| PANEL | 5 | WEIRS, FISH PASSES AND TRAPS |
| GROUPE | BARRIERES, PASSES A POISSONS ET ECHELLES |
by/par
C.J. McGrath
Department of Agriculture and Fisheries, Fisheries Division
3 Cathal Brugha Street, Dublin 1
Ireland/Irlande
ABSTRACT
Fish passes: A number of installations are described which have the advantage of capturing or recording the passage of all fish passing up or down the river. Such installations are however very expensive and the magnitude and specific variety of the runs in many places limit the use of recording apparatus to areas where few species run up rivers in moderate quantities.
Traps for migrating fish: A range of types of fish trap, both fixed and mobile, is considered. Data are readily obtained on numbers of fish moving up and down river and fish are obtained relatively unharmed for tagging. Permanent structures, however, have a high capital cost and may affect the population within the river, whereas temporary structures costing less only sample a proportion of the fish population which is not always known.
Portable traps are generally unreliable although they may be the only type of gear which will catch fish in certain situations.
RESUME
Passes à poissons: On décrit un certain nombre d'installations qui présentent l'avantage de capturer ou d'enregisrer le passage de tous les poissons remontant ou descendant le fleuve. Ces installations sont toutefois très onéreuses et l'ampleur et la varieté spécifique des remontées restreignent fréquemment l'utilisation d'appareils enregistreurs aux zones ou les espèces qui remontent le fleuve sont peu nombreuses.
Pièges pour poissons migrateurs: Une gamme de pièges fixes ou mobiles est examinée. On obtient facilement des données sur le nombre de poissons se déplaçant vers l'amont et l'aval du fleuve et les poissons destinés à être marqués ne sont généralement pas endommagés. Toutefois, les structures permanentes entrainent des dépenses en capital élevées et sont susceptibles d'affecter la population fluviatile; par contre, les structures temporaires, moins coûteuses, n'échantillonnent qu'une proportion souvent indéterminée de la population de poissons.
Les pièges portatifs sont généralement peu fiables mais constituent parfois le seul type d'engin capable de prendre le poisson dans certaines situations.
CONTENTS
1. FISH COUNTING INSTALLATIONS IN FISH PASSES
1.1 Description of Method
1.2 Use of Method
1.3 Type of Data Obtained
1.4 Summary of Advantages and Disadvantages
2. FISH TRAPS FOR ANADROMOUS FISH AND THEIR YOUNG
2.1 Description of Method
2.2 Use of Method
2.3 Type of Data Obtained
2.4 Summary of Advantages and Disadvantages
3. FISH TRAPS - SELF-CONTAINED PORTABLE UNITS
3.1 Description of Method
3.2 Use of Method
3.3 Type of Data Obtained
3.4 Summary of Advantages and Disadvantages
REFERENCES
Classification: Fish counting installations in fish passes to which modifications have been made to adapt them for the identification and counting of migratory fish passing through. In some cases the arrangement permits the removal of fish for examination, overland transport, or use as brood stock in fish rearing installations. These fish passes almost invariably have been built in compliance with statutory requirements to enable migratory fish to surmount obstacles to their passage upstream and downstream.
Design and construction:
(i) Fish traps (adult) are the simplest arrangement for the capture and counting of upstream migrants (McGrath, see p.447) consisting of a special pool in the fish pass structure which has been adapted for this purpose by providing an inscale at its downstream entrance permitting fish to enter through a narrow opening but so arranged to impede or prevent them returning downstream by this route. The upstream exit from the pool is fitted with a bar screen which can be lifted as required to allow the fish to escape. A brailing device is usually provided in the form of a false floor unit which can be raised as necessary by a winch, thereby bringing the fish to the surface of the water for examination and in some cases for removal to fish rearing units or for planting in head waters. At the end of each inspection the exit screen is raised to permit the fish to escape upstream.
A large temporary wooden cage structure fitted against the upstream face of a hydro-electric dam and covering the upstream exit from a fish pass is described by Ruggles (see p.466). This is placed in position at the beginning of each migration season and removed at the end. The unit is fitted with a brailing arrangement somewhat on the lines already described and the manner of operation is also somewhat similar. Ruggles also describes a number of other similar installations, the basic principles of which are more or less as already described, but which possess modifications in detail to take account of the particular design of the fish pass installation or which are dictated by the particular conditions obtaining at the site.
(ii) Automatic fish counters (adult) - Fish passes lend themselves very readily to adaptation for the installation of automatic fish counters. Fish may be counted electronically (see panel 4b), acoustically (see panel 4a), magnetically (Trefethen and Collins, see p.490), by electro-mechanical methods (McGrath, see p.447) or optically (Trefethen and Collins, see p. 490).
Mode of action: In all the methods described above the action of the gear is passive.
Conditions in which gear can be used: The gear described could be used in various situations but is particularly suited for use in fish passes which are located in rivers.
Fish species or groups which can be sampled: Anadromous species such as Atlantio salmon (Salmo salar); sea trout (Salmo trutta); American shad (Alosa sapidissima); the alewife (Alosa pseudoharengus); chinook (O. tshawytscha); coho (O. kisutch); sockeye (O. nerka); ohum salmon (O. keta); sturgeon (Acipenser spp.); carp (Cyprinus carpio); suckers (Castostomus spp.); chubs (Hybopis spp.); catfishes (Ictaluridae); shiners (Notropis spp.); northern squawfish (Ptychocheilus oregonensis); Pacific lamprey (Entrosphenus tridentatus).
Quantitative studies of anadromous species can be made in their mass migration in from the sea and long-term trends made apparent. Estimates of spawning escapement provided by counts at dams have been useful in forecasting the magnitude of runs of returning fish 5–6 years hence (Ruggles, see p.466). Population changes can be measured as can changes in run timing, age composition, age at first maturity, and their effects on natural reproduction evaluated. The contribution made to the run by hatchery-reared fish can be assessed and the possibility of selective breeding made practicable. Data are provided for solution of practical fishery management problems such as allocation of fish between commercial and sport fishing interests while ensuring adequate spawning escapement for natural reproduction. The effects of hydro-electric development and pollution on fish stocks can be studied.
Trefethen and Collins (see p.490) also point out that information obtained in this way permits assessment of the effect of ecological and environmental changes brought about by impoundments including the effects of the passage of juvenile fish through turbines by the recovery at the adult stage of specially treated populations of young fish prior to their downstream migration.
The variety of species as well as the magnitude of runs experienced in some rivers, especially on the Pacific coast of the U.S.A. and in some of the larger rivers elsewhere, could possibly rule out the use of trapping installations in fish passes as counting instruments or limit their use to means of sampling portions of the run diverted for this purpose. It is also unlikely that water bridge resistivity units in fish passes could deal with a large variety of species or very great numbers, while at the same time ensuring that there will be no undue delay in the movement of the fish in the pass that would cause a hold-up which would tend to subject the fish to stress.
Ruggles (see p. 466) draws attention to the need for greater attention being paid to the development of sampling theory so as to minimize the numbers of specimens physically handled and thereby lessen the stress on the entire population. The employment of video-tape recording and documentation mentioned by Trefethen and Collins (see p.490) would appear to be a promising development, especially as the present system of visual counting is expensive in manpower and susceptible to error because of lack of full coverage during daylight hours and absence of observation during hours of darkness.
Fish passes do ensure, however, that all upstream fish migrations pass by this particular route thus ensuring that the entire run is made available for counting. The manpower requirements are reduced to negligible proportions.
Classification: Fish traps for anadromous fish and their young include stationary traps which are associated with or are component parts of fish fences or fish weirs which have been erected in rivers for the sole purpose of catching migrating fish. Some have been built for commercial fishing but also supply data for scientific purposes like the Thomond Salmon Weir on the River Shannon and some are of considerable antiquity such as the brush weirs or fences pre-dating 1680, which were built by the North American Indians (Ruggles, see p.466). In some installations the barrier to fish movement extends across the entire river channel but in others it extends out from the bank for a comparatively short distance only.
Also included are a number of floating trap units which are employed in impoundments and in rivers.
Reference is made to arrangements made at hydro-electric dams whereby constructional features of the dam are used as a means of collecting young fish for sampling purposes.
Design and construction: A barrier to the movement of fish upstream and downstream is placed across all or part of the river channel in such a way that the migrating fish are diverted into trapping units. This barrier can be in the vertical plane (Shearer, see p. 571, Ruggles, see p. 466) or in the horizontal plane (McGrath, see p. 447). This latter arrangement is possible only where there is a localized drop in the river bed which will permit all the river flow to pass down through the screening device at all times and not to pass across it. If this happens it allows the migrating fish to escape past the screening device instead or being sieved out of the water flow by it.
The erection of a barrier screen, especially one containing small apertures, across the full width of a river channel is not always practicable because of the danger of it clogging with debris and causing either flooding upstream or the collapse of the structure itself under the water pressure. Considerable thought has been given, therefore, to providing alternative means of stopping the passage of fish. Among the proposals put forward have been those for the employment of electric screens (Lethlean, 1953; Hartley and Simpson, 1967; Chmielewski, 1967; McGrath et al., 1969); louver screens (Bates and Vinsonhaler, 1954), and more recently the horizontal travelling screen described by Bates (1970).
Electric barriers have been employed successfully in preventing upstream fish migration especially where the velocity of river flow at the site of the barrier is sufficiently high to sweep downstream fish affected by the electric field. The same degree of success has not been achieved up to the present with downstream migrants. Hartley and Weiss (see p.159) now report achieving an efficiency of 91.4% in the particular situation in which their experiments were carried out. McGrath (see p. 447) reports encouraging results with the Sharkey Programmed Electronic Trap.
Ruggles reports Ducharme (1972) to be of the opinion that louvers are a practical means of guiding Atlantic salmon smolts from narrow fast flowing bodies of water in which the velocity of flow is in excess of 0.6 m/second. Prentice and Ossiander (1973) report achieving diversion efficiencies of 97% in experiments with spring chinook fingerlings using the horizontal travelling screen. It would appear, however, that apart from using electric screens as a means of preventing upstream migration, the flow conditions necessary for the optimum operation of the other methods mentioned above would restrict their application to specially constructed sites or water intakes where controlled conditions of flow can be assured.
Ruggles (see p. 466) describes the use in rivers of a number of traps constructed from netting mounted on a supporting framework such as are erected by commercial fishermen and a net trap arrangement employed by Blair (1957) which can be easily erected and dismantled. In free flowing rivers where water velocities exceed 1.2 m/s floating self-cleaning scoop traps may be used in depths of water not in excess of 1.2 m (Raymond and Collins, see p.552).
Raymond and Collins also employ the so-called “Migrant Dipper”, as described by Mason (1966), in rivers where the velocity is less than 1.2 m/second.
In impoundments and river sections where the velocity of flow is minimal “Merwin Traps” are used. These consist of a floating pound unit made up of netting with an attached leader net extending to the shore. In the vicinity of the pound unit this leader net incorporates a number of guiding vane nets at each side which ensure that fish which come up against the net and travel along it seeking a route past will be guided into the pound unit. The arrangement of these guide nets is also such as to prevent the fish from escaping once they have entered the pound.
Clay (1961) described the use of the Fish Wheel to capture adult Pacific salmon for tagging in very wide silty rivers in which very large numbers of fish are present at the time of operation.
Vertical travelling screens, located in the upper level of turbine intakes, divert large numbers of juvenile salmon and trout which otherwise would pass downstream through the turbine into gate-wells. These young fish are removed from the gate-well as necessary by a large power operated dip-net (Raymond and Collins, see p. 552).
In the forebay of the turbines of Ardnacrusha Hydro-electric Station on the River Shannon in Ireland a series of floodlights is provided at intervals leading from the forebay into a nearby canal lock. These are sequentially switched on and off ending in the canal lock. The cycle is repeated at suitable intervals. They have been successfully employed to attract salmon smolts from the forebay into the canal lock system by means of which they are transferred down past the dam by the operation of the lock gates when a sufficient number of smolts has been assembled. This arrangement provides the possibility for counting and sampling the smolt run.
Mode of action: In all the methods described above except the Fish Wheel and those employing electricity the action of the device is passive.
Conditions in which gear can be used: Fish weirs or fish fences can be used only in rivers where they create no risk of flooding upstream, present no obstacle to navigation, where this exists, and in which the structure itself will not be subjected to hazards from floating objects such as tree trunks, ice and floating debris including excessive floating vegetative or algal growth from weed cutting or natural processes. In addition those installations which extend across the full width of the channel are, unless expense is no object, restricted to the smaller rivers or tributaries of large rivers where the magnitude and variations in discharge and related water depths are not very great.
Self-cleaning scoop traps can be employed in rivers where the water velocities exceed 1.2 m/s and up to depths of 1.2 m, depending on the stream profile, at those times of year when the juvenile fish are migrating to the sea and there is an absence of floating logs. The “Migrant Dipper” can be used in rivers in which the velocities are less than 1.2 m/s when juvenile fish are migrating to the sea. “Merwin Traps” are employed in impoundments and stretches of river where the velocity of flow is minimal.
The removal of juvenile fish from gate-wells is possible only at hydro-electric power stations where these gate-wells exist as an integral part of the water flow control system of the station.
Sequential-switched floodlights can be employed only at the forebays to hydro-electric stations where concentrations of smolts are seen to occur and where the layout of the site makes it a practicable proposition.
Electrical systems can be employed at any site where the nature of the site and conditions otherwise suit.
Fish species or groups which can be sampled: Fish fences and fish weirs - Atlantic salmon (Salmo salar) and their parr and smolts; sea trout (Salmo trutta) and their parr and smolts; adult eels (Anguilla anguilla); brown trout (Salmo trutta); pike (Esox lucius); perch (Perca fluviatilis); Pacific salmon (Oncorhynchus spp.); steelhead trout (Salmo gairdneri). Scoop traps, Migrant Dipper, Merwin traps, Gate-well dipnets - juvenile Pacific salmon (Oncorhynchus spp.) and steelhead trout (Salmo gairdneri). Fish Wheel - adult Pacific salmon (Oncorhynchus spp.). Sequential-switched floodlights - Atlantic salmon (Salmo salar) smolts.
Fish weirs and fish fences: Shearer (see p. 571) uses a fish fence trapping installation to capture upstream and downstream migrating salmon and their young and in particular to estimate the number of juvenile salmon produced in a river by mark and recapture methods. He has also obtained firm evidence of the movement of juvenile salmon to and from the estuary at the pre-smolt stage occurring at times of the year outside that of the normal smolt run. By means of this type of installation information can be obtained on year-to-year changes in timing, composition and relative abundance of the population of Atlantic salmon in individual rivers.
According to Raymond and Collins (see p. 552) scoop nets, Migrant Dippers, Merwin traps and gate-well dipnets are employed at various stations in the Columbia river system to study the effects of turbines, impoundments and other environmental changes arising from the harnessing of this river for the generation of electricity, on the young of the valuable stocks of fish in this river. These traps are employed throughout the downstream migration to capture samples of juvenile Pacific salmon and steelhead trout in their passage to the sea. By a mark and recapture system estimates are made of the populations of fish passing each sampling site based on the recapture of marked fish, identified as to point of release, and information is obtained to assess survival and relative abundance of fish stocks and to determine timing of runs. Information is also obtained on the duration of migration and the timing of populations from various tributaries based on the median of respective migrations as identified by the particular marking.
The effects of dams are assessed on the basis of the results obtained downstream in respect of timing, migration rate and survival after the construction of the dams compared with those obtained before they were built.
Fish weirs and fish fences:
(i) Permanent structures - Those extending across the full width of the channel can have a high capital cost depending on the width of the channel and whether the range of water discharge to be provided for is great or small. Servicing of the installation and in particular ensuring that screens do not become blocked with debris requires the presence of one caretaker full time. At times of fishing the traps, one additional attendant is required to assist the biologist in charge. Working conditions and ease of operation are otherwise quite good and also arrangements can be made to handle the fish in such a way as to cause the minimum stress. Salmon smolts in particular require very careful handling to avoid injury and mortality.
(ii) Temporary structures - Capital cost is not great but there is a continuing expense in erection and dismantling each year. In most cases a boat must be provided to enable the unit to be serviced. A full-time crew of at least three is required for servicing and fishing while gear is in operation. Keeping the netting clear of debris can be a problem in some locations. Conditions for the handling of the fish are not the optimum, and in particular, care is needed in handling smolts.
A large number of both upstream and downstream migrants are made available for counting and inspection; sampling for age determination, length-weight relationships, sex composition and calculation of the composition of runs from various year classes. Where it is possible to trap all upstream and downstream migration, valuable information on spawning escapement, smolt production, recruitment and mortality can be obtained.
Where a portion only of the run can be sampled, care is necessary in the extrapolation of the results of such sampling to the entire population. However, it can be an advantage not to have to handle the entire run when population changes could be a result from such handling. Provided sampling with a constant fishing effort is carried out, satisfactory information on population changes and character can be obtained, but fishing must extend over the entire duration of the run of fish and the gear must be capable of sampling accurately the various sizes of fish encountered. If it is selective rather than representative, estimates of the composition of the run can be subject to large error.
Where the trapping installation does not receive all the flow in the river, such as one located in a bypass channel from the main river, the number of fish captured is related to the discharge of the river (Shearer, see p.571).
Poorly designed and badly operated fish fences can cause adverse effects on anadromous fish populations. A delay in upstream migration can be caused and result in a redistribution of spawning fish below the installation with a danger of causing undersirable overcrowding of limited spawning ground area. This could also be a cause of friction with sport fish angling interests.
(iii) Scoop nets, Migrant Dippers, Merwin traps and Fish Wheels - No information is supplied about cost of the devices but this should not be very great. Information is not available on cost of maintenance but quite clearly a powered boat will be needed, possibly fitted with transport tanks to receive the captured fish. A crew of three at least is probably required for the servicing of each unit or group of units.
According to Raymond and Collins (see p. 552), efficiencies of the following order can be obtained: Scoop nets - 3 to 15% of the juveniles passing the site depending on the river flow; Migrant Dipper - 2.3 to 16.7%; Gate-well dipnets - 25% of the seaward migrants passing through the turbines enter the gate-well at present and are available for sampling. All indications point to this being increased to 80% thus providing a significant number of juveniles for sampling many sites which otherwise could only be provided at very great expense and effort, if at all possible.
Classification: Fish traps - Self-contained portable units that can be fished individually or in groups or chains. This account deals with those known as drum nets, perch traps or “Windermere” perch traps and also with fyke nets, trap nets and wing nets. Other possible units such as wooden baskets and plastic eel traps have been described by Von Brandt (1971).
Design and construction: Perch traps generally consist of a skeleton framework covered with wire mesh, one end of the unit having a re-entrant funnel. It is essential for the efficiency of the trap for perch that it be painted black, especially during the spawning season.
Fyke nets, also called double traps, consist of two fyke nets facing each other and connected from entrance to entrance by a “leader” net about 3 m in length. The “leader” netting is held upright in the water by means of a few floats and firm contact is maintained between the bottom of the net and the ground by a weighted foot rope.
The fyke nets used by Leopold et al. (see p. 519) can be fitted with wings, leaders and barriers and fished as single or double traps.
Wing nets are usually a combination of three or four traps connected by wings and leaders. Nagieć (see p.566) describes a trap net of rectangular base semi-circular in cross-section where each end forms the entrance to a funnel projecting into the trap. The trap net is ballasted and placed at dusk on bottoms from 1 to 5 m deep and removed again at the following dawn.
Mode of action: It is passive for all units referred to, which remain stationary while in use and are fished unbaited.
Conditions in which gear can be used: Perch traps are usually employed in lakes each with its own rope and marker float.
Mann (see p. 194) has employed them in rivers of rapid flow and dense growth of aquatic macrophytes, and Craig (see p.502) finds that they are the only sampling gear that can be used in areas of reed swamp.
When fishing for perch in lakes, traps are used during the spawning season, from the beginning of April until the end of October. In Ireland at the end of the perch spawning they are transferred to other areas in the lakes and employed for the capture of small pike. Here the traps are placed initially in depths of water of 8 m and are gradually moved into shallower waters until the 5 m depth is reached at which they remain until the end of April, by which time spawning usually is completed.
The most suitable type of bottom for the placing of these traps when fishing for perch in the spawning season is one that is covered with thick-stemmed vegetation or brushwood.
Craig and Le Cren et al. recommend that for population monitoring perch traps be employed over many years in fleets of five traps placed in four different places.
Fyke nets are used in Polish waters from spring until autumn in shallow areas. An area is cleared for each trap which is moved regularly to maintain its catching efficiency. The nets can be fished over a great variety of bottoms and can even be used under ice in winter. In Ireland fyke net units are set in trains laid in line with the direction of the current.
Wing nets are used throughout the year so long as the water is ice-free. The nets are usually set in shallow areas overgrown with submerged vegetation, and may be connected together to form complex barriers.
Fish species or groups which can be sampled:
Perch traps catch mainly perch of spawning size with a tendency to select males. Pike may also be caught in this gear.
Fyke nets - in rivers the nets catch eels and also smaller quantities of pike, rudd, gudgeon, roach and dace in fresh waters and flounder, mullet and bass in estuaries. In lakes the nets catch pike, tench, roach and bream.
Wing nets catch mainly eels but also pike and roach.
Trap nets catch small species including ruff and young pike-perch and perch.
Perch traps are used widely for the capture of breeding perch which serve as the basis for tag-recapture experiments. Multiple census methods are, however, suspect using traps as fish tend to lose interest in them toward the end of the breeding season (Jensen, see p.600). Furthermore, perch are susceptible to stress and handling, which will give rise to wide variations in numbers estimated by this method (Craig, see p.502). Pike and perch samples from traps have, however, been used for estimates of total population numbers over a number of years using catch-per-unit effort statistics (Le Cren et al., see p.59).
Fyke nats set in rivers are one of the few effective means of obtaining reliable estimates of population densities. Similarly, in lakes the nets may be used both for qualitative and quantitative information on a number of species.
Perch traps are cheap to make and modest in manpower requirements as regards servicing and use. They are, however, bulky to transport, particularly if large numbers are to be dealt with.
In a lake of any size it is necessary to have a boat with an outboard engine and crewed by not less than two men. These men should be capable of servicing each day 100 perch traps distributed over a length of 2 km but 150 traps if more densely placed.
All authors agree that perch traps are very susceptible to interference by the public; traps can also interfere with sport angling for trcut. A major disadvantage is the great variability in catch numbers, and to show significant differences between two years or locations the means should be based on the catch of not less than ten traps. There are also marked seasonal variations in the catch which depends on the behaviour of the fish, and considerable sex selection correlated with season and the analysis of catches to provide population estimates has presented many problems. Traps are probably of little value in short- term mark-recapture experiments except as a method of catching fish for initial marking.
No other method of sampling for perch has, however, proved more effective than perch traps and they are very satisfactory for the large-scale removal of perch.
In general, perch traps should not be used as a method of sampling fish populations due to their limitations. Stott (1970) and Bagenal (1972) criticize the use of traps for catching fish and using the fish for estimating population parameters since the catch depends on the behaviour of the fish but they may be the only gear that can be used providing their limitations be fully understood.
Fyke nets and wing nets are probably not of great value for sampling for similar reasons to those given for perch traps. In addition the labour requirements of fyke nets are particularly high and they are used only reluctantly by fishermen.
