The Problem Of Incidental Catches Of Sharks And Rays, Its Likely Consequences, And Some Possible Solutions

		Shark Conference 2000 Online Documents Honolulu, Hawaii February 21-24
Sponsored By: The Barbara Delano Foundation The Homeland Foundation The David & Lucile Packard Foundation The AVINA Foundation	Presented By: WildAid Hawaii Audubon Society Pacific Fisheries Coalition
THE PROBLEM OF INCIDENTAL CATCHES OF SHARKS AND RAYS, ITS LIKELY CONSEQUENCES, AND SOME POSSIBLE SOLUTIONS
Ramón Bonfil Fisheries Centre University of British Columbia 2204 Main Hall Vancouver, BC V6T 1Z4 Canada Email: [email protected] Introduction The incidental catch of sharks in fisheries directed to other species is a major source of mortality that has raised considerable concern over the last decade (Bonfil, 1994; Camhi, 1999). In this paper I provide a brief overview of the incidental catches or bycatches of sharks in fisheries, the problems that arise from these bycatches, and some possible solutions to them. The analysis is here extended to the skates and rays, which for most purposes are sharks with a different morphology. Skates and rays share with sharks not only the same type of biology but also a similar fate in fisheries as all of them have until recently been perceived as low-value fish. What is bycatch? There are several definitions of what bycatch or incidental catch is. On one hand, it has been proposed that bycatch is "that portion of the catch returned to the sea as a result of economic, legal or personal consideration plus the retained catch of non-target species" (McCaughran 1992). Another similar definition of bycatches points out that these are "animals other than the target species which are unmarketable because they are too small or for some other reason" (Alverson et al. 1994). Hall (1996) proposed defining bycatch as "that portion of the capture that is discarded at sea dead (or injured to an extent that death is the most likely outcome) because it has little or no economic value or because its retention is prohibited by law". In simpler words, these definitions say respectively that bycatches are: All discards plus retained non-target species Discards of non-target species All dead discards However, we have limited understanding about the survival of discarded sharks and rays in most fisheries. Additionally, retained catches of non-target species are seldom regulated in a fishery. Thus, for this paper I loosely define bycatches as all catches of sharks and rays in fisheries targeting other species. Whether they are discarded or retained, whether they survive the discard or not, or whether they are totally utilised or discarded after finning, is not important for this definition. What this definition is trying to comprise is that sharks and rays are being caught and usually killed in most cases without any kind of control, while supposedly fishing for other species, being these shrimps, flatfishes, tunas, or any others. Factors affecting bycatch There are four major factors defining the level of bycatch in a given fishery. These are: 1) The fishing grounds or type of environment. Whether fishing takes place in the sea floor or in mid water, in the temperate seas or tropical areas. 2) The fishing gear. If this is passive or active, a net, a hook or a pot. 3) The behaviour of marine animals. If they are predators or not, their association with target species, their avoidance behaviour, their food habits, etc. 4) The seasonality of movements of the bycatch species. Many species go through migrations that make them more or less available in the fishing grounds where target species are caught, and this sometimes has an important role in the level of bycatches. Who is who in the contest for the title of 'nastiest fishing gear'? There seems to be some confusion among the general public about what types of fishing gear are 'cleaner' for fishing and which are more indiscriminate on the species they catch. Table 1 lists the most common types of fishing gear on a global basis, ordered from the most indiscriminate to the relatively cleanest one. There is no doubt that the most indiscriminate of fishing gears are the towed nets. In their original mode, trawl and seine nets are the kinds of fishing gear that account for the highest number and amount of non-target species caught incidentally. This is because towed nets are meant to capture all they encounter on their path, and when used in the sea floor, they encounter a high number of non-target species. Bycatch rates from these gears range from 60% of the target fish as in some temperate groundfish fisheries, to over 10 times the amount of target species in some tropical shrimp fisheries. Gillnets of all kinds are probably the second most indiscriminate gear. They are designed to entangle or trap most of the animals that come across the wall of net. However, because they are passive gears and because of the wide range of mesh sizes in which they are built, they can be relatively more selective than trawl nets. Typical bycatch rates for pelagic gillnets before they were banned from the high seas were of about 30-40% of the total catch in numbers (Mckinnell and Seki,1998). Longlines and in general hook and line gears, are among some of the most selective fishing devices. Not only hook size plays a role on their ability to discriminate, but also the type of bait used, and the requirement that the animal approaches and bites or otherwise gets accidentally hooked. This is not to say that longlines are totally 'clean' gear. Within a given hook size, longlines are not very selective on the type of predatory fish that they get (they are known to catch even scavenging seabirds). However, typical levels of bycatch in pelagic longline fisheries are from 20% of the total catch to about 4 times the target catch in some swordfish fisheries, although they are typically 60% of the total catch in tuna longlines (Da Silva and Pereira, 1998; Hazin et al. 1998). Purse seines have some of the lowest level of bycatches mainly because they are highly active gears that are usually set upon schools of the target species. At least for some non-target species such as dolphins and large animals like whales and whale sharks, there is the possibility of unharmed release. The bycatch rate in this fisheries ranges from 2-10% of the total catch (Seret et al. 2000, Deelner and Hall 2000). It is clear that although there are no totally clean fishing gears despite our claim to technological know-it-all, longlines stand among the least indiscriminate when compared with the rest of the most common fishing gears. As we will see below, the problem with bycatches in longline fisheries is not so much their selectivity but the size of the fisheries. Main sources of shark and ray bycatches Figure 1 shows the main fisheries, at the global scale, with large bycatches of sharks and rays. In coastal areas, there is no doubt that bottom trawl fisheries cause the largest bycatches of sharks and rays. These fisheries mainly cause skates and ray bycatches because these species live on the ocean floor but some specific fisheries take also significant amounts of bottom-dwelling sharks. Due to the diversity of trawl fisheries worldwide and the fragmentation of the information it is not possible to estimate the total level of ray and shark bycatches in these fisheries at the world level. However, given their scale and typical high rates of bycatch it is likely that the total figure is in the range of a few hundred thousand tonnes of skates, rays and sharks annually. The groundfish fisheries of the Hecate Strait, in western Canada discard about 640 kg of skates and rays for every tonne of target fish they keep (Leaman, 1994), while shrimp fisheries discarded some 2,800 t of sharks annually in the northern Gulf of Mexico (NMFS, 1993, Anon., 1995). Bycatch of sharks in deep-water fisheries are typically of 10-50% of the target species catch. Pelagic trawls have also some shark bycatches, but these are poorly documented, and there are no specific numbers available. Shark and ray bycatches are likely to be small as these fisheries are not as widespread as those using bottom trawls. Nevertheless, it is known that spiny dogfish and salmon sharks are common bycatch of pelagic trawls in the North Pacific. Coastal gillnet fisheries take also bycatches of sharks and rays, but at least in most of the tropical countries gillnet fisheries are typically multispecific and normally target a mix of species. In the context of the definition used here for bycatches, these fisheries are probably of lesser importance than trawl fisheries and most of their catch should be already accounted for in official fishery statistics. An important exception to this will be cases such as basking sharks and whale sharks caught accidentally in coastal areas. High seas fisheries produce among the largest bycatches of sharks at the world level. The most important bycatches are in longline fisheries for tunas and billfish. The major reason of the enormous bycatch of sharks and rays in high seas longline fisheries is the large amount of hooks that are set in the sea by these fisheries. Tuna and billfish fisheries cover most of the oceans of the world, they occur in most of the Pacific, in the Indian and also in the Atlantic Ocean (Fig 2). With so much effort being applied it is not surprising that they account as the major source of shark bycatches. These fisheries typically take blue sharks as the main shark bycatch but when in tropical seas or closer to landmasses they take also important numbers of silky, oceanic whitetip, mako, porbeagle, hammerhead, and other sharks. As shown in table 2, it has been estimated that the total amount of sharks caught in longline fisheries in the world is similar for the three major oceans where these fisheries take place (Bonfil, 1994). Alternative estimates for the Pacific Ocean (Stevens, 1997) suggest a somewhat higher number of sharks taken as bycatch. However, there are some errors in those calculations, which inflated the results (Stevens, pers. comm. Feb 2000). The most reliable estimates suggest that a total of about 232,000 tonnes of sharks and rays (manly blue sharks) were caught incidentally in longline fisheries for tunas and billfishes (Bonfil, 1994). At the time these estimates were made it was uncertain what the fate of the sharks were, but throughout the 1990s there have been numerous reports of increasing rate of finning of sharks in longline fisheries in all oceans. So we can pretty much assume that all these estimates actually translate into deaths. Drift gillnet fisheries use to be the second major source of shark bycatch in the high seas. Available estimates from the early 1990s indicate that driftnet fisheries accounted for nearly 30,000 tonnes of shark and ray bycatches annually (Bonfil, 1994). However, high-seas driftnet fisheries have been banned since 1992. The third most important source of shark and ray bycatches are the tuna purse-seine fisheries of tropical seas (Fig. 3). The estimate presented here is a combination of Stevens' (1997) estimates for the Pacific Ocean with others available for the Atlantic and the Indian Ocean. This is almost surely an underestimate as the method used for the Atlantic and Indian Ocean was very crude compared to that of Stevens. In all, these high seas fisheries could account for almost 300,000 tonnes of incidentally caught sharks and rays (Tab. 3). This figure could be closer to ½ a million tonnes if we consider the estimate of Stevens for the Pacific Ocean longline fisheries, but we know these are overestimated. This amount of bycatch is of serious concern. To put things in perspective, just consider that the total amount of reported landings of sharks and rays in the world is currently of 760,000 tonnes. If we add to this an 'average estimate' of 400,000 tonnes per year based on the above estimates of unreported bycatches, we arrive to about 1,160,000 tonnes. So we are wasting between half and 60% of what we are utilising. In addition, there might be some few thousand tonnes of skates, rays and some sharks taken in coastal bottom trawl fisheries around the world that are not accounted for in official statistics as they are usually thrown overboard. Thus the grand total could be of about 1.36 million tonnes of sharks and rays annually. A rough approximation to the total number of sharks that are killed every year around the world can be done as follows (Fig. 4). Considering an average weight of 15 kg per individual, the total number of skates, rays and sharks killed by fisheries would be close to 90 million fish per year, but this is probably overestimated as 15 kg for the average weight might be a bit low. However, if we consider only the sharks, then the figure is closer to 55 million or probably less (again, due to a conservative estimate of average weight per fish of 15 kg; many commercially important and common bycatch shark species easily attain average weights between 50 and 100 kg per fish). Sharks are only about ½ of the reported catch (i.e. 380,000 t), they represent most of the high-seas bycatch (lets say 390,000 t/y), but only a minor part of the likely 200,000 t/y or so of skates, rays and sharks taken and discarded (i.e. unreported) in coastal bottom trawl fisheries (i.e. 50,000 t/y). Thus, the total of 55 million sharks by dividing the total kill of 820,000 t of sharks by 15 kg per fish. These very rough figures try to consider all sources of shark catches, the reported ones, the unreported bycatches in high seas fisheries, and the unreported bycatches in coastal trawl fisheries. Conservation problems directly related to bycatch of sharks and rays The next important question to ask, is how bad are these levels of bycatch for the shark and ray populations? It is now known that they are actually quite damaging. In the last decade, we have learned from tangible case studies, that the mixture of high bycatch rates and the biological characteristics of these fishes can cause severe reductions or depletion of the populations of sharks and rays. Sawfishes have been driven to very low numbers across their range of distribution due to their incidental capture in coastal multispecies fisheries (Compagno and Cook, 1998). The common skate has been extirpated from the Irish Sea since the early 1980s due to its incidental capture in bottom trawl fisheries there (Brander, 1981). The barndoor skate has also been greatly reduced in numbers in the east coast of North America by trawl fisheries for groundfish, and could become the first documented case of extinction in a large marine fish if current trends continue (Casey and Myers, 1998). Several species of deep-water sharks and rays have been reduced to a trace of their original abundance as a result of their incidental catch in groundfish trawl fisheries in the east coast of Australia. Three species of deep water dogfish were virtually absent in the area having been reduced to between <1 to 2% of their original levels. Two species of greeneye dogfish were reduced to 3% of initial abundance. Ratfishes, a close relative of sharks and rays declined to 4%; and sixgill sharks to 9% (Graham et al. 1997). Clearly, it is not any more just an assumption to say that indiscriminate fishing for commercially important species can cause severe damage to the shark and ray species taken as bycatch. Conservation considerations apart, there is an additional reason why it is beneficial to keep sharks away from fisheries targeted at other species. Sharks are known to damage the catches of the target species. This problem can sometimes be very acute and can happen in fisheries as diverse as the sablefish of the NE Pacific, where sleeper sharks can damage significant parts of the catch (Hulbert and Wright, 2000) or tuna fisheries where carcharhinid sharks often damage the valuable catch (Taniuchi, 1990). In addition, sharks and rays can damage the fishing gear, can be hazardous to fishermen on deck, and in general raise the costs of fishing. From this point of view it is also desirable to keep sharks away from fishing gear. Likely consequences of depletion of sharks in an ecosystem A key consideration is assessing what are the ultimate effects of depleting sharks in an ecosystem context. A possible way to answer this question is to simulate the dynamics of species or groups of species across different types of ecosystems when sharks are rapidly depleted due to fishing. For this purpose, Stevens et al. (2000, in press) used an ecosystem model called EcoSim (Walters et al. 1998), which is a dynamic simulation model based on the food-linkages of an ecosystem. This model looks at who eats who and in what amounts, and also at the relative abundances of the groups that compose an ecosystem. The model then uses these relations to predict the size of the populations through time, when we introduce changes in the ecosystem, for example by fishing down the sharks. One of the most interesting cases analysed by Stevens et al. is the reef ecosystem of the French Frigate Shoals which is part of the Hawaiian Archipelago. Figure 5 is a representation of the food links in this ecosystem. Tiger sharks are the apex predators, while reef sharks, jacks, seabirds, and monk seals are other top predators (Polovina, 1984). The predictions of the most likely things to happen in this reef when we fish down the tiger sharks are presented in figure 6. The figure shows the relative abundance of each species for a period of 100 years when we fish tiger sharks to extremely low levels. Turtles and reef sharks increase by a factor of 9. Tiger sharks are known to have a taste for sea turtles and also prey to a limited extent on reef sharks but mainly compete with them for food. Thus, these results are in line with expectations. Bottom fishes increase by a factor of 3 and seabirds by a factor of 2.5. This is also in line with expectations given the diet of tiger sharks. Monk seals also increase in abundance although not so dramatically. However, an unforeseen and dramatic result was found when tiger sharks were depleted: the jacks and the tunas crash down and disappear. There is nothing actually wrong with the model. Looking at the diets of all groups in the ecosystem model and after further simulations changing the abundance of other groups, it became apparent that as tiger sharks eat a reasonable amount of seabirds in this ecosystem, when we take out the tiger sharks the seabirds flourish. As marine birds prey heavily in juvenile tunas and jacks, when tiger shark are depleted sea birds are free to feed at will on juvenile tuna and jacks therefore crashing down their populations. Although this is a very simplistic model of the ecosystem dynamics, the point is not how accurate the ecosystem simulation is, but to illustrate that the consequences of depleting sharks in certain ecosystems and under certain conditions could lead to unforeseen and devastating consequences. This suggest that in some cases it is not a good idea to carelessly deplete 'useless or damaging species', as this strategy might backfire by taking valuable resources down together with the tiger sharks. Possible solutions to the bycatch problem There are different alternatives for controlling and reducing bycatches in fisheries. These are not all specifically for shark and rays but some of them can be adapted for this purpose, and can be divided in two major groups (Tab. 4). First, are the management measures, which are designed to keep fishermen from going out to sea and taking bycatches or to encourage them to reduce the bycatch by decisions they make themselves. They range from incentives for reducing bycatch (like more quota of the target fish), to bycatch quotas that can also be transferable, to restrictions on the fishing season or mode of deployment of the gear. They can go all the way to reducing effort by shortening the fishing season or taking boats out of the fishery. Some fisheries have been closed in this type of decisions, just as happened to the driftnet fisheries in the North Pacific. Another alternative, much in vogue in the minds of managers in many parts of the world, is the setting up of marine protected areas, where no fishing can take place at all. The other category is the technological advances in the form of bycatch reduction devices or BRDs. This is a very promising alternative although it is costly and takes a lot of initial research and commitment. For trawl fisheries, there are many BRDs that can be adapted to the specific conditions of each fishery. These include sorting devices that lead unwanted animals to release openings in the net, such as TEDs , or grids like the Nordmore grid (Fig.7), which is a rigid structure that sorts large fish and other animals from smaller target species. Grids in particular can be of very different arrangements and types and some can work better than others for particular species (Fig. 8) . There are also separator panels of net inside the trawl, which lead species that usually swim at higher distances from the floor into an escape exit. Larger diamond panels or square mesh panels also allow the escape of species that are not targeted (Fig. 9). There are many possibilities and the technology to adapt them is there. All that is needed is a true commitment and a willingness to spend time and money. For purse seine fisheries, there are also technological innovations that can help. Some examples are the Medina panel that aids in manoeuvres to allow the escape of dolphins, and the rigid grids that allow escape of smaller fish (Fig. 10) The main problem at the moment has been to come with BRDs for longline fisheries. However, there is currently ongoing research in Japan to try to explore the possibility of adapting the technology of the SharkPod, which is an electromagnetic field device that repels sharks very effectively from divers. If successful, we could see the day that miniature SharkPods are deployed every certain number of hooks to keep the sharks from taking them and getting caught, or even from damaging the valuable catch. Artificial baits have been also used successfully in some cases and this is an area of promising research. The blackcod fishery in Alaska uses artificial bait that does not catch as many dogfish as normal bait (Erickson and Berkeley, 2000). The deployment depth of hooks can have an influence on the shark species taken as bycatch (Strasburg, 1958). Most of the possible solutions listed above have to be accompanied by effective monitoring programs through onboard observers. It is proven that this is the best and probably only way to get better information about the levels of bycatches that will allow us to make better estimates of the dimension of the problem, as well as the only real way to assure that management measures are being followed. Conclusions The problem of bycatch of sharks and rays is a real one and one which is not going to go alone. This problem is increasingly causing the depletion of more and more populations of sharks and rays throughout their range. The problem does not stop at the level of conservation of sharks and rays. The consequences of the depletion of these fishes can under certain conditions be potentially devastating both for the marine ecosystems and for our own economies and livelihoods. There are ways to reduce bycatch. A mixture of managerial and technological tools already exists. However, we need to fully accept the principles of responsible fishing and take all the measurements needed to reduce and control bycatches. Literature Alverson, D. L., M. H. Freeberg, et al. (1994). A global assessment of fisheries bycatch and discards. Rome, FAO. FAO Fisheries Technical Paper No 339: 233pp. Anonymous (1995). 1995 Shark Evaluation Annual Report. Miami, FL, NOAA, National Marine Fisheries Service. 23 pp. Bonfil, R. (1994). "Overview of world elasmobranch fisheries." FAO Fisheries Technical Paper No 341: 119pp. Brander, K. (1981). "Disappearance of common skate Raia batis from Irish Sea." Nature 290(5801): 48-49. Camhi, M. (2000). Sharks on the Line II: A State-by-State Analysis of Sharks And Ther Fisheries in the West Coast. Islip, NY, National Audubon Society's Living Oceans Program. Casey, J. M. and R. A. Myers (1998). Near extinction of a large, widely distributed fish. Science 281: 690-691. Compagno, L. J. V. and S. Cook (1998). The exploitation and conservation of freshwater elasmobranchs: status of taxa and prospects for the future. The Biology of Freshwater Elasmobranchs. Journal of Aquariculture and Aquatic Sciences VII: 62-90. Da Silva, A. A. and J. G. Pereira (1998). Catch rates for pelagic sharks taken by the Portuguese swordfish fishery in the waters around the Azores. ICCAT Working Document SCRS/98/168: 12. Erickson, D. L. and S. Berkeley, A. (2000). Methods to reduce shark-bycatch mortality in longling fisheries. Abstracts, International Pelagic Shark Conference. Asilomar Conference Centre, Pacific Grove, California. Feb 14-17, 2000. Graham, K. J., Wood, B. R., and Andrew, N. L. (1997). The 1996-97 survey of upper slope tawling grounds between Sydney and Gabo Island (and comparisons with the 1976-77 survey). Kapala Cruise Report No. 117. NSW Fisheries Research Institute, Cronulla Australia. 96 pp. Hall, M. A. (1996). On bycatches. Reviews in Fish Biology and Fisheries 6: 319-352. Hazin, F. H. V., J. R. Zagaglia, et al. (1998). Review of a small-scale pelagic longline fishery off northeastern Brazil. Marine Fisheries Review 60(3): 1-8. Hulbert, L. and B. Wright (2000). Increased shark abundance in the Gulf of Alaska. Abstracts, International Pelagic Shark Conference. Asilomar Conference Center, Pacific Grove, California. Feb 14-17, 2000. Leaman, B. M. (1994). Bycatch mortality impacts and control for Pacific Halibut. pp. 14-22, Proceedings of the Workshop: Bycatches in fisheries and their impact on the ecosystem, Fisheries Centre, Vancouver, BC. Canada, Fisheries Centre Research Reports 1994 2(1). Lennert-Cody, C., M. A. Garcia, et al. (2000). The bycatch of pelagic sharks and rays in the Eastern Pacific tuna purse-seine fishery. Abstracts, International Pelagic Shark Conference Asilomar Conference Centre, Pacific Grove, California. Feb 14-17, 2000. NMFS (1993). Fishery Management Plan for Sharks of the Atlantic Ocean. Florida, USA, National Marine Fisheries Service, National Oceanic and Atmospheric Adminsitration, U.S. Department of Commerce, Feb 1993. McCaughran, D. A. (1992). Standardized nomenclature and methods of defining bycatch levelsand implications. Proceedings of the National Industry Bycatch Workshop, Newport, Oregon, February 4-6, 1992. Seattle WA: Natural Resources Consultants. Polovina, J. J. (1984). Model of a coral reef ecosystem. I. The EcoPath model and its application to French Frigate Shoals. Coral Reefs, 3: 1-11. Seret, B. (2000). Elasmobranch bycatch of theFrench and Spanish tuna purse-seine fleets in the eastern tropical Atlantic in 1997-1999., Asilomar Conference Centre, Pacific Grove, California, Feb 14-17, 2000. Stevens, J. D. (1997). The population status of highly migratory oceanic sharks in the Pacific Ocean. Proceedings of the symposium on managing highly migratory fish on the Pacific Ocean, 4-6 Nov. 1996, Monterey, California, National Coalition for Marine Conservation, Savannah. Stevens, J. D., R. Bonfil, N. K. Dulvy, and P. A. Walker. (in press). The effects of fishing on sharks, rays and chimaeras (chondrichthyans), and the implications for marine ecosystems. ICES Journal of Marine Science. Strasburg, D., W. (1958). Distribution, abundance, and habits of pelagic sharks in the Central Pacific Ocean. U.S. Fishery Bulletin 138: 335-361. Taniuchi, T. (1990). The role of elasmobranchs in Japanese fisheries. NOAA Technical Report NMFS 90: Elasmobranchs as Living Resources: Advances in biology, ecology, systematics, and the status of the fisheries. (Eds: H.L. Pratt, Jr., S.H. Gruber. and T.Taniuchi.): 415-426. Walters, C., Christensen, V., Pauly, D. (1997). Structuring dynamic models of exploited ecosystems from trophic mass balance assessments. Reviews in Fish Biology and Fisheries, 7 (2): 139-172.

Shark Conference 2000
Online Documents
Honolulu, Hawaii February 21-24

Sponsored By:
The Barbara Delano Foundation
The Homeland Foundation
The David & Lucile Packard Foundation
The AVINA Foundation

Presented By:
WildAid
Hawaii Audubon Society
Pacific Fisheries Coalition

THE PROBLEM OF INCIDENTAL CATCHES OF SHARKS AND RAYS, ITS LIKELY CONSEQUENCES, AND SOME POSSIBLE SOLUTIONS

Ramón Bonfil
Fisheries Centre
University of British Columbia
2204 Main Hall
Vancouver, BC V6T 1Z4 Canada
Email: [email protected]

Introduction

The incidental catch of sharks in fisheries directed to other species is a major source of mortality that has raised considerable concern over the last decade (Bonfil, 1994; Camhi, 1999). In this paper I provide a brief overview of the incidental catches or bycatches of sharks in fisheries, the problems that arise from these bycatches, and some possible solutions to them. The analysis is here extended to the skates and rays, which for most purposes are sharks with a different morphology. Skates and rays share with sharks not only the same type of biology but also a similar fate in fisheries as all of them have until recently been perceived as low-value fish.

What is bycatch?

There are several definitions of what bycatch or incidental catch is. On one hand, it has been proposed that bycatch is "that portion of the catch returned to the sea as a result of economic, legal or personal consideration plus the retained catch of non-target species" (McCaughran 1992). Another similar definition of bycatches points out that these are "animals other than the target species which are unmarketable because they are too small or for some other reason" (Alverson et al. 1994). Hall (1996) proposed defining bycatch as "that portion of the capture that is discarded at sea dead (or injured to an extent that death is the most likely outcome) because it has little or no economic value or because its retention is prohibited by law".

In simpler words, these definitions say respectively that bycatches are:

All discards plus retained non-target species
Discards of non-target species
All dead discards

However, we have limited understanding about the survival of discarded sharks and rays in most fisheries. Additionally, retained catches of non-target species are seldom regulated in a fishery. Thus, for this paper I loosely define bycatches as all catches of sharks and rays in fisheries targeting other species. Whether they are discarded or retained, whether they survive the discard or not, or whether they are totally utilised or discarded after finning, is not important for this definition. What this definition is trying to comprise is that sharks and rays are being caught and usually killed in most cases without any kind of control, while supposedly fishing for other species, being these shrimps, flatfishes, tunas, or any others.

Factors affecting bycatch

There are four major factors defining the level of bycatch in a given fishery. These are:

1) The fishing grounds or type of environment. Whether fishing takes place in the sea floor or in mid water, in the temperate seas or tropical areas.
2) The fishing gear. If this is passive or active, a net, a hook or a pot.
3) The behaviour of marine animals. If they are predators or not, their association with target species, their avoidance behaviour, their food habits, etc.
4) The seasonality of movements of the bycatch species. Many species go through migrations that make them more or less available in the fishing grounds where target species are caught, and this sometimes has an important role in the level of bycatches.

Who is who in the contest for the title of 'nastiest fishing gear'?

There seems to be some confusion among the general public about what types of fishing gear are 'cleaner' for fishing and which are more indiscriminate on the species they catch. Table 1 lists the most common types of fishing gear on a global basis, ordered from the most indiscriminate to the relatively cleanest one.

There is no doubt that the most indiscriminate of fishing gears are the towed nets. In their original mode, trawl and seine nets are the kinds of fishing gear that account for the highest number and amount of non-target species caught incidentally. This is because towed nets are meant to capture all they encounter on their path, and when used in the sea floor, they encounter a high number of non-target species. Bycatch rates from these gears range from 60% of the target fish as in some temperate groundfish fisheries, to over 10 times the amount of target species in some tropical shrimp fisheries.

Gillnets of all kinds are probably the second most indiscriminate gear. They are designed to entangle or trap most of the animals that come across the wall of net. However, because they are passive gears and because of the wide range of mesh sizes in which they are built, they can be relatively more selective than trawl nets. Typical bycatch rates for pelagic gillnets before they were banned from the high seas were of about 30-40% of the total catch in numbers (Mckinnell and Seki,1998).

Longlines and in general hook and line gears, are among some of the most selective fishing devices. Not only hook size plays a role on their ability to discriminate, but also the type of bait used, and the requirement that the animal approaches and bites or otherwise gets accidentally hooked. This is not to say that longlines are totally 'clean' gear. Within a given hook size, longlines are not very selective on the type of predatory fish that they get (they are known to catch even scavenging seabirds). However, typical levels of bycatch in pelagic longline fisheries are from 20% of the total catch to about 4 times the target catch in some swordfish fisheries, although they are typically 60% of the total catch in tuna longlines (Da Silva and Pereira, 1998; Hazin et al. 1998).

Purse seines have some of the lowest level of bycatches mainly because they are highly active gears that are usually set upon schools of the target species. At least for some non-target species such as dolphins and large animals like whales and whale sharks, there is the possibility of unharmed release. The bycatch rate in this fisheries ranges from 2-10% of the total catch (Seret et al. 2000, Deelner and Hall 2000).

It is clear that although there are no totally clean fishing gears despite our claim to technological know-it-all, longlines stand among the least indiscriminate when compared with the rest of the most common fishing gears. As we will see below, the problem with bycatches in longline fisheries is not so much their selectivity but the size of the fisheries.

Main sources of shark and ray bycatches

Figure 1 shows the main fisheries, at the global scale, with large bycatches of sharks and rays.

In coastal areas, there is no doubt that bottom trawl fisheries cause the largest bycatches of sharks and rays. These fisheries mainly cause skates and ray bycatches because these species live on the ocean floor but some specific fisheries take also significant amounts of bottom-dwelling sharks.

Due to the diversity of trawl fisheries worldwide and the fragmentation of the information it is not possible to estimate the total level of ray and shark bycatches in these fisheries at the world level. However, given their scale and typical high rates of bycatch it is likely that the total figure is in the range of a few hundred thousand tonnes of skates, rays and sharks annually. The groundfish fisheries of the Hecate Strait, in western Canada discard about 640 kg of skates and rays for every tonne of target fish they keep (Leaman, 1994), while shrimp fisheries discarded some 2,800 t of sharks annually in the northern Gulf of Mexico (NMFS, 1993, Anon., 1995). Bycatch of sharks in deep-water fisheries are typically of 10-50% of the target species catch.

Pelagic trawls have also some shark bycatches, but these are poorly documented, and there are no specific numbers available. Shark and ray bycatches are likely to be small as these fisheries are not as widespread as those using bottom trawls. Nevertheless, it is known that spiny dogfish and salmon sharks are common bycatch of pelagic trawls in the North Pacific.

Coastal gillnet fisheries take also bycatches of sharks and rays, but at least in most of the tropical countries gillnet fisheries are typically multispecific and normally target a mix of species. In the context of the definition used here for bycatches, these fisheries are probably of lesser importance than trawl fisheries and most of their catch should be already accounted for in official fishery statistics. An important exception to this will be cases such as basking sharks and whale sharks caught accidentally in coastal areas.

High seas fisheries produce among the largest bycatches of sharks at the world level. The most important bycatches are in longline fisheries for tunas and billfish. The major reason of the enormous bycatch of sharks and rays in high seas longline fisheries is the large amount of hooks that are set in the sea by these fisheries. Tuna and billfish fisheries cover most of the oceans of the world, they occur in most of the Pacific, in the Indian and also in the Atlantic Ocean (Fig 2). With so much effort being applied it is not surprising that they account as the major source of shark bycatches. These fisheries typically take blue sharks as the main shark bycatch but when in tropical seas or closer to landmasses they take also important numbers of silky, oceanic whitetip, mako, porbeagle, hammerhead, and other sharks.

As shown in table 2, it has been estimated that the total amount of sharks caught in longline fisheries in the world is similar for the three major oceans where these fisheries take place (Bonfil, 1994). Alternative estimates for the Pacific Ocean (Stevens, 1997) suggest a somewhat higher number of sharks taken as bycatch. However, there are some errors in those calculations, which inflated the results (Stevens, pers. comm. Feb 2000). The most reliable estimates suggest that a total of about 232,000 tonnes of sharks and rays (manly blue sharks) were caught incidentally in longline fisheries for tunas and billfishes (Bonfil, 1994). At the time these estimates were made it was uncertain what the fate of the sharks were, but throughout the 1990s there have been numerous reports of increasing rate of finning of sharks in longline fisheries in all oceans. So we can pretty much assume that all these estimates actually translate into deaths.

Drift gillnet fisheries use to be the second major source of shark bycatch in the high seas. Available estimates from the early 1990s indicate that driftnet fisheries accounted for nearly 30,000 tonnes of shark and ray bycatches annually (Bonfil, 1994). However, high-seas driftnet fisheries have been banned since 1992.

The third most important source of shark and ray bycatches are the tuna purse-seine fisheries of tropical seas (Fig. 3). The estimate presented here is a combination of Stevens' (1997) estimates for the Pacific Ocean with others available for the Atlantic and the Indian Ocean. This is almost surely an underestimate as the method used for the Atlantic and Indian Ocean was very crude compared to that of Stevens.

In all, these high seas fisheries could account for almost 300,000 tonnes of incidentally caught sharks and rays (Tab. 3). This figure could be closer to ½ a million tonnes if we consider the estimate of Stevens for the Pacific Ocean longline fisheries, but we know these are overestimated. This amount of bycatch is of serious concern. To put things in perspective, just consider that the total amount of reported landings of sharks and rays in the world is currently of 760,000 tonnes. If we add to this an 'average estimate' of 400,000 tonnes per year based on the above estimates of unreported bycatches, we arrive to about 1,160,000 tonnes. So we are wasting between half and 60% of what we are utilising. In addition, there might be some few thousand tonnes of skates, rays and some sharks taken in coastal bottom trawl fisheries around the world that are not accounted for in official statistics as they are usually thrown overboard. Thus the grand total could be of about 1.36 million tonnes of sharks and rays annually.

A rough approximation to the total number of sharks that are killed every year around the world can be done as follows (Fig. 4). Considering an average weight of 15 kg per individual, the total number of skates, rays and sharks killed by fisheries would be close to 90 million fish per year, but this is probably overestimated as 15 kg for the average weight might be a bit low. However, if we consider only the sharks, then the figure is closer to 55 million or probably less (again, due to a conservative estimate of average weight per fish of 15 kg; many commercially important and common bycatch shark species easily attain average weights between 50 and 100 kg per fish). Sharks are only about ½ of the reported catch (i.e. 380,000 t), they represent most of the high-seas bycatch (lets say 390,000 t/y), but only a minor part of the likely 200,000 t/y or so of skates, rays and sharks taken and discarded (i.e. unreported) in coastal bottom trawl fisheries (i.e. 50,000 t/y). Thus, the total of 55 million sharks by dividing the total kill of 820,000 t of sharks by 15 kg per fish. These very rough figures try to consider all sources of shark catches, the reported ones, the unreported bycatches in high seas fisheries, and the unreported bycatches in coastal trawl fisheries.

Conservation problems directly related to bycatch of sharks and rays

The next important question to ask, is how bad are these levels of bycatch for the shark and ray populations? It is now known that they are actually quite damaging. In the last decade, we have learned from tangible case studies, that the mixture of high bycatch rates and the biological characteristics of these fishes can cause severe reductions or depletion of the populations of sharks and rays.

Sawfishes have been driven to very low numbers across their range of distribution due to their incidental capture in coastal multispecies fisheries (Compagno and Cook, 1998). The common skate has been extirpated from the Irish Sea since the early 1980s due to its incidental capture in bottom trawl fisheries there (Brander, 1981). The barndoor skate has also been greatly reduced in numbers in the east coast of North America by trawl fisheries for groundfish, and could become the first documented case of extinction in a large marine fish if current trends continue (Casey and Myers, 1998).

Several species of deep-water sharks and rays have been reduced to a trace of their original abundance as a result of their incidental catch in groundfish trawl fisheries in the east coast of Australia. Three species of deep water dogfish were virtually absent in the area having been reduced to between <1 to 2% of their original levels. Two species of greeneye dogfish were reduced to 3% of initial abundance. Ratfishes, a close relative of sharks and rays declined to 4%; and sixgill sharks to 9% (Graham et al. 1997). Clearly, it is not any more just an assumption to say that indiscriminate fishing for commercially important species can cause severe damage to the shark and ray species taken as bycatch.

Conservation considerations apart, there is an additional reason why it is beneficial to keep sharks away from fisheries targeted at other species. Sharks are known to damage the catches of the target species. This problem can sometimes be very acute and can happen in fisheries as diverse as the sablefish of the NE Pacific, where sleeper sharks can damage significant parts of the catch (Hulbert and Wright, 2000) or tuna fisheries where carcharhinid sharks often damage the valuable catch (Taniuchi, 1990). In addition, sharks and rays can damage the fishing gear, can be hazardous to fishermen on deck, and in general raise the costs of fishing. From this point of view it is also desirable to keep sharks away from fishing gear.

Likely consequences of depletion of sharks in an ecosystem

A key consideration is assessing what are the ultimate effects of depleting sharks in an ecosystem context. A possible way to answer this question is to simulate the dynamics of species or groups of species across different types of ecosystems when sharks are rapidly depleted due to fishing. For this purpose, Stevens et al. (2000, in press) used an ecosystem model called EcoSim (Walters et al. 1998), which is a dynamic simulation model based on the food-linkages of an ecosystem. This model looks at who eats who and in what amounts, and also at the relative abundances of the groups that compose an ecosystem. The model then uses these relations to predict the size of the populations through time, when we introduce changes in the ecosystem, for example by fishing down the sharks.

One of the most interesting cases analysed by Stevens et al. is the reef ecosystem of the French Frigate Shoals which is part of the Hawaiian Archipelago. Figure 5 is a representation of the food links in this ecosystem. Tiger sharks are the apex predators, while reef sharks, jacks, seabirds, and monk seals are other top predators (Polovina, 1984). The predictions of the most likely things to happen in this reef when we fish down the tiger sharks are presented in figure 6. The figure shows the relative abundance of each species for a period of 100 years when we fish tiger sharks to extremely low levels. Turtles and reef sharks increase by a factor of 9. Tiger sharks are known to have a taste for sea turtles and also prey to a limited extent on reef sharks but mainly compete with them for food. Thus, these results are in line with expectations. Bottom fishes increase by a factor of 3 and seabirds by a factor of 2.5. This is also in line with expectations given the diet of tiger sharks. Monk seals also increase in abundance although not so dramatically. However, an unforeseen and dramatic result was found when tiger sharks were depleted: the jacks and the tunas crash down and disappear.

There is nothing actually wrong with the model. Looking at the diets of all groups in the ecosystem model and after further simulations changing the abundance of other groups, it became apparent that as tiger sharks eat a reasonable amount of seabirds in this ecosystem, when we take out the tiger sharks the seabirds flourish. As marine birds prey heavily in juvenile tunas and jacks, when tiger shark are depleted sea birds are free to feed at will on juvenile tuna and jacks therefore crashing down their populations.

Although this is a very simplistic model of the ecosystem dynamics, the point is not how accurate the ecosystem simulation is, but to illustrate that the consequences of depleting sharks in certain ecosystems and under certain conditions could lead to unforeseen and devastating consequences. This suggest that in some cases it is not a good idea to carelessly deplete 'useless or damaging species', as this strategy might backfire by taking valuable resources down together with the tiger sharks.

Possible solutions to the bycatch problem

There are different alternatives for controlling and reducing bycatches in fisheries. These are not all specifically for shark and rays but some of them can be adapted for this purpose, and can be divided in two major groups (Tab. 4). First, are the management measures, which are designed to keep fishermen from going out to sea and taking bycatches or to encourage them to reduce the bycatch by decisions they make themselves. They range from incentives for reducing bycatch (like more quota of the target fish), to bycatch quotas that can also be transferable, to restrictions on the fishing season or mode of deployment of the gear. They can go all the way to reducing effort by shortening the fishing season or taking boats out of the fishery. Some fisheries have been closed in this type of decisions, just as happened to the driftnet fisheries in the North Pacific. Another alternative, much in vogue in the minds of managers in many parts of the world, is the setting up of marine protected areas, where no fishing can take place at all.

The other category is the technological advances in the form of bycatch reduction devices or BRDs. This is a very promising alternative although it is costly and takes a lot of initial research and commitment.

For trawl fisheries, there are many BRDs that can be adapted to the specific conditions of each fishery. These include sorting devices that lead unwanted animals to release openings in the net, such as TEDs , or grids like the Nordmore grid (Fig.7), which is a rigid structure that sorts large fish and other animals from smaller target species. Grids in particular can be of very different arrangements and types and some can work better than others for particular species (Fig. 8) .

There are also separator panels of net inside the trawl, which lead species that usually swim at higher distances from the floor into an escape exit. Larger diamond panels or square mesh panels also allow the escape of species that are not targeted (Fig. 9). There are many possibilities and the technology to adapt them is there. All that is needed is a true commitment and a willingness to spend time and money.

For purse seine fisheries, there are also technological innovations that can help. Some examples are the Medina panel that aids in manoeuvres to allow the escape of dolphins, and the rigid grids that allow escape of smaller fish (Fig. 10)

The main problem at the moment has been to come with BRDs for longline fisheries. However, there is currently ongoing research in Japan to try to explore the possibility of adapting the technology of the SharkPod, which is an electromagnetic field device that repels sharks very effectively from divers. If successful, we could see the day that miniature SharkPods are deployed every certain number of hooks to keep the sharks from taking them and getting caught, or even from damaging the valuable catch.

Artificial baits have been also used successfully in some cases and this is an area of promising research. The blackcod fishery in Alaska uses artificial bait that does not catch as many dogfish as normal bait (Erickson and Berkeley, 2000). The deployment depth of hooks can have an influence on the shark species taken as bycatch (Strasburg, 1958).

Most of the possible solutions listed above have to be accompanied by effective monitoring programs through onboard observers. It is proven that this is the best and probably only way to get better information about the levels of bycatches that will allow us to make better estimates of the dimension of the problem, as well as the only real way to assure that management measures are being followed.

Conclusions

The problem of bycatch of sharks and rays is a real one and one which is not going to go alone. This problem is increasingly causing the depletion of more and more populations of sharks and rays throughout their range. The problem does not stop at the level of conservation of sharks and rays. The consequences of the depletion of these fishes can under certain conditions be potentially devastating both for the marine ecosystems and for our own economies and livelihoods. There are ways to reduce bycatch. A mixture of managerial and technological tools already exists. However, we need to fully accept the principles of responsible fishing and take all the measurements needed to reduce and control bycatches.

Literature

Alverson, D. L., M. H. Freeberg, et al. (1994). A global assessment of fisheries bycatch and discards. Rome, FAO. FAO Fisheries Technical Paper No 339: 233pp.

Anonymous (1995). 1995 Shark Evaluation Annual Report. Miami, FL, NOAA, National Marine Fisheries Service. 23 pp.

Bonfil, R. (1994). "Overview of world elasmobranch fisheries." FAO Fisheries Technical Paper No 341: 119pp.

Brander, K. (1981). "Disappearance of common skate Raia batis from Irish Sea." Nature 290(5801): 48-49.

Camhi, M. (2000). Sharks on the Line II: A State-by-State Analysis of Sharks And Ther Fisheries in the West Coast. Islip, NY, National Audubon Society's Living Oceans Program.

Casey, J. M. and R. A. Myers (1998). Near extinction of a large, widely distributed fish. Science 281: 690-691.

Compagno, L. J. V. and S. Cook (1998). The exploitation and conservation of freshwater elasmobranchs: status of taxa and prospects for the future. The Biology of Freshwater Elasmobranchs. Journal of Aquariculture and Aquatic Sciences VII: 62-90.

Da Silva, A. A. and J. G. Pereira (1998). Catch rates for pelagic sharks taken by the Portuguese swordfish fishery in the waters around the Azores. ICCAT Working Document SCRS/98/168: 12.

Erickson, D. L. and S. Berkeley, A. (2000). Methods to reduce shark-bycatch mortality in longling fisheries. Abstracts, International Pelagic Shark Conference. Asilomar Conference Centre, Pacific Grove, California. Feb 14-17, 2000.

Graham, K. J., Wood, B. R., and Andrew, N. L. (1997). The 1996-97 survey of upper slope tawling grounds between Sydney and Gabo Island (and comparisons with the 1976-77 survey). Kapala Cruise Report No. 117. NSW Fisheries Research Institute, Cronulla Australia. 96 pp.

Hall, M. A. (1996). On bycatches. Reviews in Fish Biology and Fisheries 6: 319-352.

Hazin, F. H. V., J. R. Zagaglia, et al. (1998). Review of a small-scale pelagic longline fishery off northeastern Brazil. Marine Fisheries Review 60(3): 1-8.

Hulbert, L. and B. Wright (2000). Increased shark abundance in the Gulf of Alaska. Abstracts, International Pelagic Shark Conference. Asilomar Conference Center, Pacific Grove, California. Feb 14-17, 2000.

Leaman, B. M. (1994). Bycatch mortality impacts and control for Pacific Halibut. pp. 14-22, Proceedings of the Workshop: Bycatches in fisheries and their impact on the ecosystem, Fisheries Centre, Vancouver, BC. Canada, Fisheries Centre Research Reports 1994 2(1).

Lennert-Cody, C., M. A. Garcia, et al. (2000). The bycatch of pelagic sharks and rays in the Eastern Pacific tuna purse-seine fishery. Abstracts, International Pelagic Shark Conference Asilomar Conference Centre, Pacific Grove, California. Feb 14-17, 2000.

NMFS (1993). Fishery Management Plan for Sharks of the Atlantic Ocean. Florida, USA, National Marine Fisheries Service, National Oceanic and Atmospheric Adminsitration, U.S. Department of Commerce, Feb 1993.

McCaughran, D. A. (1992). Standardized nomenclature and methods of defining bycatch levelsand implications. Proceedings of the National Industry Bycatch Workshop, Newport, Oregon, February 4-6, 1992. Seattle WA: Natural Resources Consultants.

Polovina, J. J. (1984). Model of a coral reef ecosystem. I. The EcoPath model and its application to French Frigate Shoals. Coral Reefs, 3: 1-11.

Seret, B. (2000). Elasmobranch bycatch of theFrench and Spanish tuna purse-seine fleets in the eastern tropical Atlantic in 1997-1999., Asilomar Conference Centre, Pacific Grove, California, Feb 14-17, 2000.

Stevens, J. D. (1997). The population status of highly migratory oceanic sharks in the Pacific Ocean. Proceedings of the symposium on managing highly migratory fish on the Pacific Ocean, 4-6 Nov. 1996, Monterey, California, National Coalition for Marine Conservation, Savannah.

Stevens, J. D., R. Bonfil, N. K. Dulvy, and P. A. Walker. (in press). The effects of fishing on sharks, rays and chimaeras (chondrichthyans), and the implications for marine ecosystems. ICES Journal of Marine Science.

Strasburg, D., W. (1958). Distribution, abundance, and habits of pelagic sharks in the Central Pacific Ocean. U.S. Fishery Bulletin 138: 335-361.

Taniuchi, T. (1990). The role of elasmobranchs in Japanese fisheries. NOAA Technical Report NMFS 90: Elasmobranchs as Living Resources: Advances in biology, ecology, systematics, and the status of the fisheries. (Eds: H.L. Pratt, Jr., S.H. Gruber. and T.Taniuchi.): 415-426.

Walters, C., Christensen, V., Pauly, D. (1997). Structuring dynamic models of exploited ecosystems from trophic mass balance assessments. Reviews in Fish Biology and Fisheries, 7 (2): 139-172.