Principles involved in reducing species bycatch
Principles involved in reducing species bycatch and to size-select fish and shellfish in mobile gears
Issue
Selectivity in mobile gears like trawls and seines simply means that targets are caught whereas non-targets are allowed to escape during the catching process. Targets might be some wanted species of certain size-ranges, often above a pre-set minimum whereas the non-targets will consist of all other species and size-ranges that are incidentally caught.
In principle there are two ways organisms can escape from being caught in a trawl or seine gear, one is through passive filtering whereas the other depends on active escapement from the gear while inside its influence. As all selection processes in such gear involve one of these two principles or a combination of both, I will in this paper attempt to describe the fish behaviour and gear characteristics of importance for this process. An elaboration of the underlying principles for the selectivity process might hopefully facilitate the understanding of how various selection devices work and may also be useful as guidance for development of new devices when improved selectivity in a specific fishery is required.
Selectivity Process
Interaction between the behaviour of living organisms and the moving gear is the key factor in any selection process. Above anything else, the individual behaviour pattern of species affects the success of selectivity. Organisms that are passive towards external stimulation will simply be overtaken by the moving gear and will only be released from the gear if there is an opening large enough to penetrate where it is taken to by the water flow. Some organisms, however, have the capability to sense parts of the approaching gear and will react in various ways. The most common reaction is to swim away from it. This behaviour, also known as herding, is seen in different parts of the gear and will normally result in a separation of organisms with different swimming capabilities, e.g. large and small individuals of the same species. Another behaviour pattern typical for some organisms is that they can sense escape holes that they will try to penetrate when in a frightening situation as in the codend. When organisms become more densely packed than they commonly are in their natural environment, some will become frightened and react with panic. The resulting behaviour is vigorous flight in all directions, a reaction often seen in the aft belly of trawls when the cross section of the trawl has been reduced to a fraction of what it is at the entrance
The following is a description of how these basic behaviour patterns can be used to select species of various sizes in four different zones of a trawl gear, the pre-trawl area (PTZ), the trawl belly (TBZ), the selection zone (SZ) and the codend proper (CP).
Pre-Trawl Zone (PTZ)
This zone extends forward from the trawl wings and as far as the point where the organisms can sense the coming fishing vessel. Laterally the zone is identical to the distance between the trawl doors, or in case of pair trawls operation the area between the towing vessels.
This zone in front of the trawl itself is not normally considered of much importance for selectivity action, an assumption based on the fact that selective performance arising in this area is difficult to measure and thus document. There is, however, evidence that some size and species selection occurs in this area.
Organisms can sense the towing vessel either by the noise generated from its engine or when the vessel is illuminated during darkness. A common reaction to both these types of stimulators is avoidance inasmuch as the stimulated organisms can flee in all directions; upwards, downwards, laterally or directly forward away form the approaching stimulus. In general, larger fish swim faster than smaller individuals of the same species. Some species are also superior to others in their swimming capability. The impact this might have on selectivity is that the largest organisms might swim away from the path between the trawl doors and this to some extent reduces the catch of targets. On the other hand, this reaction pattern can also be used to reduce bycatch of "reacting fish" when the targets show no directional avoidance, e.g. in shrimp fisheries with beam trawls. Pair trawling is another example where selectivity is dependent on size and species, where some can avoid the vessel and where, of course, the best swimmers are more successful than the weaker ones. Some of the smallest fish closest to the vessels will be overtaken by the towing warps and thus avoid capture.
The sweeps between the doors and the wings are probably the most important stimulators in the PTZ with impact on selectivity. The larger individuals of a species are better swimmers than the smaller ones and will thus be more efficiently herded into the catching zone between the trawl wings, whereas the smaller ones might be overtaken by the sweeps and thus avoid being captured. As some species are more efficiently herded by sweeps than others, the length of sweeps and their angle relative to the towing direction will significantly effect the species composition entering the trawl belly zone. Crustacean species are poorly herded by sweeps unlike many fish species. A trawl rigging without or with very short sweeps is thus beneficial in some fisheries, e.g. shrimp, where fish bycatch is to be avoided.
Trawl Belly Zone (TBZ)
The Trawl Belly Zone extends from the area below the headline, inside the trawl back to where its internal diameter is reduced to approximately two meters. The zone is typically a cone of net panels with decreasing mesh sizes form the front to the rear. The slope of the panels might vary according to the gear design, ranging from 5 to 20 degrees.
As for the PTA, neither this zone is regarded as a very important selectivity area. Some specific behaviour patterns might, however, be utilized to effect the selectivity also here. When different species enter the mouth of a trawl they exhibit various kinds of behaviour. Some species might move upwards and others have a tendency to go downwards to the bottom, while the majority just drop backwards in the same level as they enter into the trawl mouth. This behaviour is sometimes used to separate species, e.g. crustaceans from fish into an upper and lower codend each with different mesh sizes to suit the target sizes of the two groups of organisms. Organisms with minor directive reaction patterns will be taken by the water flow towards the netting panels of the belly. The smallest organisms will pass through the meshes more easily than larger ones. Inserting slack side panels to open the meshes even more sometimes increases this effect. A third behaviour pattern worth mentioning is the panic reaction often seen when fish become densely packed in the narrow part aft. When the meshes are big enough, significant numbers of fish can escape. Size and body shape in relation to the mesh opening will influence the success of escape effort in this part of the trawl. This escape behaviour sometimes results in meshing or gilling of fish when the girth of the fish is comparable to the mesh circumference.
Selection Zone (SZ)
The trawl selection zone is the narrow netting tunnel connecting the belly and the codend proper of the gear. The reason for naming it a Selection Zone, is its prospect as an area for improved selectivity, which indeed has proven to be the case in recent developments. There are three obvious reasons why this part of the trawl is so important for selectivity purposes.
- the density of organisms passing the zone is high
- the water flow entering the zone is comparable to the towing speed
- the netting tube is narrow, allowing small but complicated devices to be inserted
Typical behaviour of non-reacting organisms, like most crustaceans, is that they move backwards with the water flow and will pass this area rapidly if no special selectivity device is inserted. Some "reactive" organisms will often enter the zone tail first and try to swim forward but gradually drop backwards. .
Filtering is no doubt the most effective sorting principle in this zone. The filter can be netting that is angled relative to the moving direction; or more effectively, a solid grid of parallel bars made from metal or plastic arranged in such a way that when organisms hit the grid, they will either penetrate the bars or pass it by.
The Nordmoere grid principle as illustrated in Figure 1, designed to reduce fish bycatch in shrimp trawls is probably the most successful application of this principle. When properly designed, this device releases all fish bycatch above a certain size as the loss of target shrimps are considered minimal and economically acceptable. This device is now in commercial use in most fisheries for deepwater shrimp in northern waters. Another important application of the grid concept is in the tropical shrimp fishery where turtle bycatch are efficiently released unharmed in the catching process. In some cases sloping netting panels are used for the same purpose.
The filtering mechanism has also proven to be efficient to size select organisms of the same species. Two important success factors are that all sizes of organisms are equally exposed to the filter and that the filter is kept clean and thus functional throughout the haul. Maximum exposure to the filter (grid) is often achieved by using guiding panels in front of it as illustrated in Figure 2. To keep the grid free from blocking organisms and other objects throughout the haul is often the most critical part of the operation of such devices. The slope of the grid and its smoothness are particularly important in this respect.
In instances when the targets and non-targets are of similar sizes, the filtering principle is inefficient in separating the two groups. Behaviour differences can then be used, which allow the non-targets to escape through specially designed escape openings, whereas the targets are guided into and kept in the codend. Such openings might be meshes which are big enough to allow non-targets to penetrate or larger escape holes which the non-targets are stimulated to seek and swim through. The principle is illustrated in Figure 3. One variant of the principle is descriptively called a Radial Escape Section (RES) as it relies on a central guiding funnel where the non-reacting organisms will be transported to the codend whereas some species will be stimulated to radial flights through openings outside this path. The stimulator is a critical factor for the operation of such a device, and although simple devices are used for this purpose, a better understanding of the reaction patterns of organisms in this particular phase is required to successfully develop devices that can separate different species of similar sizes.
Codend Proper (CP)
The codend has until recently been regarded as the major area for selectivity action, in particular by size of organisms. Most existing regulations therefore concentrate on a minimum mesh size in the codend. The mesh size has no doubt a significant effect on the selectivity of trawl and seine gears, and should therefore be regarded as the primary tool for size selectivity also in the near future. A standard codend is, however, imperfect for selectivity action for at least two major reasons: one is when full of catch, the mesh openings in front of the accumulated catch are reduced and thus prevent escapement of gradually smaller individuals. When the catch fills the rear part of the codend the direct backward flow of water is blocked, and thus passive organisms will start to be stacked up in front of the accumulated catch. Consequently the release of the small organisms is reduced.
The first of these constraints can partly be solved by codend designs where the meshes stay open independently of the catch. Turning the netting panel 45 degrees so that the meshes attain a square configuration is one solution, which has proven workable. Another is to let longitudinal ropes take the strain of the catch allowing the meshes to open laterally in the area in front of the accumulated catch as it moves forward. Diamond meshes can also be reinforced with some stiff material, which will keep the meshes open independently of the catch.
Accumulation of catch in the codend is of course the ultimate aim of any catching process and is therefore impossible to avoid. Is it then possible to do anything with the codend design that improves the filtering through codend meshes? As the objects have to be directed to netting parts where the meshes stay open, a possible solution might be to change the straight backward line of the codend to a bent one. This can be achieved by inserting extra weight or floats in the rear part of the codend. Such a bending performance is in fact observed when non-buoyant organisms like shrimp fill the codend and it is then common to see small shrimp passing through the meshes on top and in front of the accumulated catch.