight, and this requires that conversion factors be used to convert the weights provided to round weights. Frequently there is lack of information or misunderstanding as to the nature of the weight provided, and sometimes the conversion factors are crude or incorrect. This causes bias and/or significant imprecision. This problem becomes even more serious when trade data are used for landing estimates. In many cases the conditions of products are not known, and the relationships between the trade weights and the round weights have not been documented outside of industry. This problem is further complicated when individual companies view their raw product utilization rates as confidential data critical to market competition. Furthermore, many persons involved in processing and compiling statistics from fisheries are not aware that these problems exist.

The training and education of samplers and statisticians is essential. A bibliography of existing conversion factors should be developed, and copies of the referenced papers, manuals, and/or articles should be maintained in an archive from which copies are made readily available to the RFBs and national programs. A technical manual detailing the common terms and factors required to correctly determine factors required for conversions, and the various conversion factors for tunas and tuna-like fishes, should be developed.

Fortunately, a conversion factor calculated for a species in a particular area is likely to be usable in another area. Nevertheless, the applicability of conversion factors across large area-time strata should be determined, and when the factors are not sufficiently accurate and precise, additional studies should be undertaken.

The continuing applicability of conversion factors should be determined with small-scale validation studies on a regular basis.

Common expressions for processed conditions must be clearly defined, since an expression may have different meanings among companies, regions, and/or fisheries. Dressed weight, for example, can mean gilled and gutted, or it can mean gilled, gutted, and beheaded.

Catches in numbers or weights

Catches are usually reported as weights, but sometimes, especially in the longline fisheries, they are reported only as numbers of fish. For most analyses the catch in weight is required for stock assessment, but in some cases catches in numbers of fish should be used. In either case there is a need for factors for conversion between individual fish size and weight. These are based on the size compositions of the catches and weight-length or length-weight relationships. The size compositions of the catches will almost certainly vary considerably from area to area and year to year, and the weight-length and length-weight relationships may also vary, perhaps in response to long-term shifts in the environment, which might be introduced by shifts in climate regime. Lack of accuracy in these conversion factors may result in significant errors in the estimated catches, independent of errors that may be introduced by failure to obtain representative size-frequency data (discussed further in a later section).

Biological studies on weight-length and length-weight relationships that include terms for areas and seasons should be conducted. Existing studies may prove valuable in establishing criteria for experimental design. Validation studies should be made on a regular basis.

Misidentified and unclassified species

As previously noted, in industrialized tropical surface fisheries the catches and/or landings of juvenile skipjack, yellowfin, and/or bigeye are usually reported as "skipjack" or "tunas," as the prices of the fish do not differ among species. Juvenile yellowfin and bigeye are quite similar in appearance, so untrained people, even if they want to, cannot distinguish between the two species. At larger sizes, when the prices paid for yellowfin and bigeye are the same, there is a tendency to report both species as yellowfin. When the prices for the two species are different, in some instances, the lower-priced species is over-reported and the higher-priced one is under-reported. Furthermore, if there are regulations that restrict the catches of one species, but not the other, fish of the first species may be reported as the second species. For many years, scientists have been working to correctly determine and report the catches of small fishes by species. In the eastern Atlantic substantial port-sampling programs for catches landed by purse seiners and pole-and-line vessels have been implemented to accomplish this objective. This problem has more recently been recognized in other areas, and corrective measures, including redesigned and more intensive port-sampling programs, have been implemented.

Identification guides that include species recognition of fish of all sizes and both whole and dressed fish should be produced, and observers and port samplers should be well trained in species identification.

Scientists who analyze the data, especially historical data, should be mindful of the problem of species misidentification. When potential bias has been identified, consideration should be given to revising the data to adjust for species misidentification. Because sampling for species composition is labor-intensive and costly, the sampling strategies and experimental designs should be reviewed to determine how to get the required accuracy and precision at the minimum cost. For the areas where sampling for species composition is not being conducted, the catch of juvenile tropical tunas should be carefully documented, and, if problems similar to those discussed are found to exist, corrective measures should be introduced immediately.

Port sampling must also be sufficient to validate species estimates provided by observer programs and to provide checks against falsification of catch reports.

Double reporting

With the introduction of regulations to fisheries harvesting tunas and tuna-like fishes, joint-venture operations have become more common. Most of these operations are charter vessel arrangements between coastal states and vessels from other regions. Under agreements for data reporting developed through the Coordinating Working Party (CWP), the responsibility for reporting of catches and fishing effort rests with the nations in which the vessels are registered. Under certain conditions, however, exceptions to this responsibility are made. Unfortunately, in many instances it is unclear whether a joint venture qualifies for exemption, and the catches are not reported by either nation, or they are reported by both. This problem is likely to become more serious in the near future.

The definitions of responsibility of reporting agreed upon at the CWP should be taken into account. When there is any doubt, the RFB should decide where the responsibility resides.

4.4.3 Effort data

Effort data present even more problems than do catch data. Nominal effort data are collected for most of the large-scale commercial fisheries, but seldom for artisanal fisheries. The fact that the efficiency of the vessels tends to improve over time complicates the interpretation of the data. In general, the changes occur gradually, but sometimes there are sudden and drastic changes, such as the adoption of deep longlines, which are more effective for catching bigeye, and the use of FADs, which increase the catches of skipjack and bigeye by purse-seiners. The problem is complicated by the fact that most of the fisheries for tunas and tuna-like species are directed at more than one species, the fishermen may change their fishing strategies in accordance with changes in the relative abundance of various species, the prices paid for various species, and/or fishing regulations. Scientists must develop techniques to standardize the effort data to compensate for both gradual and sudden changes in fishing efficiency, which requires considerable knowledge of the fisheries and the effect of the environment on the behaviour of the fish, and sophisticated analyses of large amounts of data.

Collection of data on nominal fishing effort should include collection of ancillary data for factors which are known or suspected to have impacts on effective fishing effort, including directly questioning or requesting in logbooks that the targets of fishing effort be clearly identified. Logbooks should also request information on the gear configuration on each set of the gear (for longlines), and for the trip (for other gear types). The RFBs and national organizations should develop and maintain data bases on vessel specifications and gear configurations and operations, preferably on a trip-by-trip basis. If aggregated data are provided to the RFBs by the national organizations, the data provided and the levels of stratification should include detailed gear configuration and targeting data, including the number of FADs involved in the operation and the characteristics of the data provided by the FADs, i.e. location only or information on fish presence and/or oceanographic conditions.

A non-exhaustive list of the kinds of data which should be obtained in association with nominal fishing effort data includes gear specifications, gear construction materials, other auxiliary equipment and gears used, school type, times and positions for starting and ending fishing operations, bait used, use of supply vessels, including transfer dates and amounts, use of oceanographic data obtained directly from satellite or from companies providing current oceanographic and/or weather data, use of aircraft, and use of FADs.

Contribution to the understanding of the efficiency of effort may be gained from discussions with representatives of the fishing industry, but it should be considered that information received may be confidential, due to the perception of those implementing new technologies that the technology gives them a competitive edge. Nevertheless, it is vital that information on developing technologies and vessel characteristics be maintained. The experience and skill of the captain or fishing master is a significant factor in understanding the performance of fishing vessels and effective fishing effort.

The logbook data systems should include records of the captain or fishing master for each trip.

4.4.4 Incidental catch data

As the Precautionary Approach is applied in management, restrictive measures may be applied to tuna fisheries that are based not only on the condition of the stocks of tunas and tuna-like fishes, but also on considerations of species inadvertently caught in association with tunas and tuna-like fishes. Some of these, e.g. sharks, dolphinfish, wahoo, and rainbow runners, are important to artisanal fishermen, others, e.g. billfishes, are important to recreational fishermen, and others, e.g. sea turtles, sea birds, and marine mammals, are important to society, in general. Therefore, the studies of these incidental catches, regardless of whether they are retained or discarded, are of utmost importance to the RFBs and to all nations with fisheries for tunas and tuna-like fishes. Except for the data obtained from observer programs, information on the catches of species other than those that are marketable are minimal. However, data on some species and aggregates of species are available in logbooks, particularly those of longliners. Attempts have been made to estimate the catches of selected species (e.g. Pacific and Atlantic blue shark) by using these logbook records in conjunction with data on the species compositions of the catches of research vessels. These estimated catch data are important sources of information for implementation of the Precautionary Approach.

Emphasis should be given to the collection of data on the incidental catches of fisheries for tunas and tuna-like fishes.

Some RFBs (e.g. ICCAT) require that the Contracting Parties report the incidentalcatches, and some have developed reporting forms for incidental catches (retained and discarded) and are adjusting the data as described above. Logbook and observer programs should be modified, if necessary, to require the reporting of all catches, regardless of category or species. Historical data from logbook and observer programs should be examined for opportunities to estimate the catches of particular species directly or with the use of ancillary information.

4.4.5 Vessel-monitoring and global-positioning systems

Vessel-monitoring systems (VMSs) and global-positioning systems (GPSs) have been discussed in Section

The RFBs should develop requirements that each vessel operating in fisheries which may capture tunas and/or tuna-like fishes carry a VMS or keep records of positions obtained from an onboard GPS, and that the data from this system be available to the RFBs.

4.4.6 Biological data


A well-designed and -executed tagging program can be of benefit in undertaking resource assessments. Execution must include placement of the tags on the fish, collection of the tags and corresponding recapture data, maintenance of the tagging data base, and analysis of the data. A poorly-designed or -executed program, however, is likely to be of little or no value, and may even complicate or compromise concurrent well-designed and -executed programs.

The RFBs and national organizations should actively encourage the implementation of well-designed tagging programs and, when appropriate, seek to improve existing programs.

Size frequencies

For stock evaluation and other biological studies of tunas and tuna-like fishes adequate size-frequency data are essential, because the principal means for estimating the age of individual fishes in the catch are based on measures of size. These age-at-size data are generally validated with studies of hard parts or tagging data, but for some species validation has not yet been accomplished. The port-sampling programs of the RFBs and the national organizations obtaining data from the catches of large purse-seine vessels provides a high level of coverage in terms of areas and times of the catches, even when stratified sampling designs are not used. The reasons for this include the ease of access to the catches during unloading or in the processing facilities and the relative accuracy with which the areas and time periods in which the catches were made can be identified, even when the vessels are in port. In general, the sampling of the catches of longliners is less satisfactory. The reasons for this include the facts that the vessels operate far from their home ports, the trips last for very extended periods, most of the catches are transferred from fishing vessels to freezer vessels before they are landed, and the fish are dressed at sea. Until recently the sampling of the catches of longline vessels has been conducted by the fishermen, which has resulted in low coverage rates and inaccurate measurements, particularly for species that are less abundant in the catches, but which are often the most important from the standpoint of assessment and management. Also, although fishermen are instructed to sample randomly, it is suspected that this has often not been the case. An exception is the Japanese program for southern bluefin, for which every fish is measured. Low coverage and sampling rates result in inaccurate raising factors being used to make estimates of the catches at size. Also, no samples, or inadequate samples, are obtained for some area-time strata, in which cases size-frequency data for other area-time strata are used to approximate the size distributions of the catches in strata for which there are few or no data, which may add significant uncertainty to the analyses. These are serious problems, since often the size-frequency data that are available represent only one in more than 1,000 fish caught in an area-time stratum. Also, for various reasons, data for catches made by artisanal fisheries are scarce. This can be relatively easily solved in some cases by developing and implementing well-designed sampling programs at the major ports where the fish are unloaded. Fortunately, determination of the area-time strata in which the fish were caught is not a problem with artisanal fisheries.

Well-designed, stratified sampling programs for longline catches should be developed and implemented.

It is believed that observer programs, much increased from present levels, would provide these invaluable, required improvements in data quality and quantity. In cases for which observer data are not available, port sampling is the only data source. It is not clear that port sampling alone would provide the required information, due to the difficulties of validated identification of catches to specific area-time strata. The port-sampling programs of ICCAT have provided cross-validation checks and information that identified major errors in national data collection programs. Thus it is considered that port-sampling programs should be designed to cross check and validate data collected by observers. It is expected that technological developments will lead to improvements in collection of size-frequency data in port and at sea. Obtaining size-frequency data from camera records obtained at sea during fishing operations may provide significant improvements in data quality and quantity. Measurements using infrared or laser technologies may also be developed and routinely adopted. Research on new technologies that appear promising for increasing the quantity and quality of size-frequency data should be conducted. Studies on age determination should also be conducted to obtain and/or validate the age-size relationships used in stock assessments.

Sex ratios

Information on the sex ratios of the catches must be collected for species for which sexual dimorphism is known to exist. Obtaining information on sex ratios is presently possible only by examining individual, whole (ungutted) fish, which is difficult to do. Most of the longline-caught fish are dressed at sea, so it is not possible to examine their gonads. Also, at some landing ports and stations industry personnel do not allow samplers to cut the fish open to examine their gonads. Problems caused by inadequate sample sizes and data substitution for missing information in size-frequency data apply as well to these data.

Studies of the distribution of sex ratios and reproductive activity on appropriate area-time scales should be conducted, and gonad indices should be validated for individual species of billfishes. Data required to estimate sex-specific catch at size should be collected at sea by observers and in port by samplers.

4.4.7 Environmental data

As noted previously, environmental data collected for purposes unrelated to fisheries science are playing an ever-increasing role in meeting the objectives and mandates of the RFBs. There is a great concern that the sources of data that have proven most useful in these respects may suddenly become unavailable or unreliable as the programs generating the data are modified or completed.

The sources of hydrographic and environmental data that are identified for use in research and applications by the RFBs and national programs should be cataloged, and the providers should be informed of the uses to which the data are put and of their value in meeting the mandates of the RFBs and the Precautionary Approach. The RFBs and national organizations should coordinate efforts to keep useful environmental data series available and current.

4.4.8 Fleet data

As noted previously, detailed data on the characteristics of individual fishing vessels and their gear are not usually collected by the RFBs or national organizations. Such data are necessary for standardizing fishing effort, and if this is not done increases in vessel efficiency may cause the fishing effort to be under-estimated which could, in turn, cause serious mistakes in stock assessment.

The RFBs and national organizations should collect detailed information on the vessels participating in fisheries that capture tunas and tuna-like fishes and on their equipment.

These data collection programs should to be designed in consultation with scientists, engineers, and fishery technicians. The data should be considered highly confidential. Data collected by national organizations should be made available to the RFBs. Periodic reviews of these data should be conducted to determine whether changes in the vessels, gear, and/or techniques employed are affecting the scientific analyses.

Previous Page Top of Page Next Page