Reducing turtle mortality in shrimp-trawl fisheries in Australia, Kuwait and Iran (Steve Eayrs)

Steve Eayrs
Australian Maritime College
P.O. Box 21
Beaconsfield, Tasmania, 7270 Australia

Abstract

The work achieved to date to prevent the capture of turtles in Australia’s Northern Prawn Fishery, the Iranian Shrimp-trawl Fishery and the Kuwait Shrimp-trawl Fishery is described. Attention is given to the design and operational performance of TEDs used in these studies and the steps taken to extend the results to fishermen. A framework for the development and introduction of these devices into a shrimp fishery is also discussed.

INTRODUCTION

In recent years there has been increased global concern over the impacts of fishing activity on turtle populations, in particular, shrimp-trawl fisheries in tropical waters have come under increased scrutiny due to the capture and drowning of turtles caught in trawl nets. In many countries this scrutiny has resulted in the introduction of stringent legislation requiring the implementation of measures to prevent turtle capture. These measures may include temporal or spatial closure of fishing grounds, particularly where the risk of turtle interaction is high, such as adjacent to turtles nesting sites, although the most commonly favoured option seems to be the development of gear modifications known as turtle excluder devices.

The development of turtle excluder devices (TEDs) probably originated in the United States in the 1980s following concerns over the impact of shrimp trawling on turtle populations in the Gulf of Mexico and southwestern Atlantic shrimp fisheries. News of these developments spread quickly to other countries, particularly where turtle populations were also thought to be threatened by similar fishing activity. In Australia this work commenced in the early 1990s, following the listing of all six species of turtles recorded in shrimp-trawl catches as either threatened or endangered in the Endangered Species Protection Act 1992. This led to major efforts in all of Australia’s tropical shrimp-trawl fisheries to develop suitable TED designs, and currently all require, or are in the process of requiring, the mandatory use of these devices in all trawl nets. In Australia’s Northern Prawn Fishery (NPF) for example, TED-related research has resulted in almost total elimination of turtle catches, with nearly 100 percent of turtles that enter a trawl being excluded by these devices. Other countries with tropical shrimp-trawl fisheries, such as those in Southeast Asia, the Persian Gulf and the Caribbean, have also been addressing the issue of turtle capture. Some of these countries have made substantial progress towards developing and testing suitable devices, and a few have made their use mandatory. However, in many cases progress towards the adoption of these devices has been slow and some of these countries have been impacted by a United States embargo on shrimp imports from fisheries where approved TED designs are not being used. Reasons for the lack of progress include the absence of incentives for fishermen to use and adopt TEDs, inadequate extension programmes, poor collaboration between fishermen, scientists, managers and government officials, inadequate management or poor enforcement of fishery regulations.

This paper presents experiences in reducing sea turtle mortality in shrimp fisheries fisheries in Australia, Kwait and Iran.

Australia - The Northern Prawn Fishery

In the early 1990s fishermen in the Northern Prawn Fishery (NPF) were under considerable pressure to reduce the capture of turtles and other bycatch from trawl operations. During this time the estimated annual mortality of turtles caught in this fishery ranged between 500 and 1 000 individuals (Poiner and Harris, 1996). Records indicated the capture of six turtle species and all were listed as either endangered or vulnerable under Australia’s Endangered Species Protection Act 1992. Under the Act shrimp trawling was listed as a “key threatening process” to turtles, this essentially meaning that trawling could be prohibited if a suitable threat abatement plan to mitigate their capture was not introduced into the fishery. During this period additional pressure on NPF fishermen included the mandatory requirement by the year 2000 for all nets to be fitted with an approved TED (plus a bycatch reduction device (BRD) to reduce the capture of small fish bycatch), and the introduction of the shrimp embargo by the United States on fisheries not using approved turtle excluder devices. Despite the majority of exports from the NPF being destined for Asian markets the desire to maintain access to the US market was strong.

The Northern Prawn Fishery is Australia’s largest and most valuable shrimp fishery, landing approximately 10 000 tonnes of shrimp per year valued at an estimated A$150 million. The fishery covers a large geographical area of some one million square kilometres extending across much of the northern coastline from Queensland to Western Australia. The main target species are Penaeus semisulcatus, P. esculentus, Fenneropenaeus merguiensis, and F. indicus. The fishery presently supports 95 boats with an average length around 24 m. Each boat is permitted to tow two trawls measuring between 11 and 29 m (headline length) each. The mesh size in the main part of the trawl is typically around 63 mm and the codend mesh around 50 mm. The fishery has two fishing seasons; during the first season fishing can occur during the night and day while only night operations are permitted in the second season. Fishing depth is typically less than 40 m.

Attempts to introduce both TEDs and BRDs into this fishery have been extensive and overall very successful. These attempts began in 1995 when 16 TED and BRD combinations were tested in the fishery using a 65-m research vessel. This was followed by testing the best performing devices onboard a 24-m shrimp trawler. The TEDs tested included a Super Shooter TED (Figure 1), a Nordmore grid (Figure 2) and the AusTED. The Super Shooter was originally developed for the Gulf of Mexico and southwestern Atlantic shrimp fisheries. The grid has an oval shape and is constructed from aluminium rod or pipe. The bars of the grid are bent near the escape opening to facilitate the removal of weed that may foul the bars and prevent the entry of shrimp into the codend. In this study bar spacing was 100 mm. A guiding panel ahead of the grid was used to guide animals towards the top of the codend. Large animals are then guided by the bars towards the escape opening in the bottom of the codend. These animals then push aside a netting escape panel (cover) located over the escape opening and are excluded from the trawl net. Small animals exit the guiding panel, pass through the bars and into the codend. The escape cover sits tightly against the escape opening and prevents the escape of small animals. The Nordmore grid used in this study was a hybrid of a larger grid of the same name that originated from Norway. This grid is rectangular in shape and is designed to exclude large animals through an escape opening located on top of the codend. It was also constructed from aluminium but the bars were not bent near the escape opening. Bar spacing was 100 mm. A guiding funnel ahead of the grid guides animals away from the escape opening towards the bottom of the codend. An escape cover was not fitted over the escape opening. The AusTED was a unique design with the grid being constructed from steel wire rope encased in plastic. In this way the grid was flexible thus avoiding claims by fishermen that TEDs were a safety hazard to crew when hauled aloft. Bar spacing was also 100 mm and an escape cover was not fitted. Detailed construction details of these devices are provided in Eayrs and Prado (1998).

Figure 1 The Super Shooter TED has an inclined oval grid to exclude animals through an escape opening in the bottom of the codend. The guiding funnel guides shrimp away from the escape opening.

Figure 2 The Nordmore grid is a rectangular TED designed to exclude animals through an escape opening in the top of the codend.

All three TEDs tested in this study performed well and excluded large animals such as turtles, sharks and stingrays (Table 1). Shrimp loss was a problem with some devices, particularly when a BRD was added to the codend to reduce catches of small fish bycatch. In many instances this loss was due to inexperience and incorrect location of the BRD. A detailed description of the catch results from all devices are reported in Brewer et al., 1998. The Super Shooter was deemed the best performing TED in this study and was subsequently selected for further testing onboard a commercial shrimp trawler under normal operating conditions (Table 2). There were several reasons for conducting the initial tests onboard a research vessel, including greater control over the testing environment and difficulties gaining voluntary involvement from fishermen. This difficulty was mainly linked to fears that the TED would cause massive shrimp loss and pose a safety hazard to the crew during onboard handling of the net.

Table 1. Summary results of TED performance during two research cruises. Catches are compared to a standard trawl. Tow duration in the initial cruise was 30 minutes and in the second cruise it was 120 minutes

# turtle caught ahead of grid and released alive

Table 2. Summary results of TED performance onboard a commercial shrimp trawler under normal operating conditions. Catches are compared to a standard trawl. Fish bycatch was not recorded due to constraints on operating practices. No. of tows in brackets

The completion of the initial study was followed by a major three-year extension programme to help prepare fishermen for the mandatory introduction of these devices in 2000. This programme included the following:

• practical workshops and fishing-gear displays at major fishing ports;

• production of bi-annual newsletters to provide performance results and information updates;

• production of videos including underwater footage of TEDs and BRDs in operation;

• awarding two tied travel grants valued at A$10 000 to fishermen identified by their peers as having made a major contribution towards TED and BRD development;

• providing a range of TEDs and BRDs free of charge to fishermen wishing to test a device onboard their boat;

• the production of a practical guidebook to bycatch reduction, and

• a fishing technologist “boat-hopping” between the fleet over a three-year period.

The practical guidebook was written to provide fishermen and other stakeholders with a guide to the type, design, use and operation of these devices. It also included technical details for their construction and maintenance. The guidebook was small in size, wire-bound and printed on water resistant paper so that it could be safely used on the deck of a trawler. The task of the fishing technologist was to demonstrate the performance of TEDs and BRDs. Armed with a range of devices the technologist would board a boat and initially describe their design and operation. If the skipper was amenable to the idea, one of the devices would be fitted to a trawl to assess operating and catching performance. Depending on the willingness and enthusiasm of the skipper (which was often linked to the ability of the device to retain shrimp as opposed to excluding bycatch) the device would be tested over several days. Catch data was recorded and preliminary assessment of the devices possible (lack of control over the fishing environment precluded more scientific assessment). The detailed assessment of a TED was not the main priority of this study; the priority being to provide the fisherman an opportunity to test a device onboard his own boat and thus gain confidence in its use and operation. A detailed description of this extension programme and associated results are reported in Robins et al., 2000.

As the date for the mandatory introduction of these devices neared, it became necessary to provide training for netmakers in the design, construction and specification of these devices. In addition compliance officers needed to be trained so they could identify approved designs and measure them against approved specifications. These specifications were also promulgated to the fishermen, both formally via official documents provided by the Australian Fisheries Management Authority (AFMA) and informally via a brochure using simple language and diagrams. The specifications were not only designed to reduce bycatch, but to provide fishermen with an opportunity to be innovative and develop new, more effective designs. For example, the specifications for a TED only prescribe limits on bar spacing, the attachment of the grid to the codend, the minimum size of the escape opening and the rigidity of material used to construct the grid. There are no specifications for the use of a guiding panel or funnel ahead or behind the grid, the placement of floats nor the size of the grid. Incorrect use of these modifications will result in excessive shrimp loss and they have little or no impact on turtle escape. The use of these modifications is therefore self-regulating yet simultaneously allows fishermen some opportunity to be innovative and optimize the performance of their devices.

As a result of these efforts the introduction and use of both TEDs and BRDs into this fishery has been very successful. Recent assessment from an observer programme, in which five observers on 24 boats recorded catches from over 1 600 trawl shots, indicated that turtle exclusion rates are 99 percent, with the rare capture of a turtle as the net is hauled (Table 3). Random at-sea inspection by compliance officers have found compliance rates nearing 100 percent and fishermen are well aware of the financial implications of poorly operated and maintained TEDs. The successful adoption of these devices by NPF fishermen and their demonstrable commitment towards reducing the threat of trawling on turtle populations, resulted in the lifting of the US embargo soon after TEDs became mandatory in this fishery. The introduction of TEDs has also had wider ecological benefits, including exclusion rates of large animals such as sharks and stingrays nearing 100 percent and to a lesser degree some large fish species. Damage rates to shrimp by these large animals has also dropped markedly, hence overall catch value has improved. This value was estimated at A$750 per week to each boat by one study, but this is dependant upon catch volume, trawl duration, size and type of large animal caught.

Table 3. Summary results from observer programme in Northern Prawn Fishery in 2002 comparing the performance of a net with a TED fitted against a net without a TED. Shark and stingray data include individuals of all lengths. No. of tows = 1 612

Kuwait shrimp-trawl fishery

Attempts to reduce catches of turtles in the Kuwait shrimp-trawl fishery commenced in earnest in 2003 when the Australian Maritime College and the Kuwait Institute for Scientific Research (KISR) began a one-year collaborative bycatch reduction programme that included practical training workshops and at-sea tests of several TED and BRD designs.

The at-sea tests occurred on the commercial fishing grounds adjacent Kuwait Bay. This fishery extends along much of the Kuwait coast and is typical of many tropical shrimp fisheries worldwide, being characterized by a wide variety of bycatch species and variable but often high bycatch to shrimp ratios. Bycatch to shrimp ratios up to 15:1 (by weight) have been recorded in this fishery (Ye, Alsaffar and Mohammed, 2000). The main bycatch species in this fishery are juvenile fish or species of low commercial value. Catches of turtles do not appear to have been documented, but the US embargo is in place and has affected shrimp exports.

The Kuwait shrimp-trawl fishery occurs along much of the Kuwaiti coastline in depths to about 40 m. The main shrimp species caught are Penaeus semisulcatus, Metapenaeus affinis and Parapenaeopsis stylifera. The fishery supports two types of fishing vessels: large steel trawlers measuring around 25 m and smaller timber or fibreglass dhows. The fishing gear used by the steel trawlers usually comprises of two flat, semi-balloon or balloon shrimp trawls measuring around 27 m (headline length) towed simultaneously in a double rig configuration. The mesh size in the main part of the trawl (body) is typically 51 mm and codend mesh size is typically 45 mm. The smaller vessels usually tow a single net of similar design. The headline length of these nets are usually around 32 m and mesh size in the main part of the trawl (body) is typically 45 mm. Codend mesh size is typically 38 mm. Fishing occurs both at night-time and day-time, although until recently regional conflict necessitated a ban on night trawling.

The collaborative programme included pre-sea training workshops and three five-day periods at sea testing the performance of one TED and several BRDs. During the workshops KISR staff had the opportunity to assist in construction of fishing gear in preparation of the at-sea tests, and in this way gain knowledge of the design, rigging and operation of these devices. This was then complemented by participation in the at-sea tests. These tests occurred on both steel trawlers and the smaller dhows. The specific location of the tests was selected by the skipper of the vessel as per normal commercial practice. The catching performance of these devices was compared against a standard trawl.

The TED used in these tests was constructed by Popeye Netmaking (Cairns, Australia) and is known in the local industry as the Popeye TED. To suit the larger trawl nets used in the fishery the overall size of the TED was increased to match the circumference of the codend, otherwise, no other changes were made to this TED. The grid used in the construction of the TED was oval in shape, measuring 1.8 m high and approximately 4.8 m in circumference, and was designed with a bar spacing of 120 mm. The entire grid was constructed from stainless steel bars to provide adequate strength.

The performance of the TED and other devices tested in this study have not been assessed in detail as the final at-sea tests were only recently completed. However, the overall performance of the TED was satisfactory; no turtles were caught in the net fitted with this device and shrimp catches were little different to the standard net (Table 4). The standard net did catch one turtle (during the second trip) however after a period of recovery onboard the boat it was released alive into the water. The TED did not pose a safety hazard to the crew and they very quickly developed the skill to safely and correctly handle this device.

Table 4. Summary results of TED performance onboard commercial shrimp trawler in the Kuwait shrimp-trawl fishery. Catches are basket numbers and are compared to a standard trawl. No. of tows in brackets

At this stage little additional progress has been made towards the removal of the US embargo.

Iranian shrimp-trawl fishery

In October 1997, the Food and Agriculture Organization of the United Nations (FAO), the Ministry of Jehad - E-Sazandegi, Fisheries Company of Iran (Shilat) and the AMC conducted a bycatch reduction workshop in Bandar-Abbas, Iran. The objectives of this workshop were to test a TED (the NAFTED) and several bycatch reduction devices, and demonstrate to fishery officers from the Gulf region the installation, operation and performance of these devices under commercial fishing conditions.

The at-sea tests occurred on commercial fishing grounds that are part of the Iranian shrimp-trawl fishery. This fishery extends along much of the Iranian coast and is typical of many tropical shrimp fisheries worldwide with regard to the variety of bycatch species caught and variable but often high bycatch to shrimp ratios. The main species caught in this fishery are Penaeus merguiensis, P. semisulcatus, Metapenaeus affinis and Parapenaeopsis stylifera. There was little information available regarding the impact of shrimp-trawling on turtle populations and despite the export of some shrimp from this fishery there was little concern for the US embargo. The main bycatch species in this fishery was juvenile fish and fishing activity occurred mainly in daylight hours.

The fishery supports three types of fishing vessels: small fibreglass trawlers measuring about 7 m in length dominate the fishery and number approximately 1 500. The largest vessels are constructed from steel and measure between 20 and 25 m and they total 39 in number. Approximately 850 timber dhows also operate in this fishery averaging around 16 m in length. The fishing gear used by the steel trawlers usually comprises of two flat, semi-balloon or balloon shrimp trawls measuring 29.3 m (headline length) towed simultaneously in a double rig configuration. The smaller vessels usually tow a single net of similar design. The mesh size in the main part of the trawl (body) measured 40 mm while codend mesh size measured 30 mm.

The NAFTED was initially designed by the Australian Maritime College (AMC) to exclude large animals such as turtles, sharks and stingrays from trawl operations in the NPF. This device was similar to the Nordmore grid tested in Australia but with bent bars near an escape opening located at the top of the codend. The grid was constructed from aluminium and grid angle was 45 degrees. This device is typically designed with a bar spacing of 100 mm, however, in this instance a bar spacing of 60 mm was used to improve exclusion rates of smaller animals. A panel of netting located ahead of the grid was used to guide all animals to the bottom of the codend. Large animals are then guided by the grid towards an escape opening on top of the codend while smaller animals pass through the bars and into the codend. No escape cover was fitted to the TED because it was feared this would reduce fish exclusion rates.

Due to the structure of the testing programme, catches from a trawl fitted with the NAFTED was compared against a trawl with a square-mesh window BRD. The NAFTED reduced bycatch by 34 percent and maintained shrimp catches (Table 5). The reduction in commercial fish was 50 percent and it is likely these fish were excluded through the large escape opening. The catch of large stingrays was reduced by 95 percent.

Table 5. Catch comparison (number of full baskets) between a net fitted with a NAFTED and a net with a square-mesh window and fish cone stimulator attached. No. of tows = 3. The numbers in brackets indicate the number of large stingrays caught in each net

The workshop participants and crew were most impressed with the performance of the NAFTED, particularly as the hazardous operation of removing stingrays from the catch was virtually eliminated and sorting times were substantially reduced. The loss of commercial fish was of some concern but this was offset by the benefits of stingray exclusion. The NAFTED posed no handling problems for the crew although it did require correct deployment from the trawler. This simply entailed making sure the codend was not twisted and the grid was sitting upright in the water prior to shooting the gear away. At no stage did the large aluminium grid pose a safety hazard to the crew.

Lesson learnt

The development and testing of TEDs in these shrimp-trawl fisheries has provided some valuable lessons useful for consideration to those about to embark on similar work.

Extension and enforcement

The successful introduction of TEDs into the Northern Prawn Fishery is in part due to the willingness of fishermen to adopt new technology. Although in most instances they initially regarded these devices with suspicion, the initiatives used to extend the results of the testing programmes on both research and commercial boats played a major role in this success. In particular the use of underwater cameras to film TEDs and associated shrimp and fish behaviour was tremendously successful. In many instances fishermen had never observed their fishing gear in operation, and the attraction of observing this video was strong and useful for stimulating new ideas and demonstrating the importance of correct rigging and operation. Gear workshops at the fishing ports were also very successful and provided fishermen an opportunity to ask questions about the design and operation of these devices. The provision of literature for fishermen to take to sea was useful because it allowed them to learn about recent developments at their leisure. This approach provided fishermen the opportunity to test and learn about TEDs before their mandatory introduction, thus giving them time to become suitably prepared, although a few fishermen made no such attempts and paid the price accordingly.

It is unclear how effective this extension work would have been in an environment whereby checking of gear by compliance officers was inadequate. In the NPF up to 80 percent of the fleet is boarded in any one year and the penalties for infringements can be severe. This serves as a suitable inducement to comply with the specifications. However, it is probable that the fear of shrimp loss and associated income in an export-orientated fishery served as an even greater inducement to comply with the specifications and optimize TED performance.

Selecting the correct TED

Probably the most important aspect to TED selection is the decision to orientate the grid either upwards to exclude large animals through the top of the codend or downwards to exclude these animals through the bottom of the codend. While grids orientated in either direction seem equally effective in excluding turtles, orientation plays a major role in the ability of the TED to exclude rocks, sponges or heavy debris from the trawl. In locations where these items are present a bottom excluding TED is required so that they can roll towards the escape opening and be excluded. In locations where the seabed is largely devoid of these items, an upward excluding TED can be used.

The overall size of the grid is arguably the next most important aspect of TED selection because it influences the ability of the escape cover to sit tightly over the escape opening. The size of the grid should be sufficiently large to slightly distort and increase the circumference of the codend. This in turn prevents shrimp loss by allowing the trailing edge of the escape cover to make good contact with codend netting adjacent the escape opening.

Grid angle also influences the ability of the TED to exclude turtles and other large animals. Typically a grid angle of 45-55 degrees is required for both upward and downward orientated grids to rapidly exclude turtles and other large animals from the trawl. Such angles also reduce the likelihood of shrimp loss. If grid angle is excessive, irrespective of grid orientation, the rapid exclusion of turtles is delayed, thus increasing their potential to be drowned. Excessive or high grid angle may also result in blockage by rocks, sponges etc., and hamper the rapid passage of shrimp into the codend. They may also partially push the escape opening aside and cause massive shrimp loss. If grid angle is inadequate the escape cover may not sit tightly over the escape opening and shrimp loss is likely. The shape of the escape opening may also become distorted. Low grid angles do not seem to have a negative impact on the exclusion of turtles from the trawl.

The distance between the bars of the grid is important because it influences exclusion rates of small or juvenile turtles and the passage of shrimp into the codend. In the United States and Australian fisheries bar spacing is typically 100-120 mm. Greater distances between bars is thought to increase the potential for the head or flippers of large turtles to become fouled in the grid. Lesser distances will have little effect on turtle exclusion and may increase escape rates of fish and other animals. Increased shrimp loss may also result from smaller bar spaces.

Other factors to consider when selecting a TED is the use of guiding panels or funnels of netting ahead of the grid. These are usually constructed from netting material and are designed to guide shrimp away from the escape opening. In Australia some shrimp fishermen have decided to do away with these funnels with seemingly little impact on shrimp catches. Another factor is the use of a netting (escape) cover over the escape opening. These are always used in bottom excluding grids to prevent shrimp near the bottom of the codend from escaping. With a top excluding grid they do not seem to be as important - studies in Australia and Iran indicated that the NAFTED lost few shrimp - but many fishermen do not like to operate a trawl with the escape opening exposed. Grids are typically constructed from aluminium or stainless steel rod or tubing, the latter material often being preferred in large grids for additional strength. Several floats are usually always attached to the TED irrespective of grid orientation. Floats serve to provide buoyancy and stability, particularly to large, heavy grids. They can also be useful when the gear is at the sea surface to indicate the orientation of the grid prior to deployment.

Over-tuning of TEDs

Over-tuning of TEDs is a term coined to describe excessive or inappropriate modifications to the TED to reduce shrimp loss. This loss typically arises from poor design, rigging or maintenance of a TED, or the poor selection of a TED for a particular fishing ground. Examples include incorrect grid angle or orientation, and the use of small grids in grounds with high sponge numbers. Many fishermen inexperienced in the use of TEDs are then tempted to modify their device to ameliorate the problem. The temptation is particularly high if two nets are towed simultaneously and the shrimp catch from one net is substantially smaller than that of the other - this being colloquially known in Australia as being “TED-ed” because the TED is blamed for the loss. It is very common for these fishermen to attempt to solve the problem by making the escape cover sit tighter over the escape opening. This can be achieved by the use of excessively long escape covers, heavy weights attached to the trailing edge of the cover (on a top opening TED), or the use of floats on the cover (of a bottom opening TED). However, in many instances the unfortunate outcome of these modifications is to increase the frequency of being TED-ed because large animals now struggle to push aside the escape opening. This in turn increases the period that the cover is not seated tightly over the opening thus providing shrimp a greater opportunity to escape. The answer to this problem often seems counter-intuitive; that is, to loosen the escape cover to increase the speed and ease with which turtles and other large animals can escape. A well-designed and tuned TED will rapidly exclude these animals from the trawl, thereby reducing blockage of the grid or escape opening. The escape cover will then be seated against the codend for longer periods and the opportunity for shrimp to escape will be minimized.

Testing protocol

Following the mandatory introduction of these devices in the NPF it became clear that a testing protocol was required that allowed innovative fishermen the opportunity to test and develop their own TEDs (and BRDs). A TED and BRD committee, with membership including a fisherman, a fishing technologist and the fishery manager, was established by AFMA to develop the protocol and monitor progress. This protocol was designed to provide fishermen with a quick and simple means of developing new devices and measuring their performance whilst providing sufficient rigour to demonstrate the achievement of bycatch reduction targets and satisfy the concerns of other stakeholders. Underpinning the design of the testing protocol was the desire to encourage and foster the development of innovative bycatch reduction devices and to provide flexible testing requirements that accommodate the difficulties of testing these devices onboard commercial fishing boats.

The TED and BRD testing protocol has three main assessment phases: an initial assessment phase, a visual assessment phase and an at-sea testing phase (Figure 3). The initial assessment phase involves the operator providing members of the TED and BRD committee for review and comment written details of the device to be tested, including technical details and specifications, and a description of how the device will reduce bycatch. Approval (or otherwise) of the device is usually made within 48 hours of receipt of the application. The criteria for initial approval of the device are based on the expected likelihood of reducing bycatch to a level at least commensurate with approved devices and the potential threat the device may pose to endangered or threatened species. The judgement of this device and ability to meet these criteria relies almost exclusively on the collective experience of the three committee members (although additional advice may occasionally be sought). The recommendation for the next phase of assessment may include a request for visual assessment of the device and/or at-sea testing.

The visual assessment phase involves one or more of the committee members either physically viewing the device or briefly (~1 day) testing it in a flume tank at the Australian Maritime College. The aim of this phase is to gain a better appreciation of the device, its operation and likelihood of achieving the claimed reduction in bycatch. This phase is only expected to be required when an entirely new or complex bycatch reduction device is developed or if the details submitted in the initial assessment phase are inadequate or inconclusive. In most instances a device will not require this assessment phase. The criteria for visual approval of the device is similar to that of the initial assessment phase, but with the possible additional expectation that the shape and orientation of the device in the flume tank is as claimed. The cost of transportation and flume tank testing is borne by AFMA.

1. Initial assessment of TED or BRD design, including rigging details, specifications and location in the codend.

2. Visual assessment of actual device fitted to a codend.

3. At-sea assessment initially by fisher, then by TED/BRD expert if required.

Figure 3: The TED and BRD testing protocol used by fishermen in Australia’s Northern Prawn Fishery to develop and test their own devices.

The final assessment of the device involves at-sea testing onboard a commercial fishing boat. The fisherman is provided with a scientific permit from AFMA to test the device under normal commercial fishing conditions, but which may include specific requirements related to the location fished (e.g. region of relatively high bycatch density), data collection methodology and the period of assessment. In the case of a new TED being tested a permit is initially granted for four to six weeks. During this time the fisherman operates as per normal commercial practice but if two turtles are caught in the net with the new device the permit is revoked and the device fails the assessment. In this way an allowance is made for the exceptional or fluky capture of a turtle, such as when a turtle has had insufficient time to escape from a TED during haul-back (steps are currently being taken to reduce this figure to zero).

Following completion of the initial at-sea testing phase the fisherman is required to provide a report with shot by shot catch details. If a new or modified TED is being tested then details of turtle catches in the net fitted with this device is required. Following receipt of this report, the members of the committee may then decide to either approve the device or request further testing and assessment. In most instances where a TED is tested a trained observer is required to board the boat and independently assess the performance of the device. The observer may spend several weeks onboard the boat and if two turtles are caught during this time the device fails the assessment. While onboard the observer will collect relevant catch data and provide a detailed report analysing the performance of the device over the test period. Importantly, the costs of the assessment by an observer are borne by AFMA.

The members of the committee then review the observer’s report and decide whether the device is acceptable for inclusion as an approved TED. The absence of turtle catch in the net fitted with the TED may simply reflect the absence of these animals in the area fished. The TED may therefore be given approval subject to further monitoring by the fisherman and/or by another observer at a later date if in the region. The approval of this device is then indicated to AFMA, other NPF fishermen and stakeholders.

TEDs in Iran and Kuwait

The TEDs tested in Iran and Kuwait originated in Australia and they successfully excluded large animals from the trawl with little shrimp loss. This suggests that TEDs originating from one country can be a suitable starting point for other countries about to embark on similar research. The testing framework applied to introduce these devices in each country has proven successful and can be confidently applied with little alteration to other fisheries (Figure 4). As fishermen become more adept at using these devices they can begin to modify or develop their own devices to better suit their fishing operation. In both Iran and Kuwait catch rates of shrimp are often highly variable due to their schooling behaviour, and this makes assessment of shrimp loss difficult unless an extensive testing programme is undertaken. In both countries fishermen were pleased with the exclusion of hazardous large animals and reduced catch sorting times. Increased leisure time is a good incentive for fishermen to adopt these devices, and there were indications that a small loss of shrimp was an acceptable price to pay. This was no doubt driven by the fact that the income of these fishermen was not totally derived from shrimp alone.

However, the at-sea testing of TEDs in these countries is only the first step in the process of successfully introducing these devices to the satisfaction of all stakeholders. Research organizations alone cannot follow the entire process to its logical conclusion, and government involvement is usually required to provide relevant funding for extension programmes, the training of fishermen, netmakers and compliance officers. It is unlikely that fishermen will fully develop, test and adopt these devices on a voluntary basis alone, particularly in a competitive environment where the testing of innovative fishing gear is fraught with risk of shrimp loss. The removal of the US embargo requires a detailed data-set of trawl and turtle interactions. Ideally this needs to be from both fishery dependent and independent sources to verify that fishermen are reducing turtle capture as claimed. Such efforts take substantial time and effort, and requires a major commitment by fishery managers and others. There also needs to be an extensive monitoring programme to confirm the uptake and effective compliance of these devices by fishermen. Such programmes are expensive and usually require funding and assistance from governments or other sources. Since the workshop in Iran, for example, there seems to have been little documented progress in TED development in the region. Anecdotal reports suggest that this is partly due to lack of political will, indifference by various stakeholders and insufficient resources to fund further research. Clearly these issues need to be addressed to effectively protect turtle populations.

Figure 4: A framework for the development of an effective TED (or BRD) into a shrimp-trawl fishery (adapted from Kennelly, 1996)

Future research

The use and operation of TEDs in a shrimp-trawl fishery is only the first step towards protecting turtle populations. There is wide-spread consensus that contact with a TED and subsequent exclusion does little harm to these animals providing that the TED is well tuned and escape occurs quickly. However, it is not known if there are any long-term health repercussions associated with repeated exclusion of an individual over a short time period. This might occur in a fishery where several boats using TEDs are operating in close proximity to one another. It is also not known to what extent escape from a trawl can be delayed by a poorly designed or tuned TED without causing mortality. Further work is required to evaluate the effect of such incidents on turtle health.

TEDs have also demonstrated good capacity to exclude other animals from the trawl, including large sharks, stingrays and fish. In many instances the exclusion of these animals can also approach 100 percent. Moreover, in some instances these devices can be modified to exclude smaller animals using a narrower bar spacing, although the risks of shrimp loss are increased. The use of underwater cameras can be a valuable tool to observe the behaviour of animals as they approach a TED, and this information help identify other species suitable for exclusion from the trawl. Where the exclusion of these animals is a desirable outcome, further work is required to modify TEDs to optimize exclusion rates while maintaining catches of shrimp.

In many countries there is a need to identify the extent of turtle populations, including timing and location of migrations be they diel or otherwise. This information is useful because it can help predict times when turtles are most vulnerable to capture from shrimp trawling. The potential introduction of temporal or spatial closures can then be used as an additional (or alternative) technique to prevent turtle mortality in trawl nets. Those countries attempting to overcome the US embargo will also need to collect additional information, including demonstrable uptake and use of TEDs by fishermen, proven ability to exclude turtles and a demonstrable means of ensuring fishermen are abiding by the TED specifications and associated regulations.

CONCLUSION

Over the past few decades major strides have been made in many shrimp-trawl fisheries to prevent the capture and mortality of turtles. A well-designed and maintained TED can exclude almost 100 percent of turtles that enter the trawl, with individuals usually only being caught immediately prior to hauling of the gear. In such instances the turtle can be released safely with little ill-effect. Other large animals can also be excluded at similar rates, including sharks, stingrays, sponges and fish.

The at-sea testing of these devices is a step in the right direction, however, if not supported by a well-funded extension programme to provide fishermen with important design, operational and performance information then the effective uptake and adoption of these devices is hampered. Moreover, compliance rates are likely to be less than would otherwise be achieved with such a programme, thus increasing the threat of trawling to turtle populations. Every effort should be made to encourage the voluntary testing and use of these devices and addressing the needs and concerns of fishermen is an essential requirement towards the successful uptake of these devices. This work should also be coupled with a strong monitoring and compliance programme to ensure all fishermen are complying with TED specifications and related regulations.

REFERENCES

Brewer D., Rawlinson, N., Eayrs, S. & Burridge, C. 1998. An assessment of Bycatch Reduction Devices in a tropical Australian prawn trawl fishery. Fisheries Research, 720: 1-21.

Eayrs, S. & Prado, J. 1998. Tests, demonstrations and training for the utilization of by-catch reduction devices in shrimp trawling fisheries in the Gulfs. Bandar-Abbas, Iran, 12-16 October 1997. FAO Fisheries Circular No. 936. Rome, FAO. 19p.

Kennelly, S. 1996. A framework for solving bycatch problems: examples from New South Wales, Australia, the eastern Pacific and the northwest Atlantic. In World Fisheries Congress. Developing and sustaining world fisheries resources: the state of science and management, pp. 544-550. 2^nd World Fisheries Congress Proceedings.

Poiner, I.R. & Harris, A.N.M. 1996. The incidental capture, direct mortality and delayed mortality of sea turtles in Australia’s Northern Prawn Fishery. Aust. J. Mar. Freshwater Res., 41: 97-110.

Robins, J., Eayrs, S., Campbell, M., Day, G. & McGilvray, J. 2000. Commercialisation of bycatch reduction strategies and devices in northern Australian prawn trawl fisheries. FRDC Project 96/254 final report. 40p.

Ye, Y., Alsaffar, A.H., & Mohammed, H.M.A. 2000. Bycatch and discards of the Kuwait shrimp fishery. Fisheries Research, 45: 9-19.