| Convener: | E.D. Le Cren |
| Panel Members: | J.S. Alabaster |
| T. Backiel | |
| H.A. Regier | |
| D.E. van Drimmelen |
(a) Objectives
It has become axiomatic that to be successful in research one must be able to ask the right questions of nature and it is equally true that questions should be framed in such a way that the answers obtained are relevant and unambiguous. Therefore, prior to any consideration of particular methodologies for survey, monitoring or appraisal of stocks, it is desirable to look at the general framework within which the activities are carried out, and particularly the objectives for which populations of fish are sampled.
There is still considerable ambiguity in the definition of terms relating to sampling, and while it is not practical here to consider the standardization of such terms, it is desirable that this be done at some time in the future. For the purpose of this meeting at least, survey can be defined as an activity which seeks to establish a spatial distribution at a given point in time, and monitoring is the investigation of changes with time at a particular point in space; appraisal is the evaluation of the resource in terms of some predefined value or purpose.
Fish or populations of fish are in fact sampled for a variety of reasons within these broad categories, of which the following may be regarded as representative:
To assess the availability of stocks for commercial exploitation and to appraise their magnitude and suitability;
To monitor changes in stocks resulting from environmental variations induced either by natural events, or by human interventions as pollution, construction of dams, etc., where the fish serve as indicator organisms and are not necessarily considered for their own sake;
To monitor the status of stocks and their response to a fishery or to forecast catches for a fishery;
To assess the results of management practices; for instance to evaluate remedial measures taken to control pollution or the improvement of stocks due to stocking programmes;
To carry out basic research on the fish themselves or their communities and populations and, where fish or fisheries exist in the presence of pollution, to formulate water quality requirements for this resource.
The choice of objective or objectives will ultimately determine the characteristics of the sampling programme including sampling locality and the gear selected. In some cases it may be desirable that sampling programme be directed at more than one objective provided that the validity or relevance of the data thus obtained is not thereby jeopardised.
Sampling programmes may often be user-oriented and determined either by socio-economic factors (in for instance sport fisheries) or by the needs of managerial decision-making. This may impose particular restrictions, for example on the time limits for obtaining an answer, and thus the type and accuracy of sampling methods which can be used or alternatively traditional sampling activities may be inadequate. In particular, survey, monitoring and appraisal as understood here cannot adequately describe events falling outside the normal experience of the system for which the programme is designed and for these we would have to rely on experimentally derived data.
(b) Scope
To fulfil such objectives a range of information may be needed which varies in precision and detail and which thereby influences the design of the sampling experiment. It may, for instance, suffice for some purpose to determine simply the presence or absence of a species, though the latter would not be easy. More generally exact determinations of the relative or absolute abundance of fish communities or their component species may be needed or detailed investigations of the biology of the fish themselves may be necessary including such factors as behavioural changes, health, presence or absence of parasites or even the acceptability of edible fish products to man or other consumers.
(c) Strategy of Sampling
Fisheries show a great range in complexity both with respect to the ecosystem, and the fish communities, and it is helpful from the planning point of view to look at those elements that are common to all systems, rather than at the differences between them.
These elements are:
Any of these elements may be the subject of surveys, monitoring and/or appraisal and it is suggested that the “strategy” of sampling is the process of allocating effort among those components or their diverse interactions. Thus in certain cases studies may be confined to one component, for example, the environmental as in the case of monitoring the effects of pollution, or in others may even be limited to one species, as in the case of biological investigation. Alternatively several interacting components have to be considered and here factors external to the system such as differences in sociological, economic and personal demands may impose differences in emphasis. Consequently the particular relationships between the component to be examined in the case of, for example, sport fisheries may be very different from those chosen among the same components in the case of commercial fisheries.
Strategy, therefore, is the broad planning of the sampling programme to fulfil its objectives. The more detailed logistics of the sampling programmes can be described by a system of six components as follows:
Within these components interactions represented by two arrows imply that a decision is reached iteratively after considering various alternatives. Very often the choice of gear and the design of the sampling experiment depend on the available money and manpower, and through the iterative process may ultimately lead to the modification of the scope of the objectives. In more extreme instances where the objectives depend on some success criteria considerations of cost may lead to the abandonment of the exercise if the criteria obviously cannot be met with the material means available.
One important component of the sampling design is the validation and calibration of the gear. This is often overlooked but for valid results the gear should be calibrated in terms of its efficiency. The criterion chosen for such calibration may vary depending on the accuracy and scope of the result desired and may even be interpreted as a percentage capture of the fish present. This implies that the performance of any gear should be compared with that of other gears sampling the same stock. Such calibration is difficult for absolute methods and supposes a complete understanding of the behaviour of at least one gear in the environment being sampled. For relative methods it may be assumed that if two or more gears show similar trends in variation the results are likely to be correct, whereas if differences in the performance of the gears are apparent both may be wrong and the sampling thus invalidated.
| Convener: | A. v.Brandt |
| Panel Members: | R. Steinberg |
| Kj.W. Jensen | |
| N. Bacalbasa-Dobrovici | |
| C. Kipling |
Documentation: EIFAC/SC-I-Symp. 63 [Panel Review]
The most important netting techniques used for sampling of fish populations are:
Dragged gear (trawls)
Seine nets
Surrounding nets (purse seines, etc.)
Gill nets
Entangling nets (trammel nets, etc.)
Other netting techniques such as bag nets, scoop nets and frame nets have been used occasionally for sampling, but very little is known about the efficiency or usefulness of such gears.
In considering the netting techniques detailed below, it must be emphasized that it is impossible to recommend any one gear for all sampling purposes. Furthermore, several types of gear should be used wherever possible.
(a) Trawls - Documentation: EIFAC/SC-I-Symp.5, 28, 34
Although originally marine trawls proved too cumbersome for general use in freshwaters, improved forms, such as two panel bottom trawls and four panel mid-water trawls, are now used for fishing not only in large lakes but also in smaller ones. The use of otter-boards should, however, be avoided wherever possible as these increase towing resistance and require much larger motors. The alternative of using two boat trawls has been successfully adopted in several areas.
The use of bottom trawls is limited by the nature of the substrate, and the gear is in general not effective in too swift a current. The method is therefore useful for sampling in lakes and rivers with low flow. Providing adequate information on the factors influencing the catching efficiency and selectivity is available, the method may be used for absolute estimates.
(b) Seine nets - Documentation: EIFAC/SC-I-Symp.4, 6, 7, 14, 24, 29, 33, 37, 43, 44
Seine nets depend on the existence of clear bottoms, usually at beaches, and are easily rendered inefficient by the presence of submerged vegetation. They can, however, be operated under ice and are then very efficient. In comparison with other methods the fish are caught relatively unharmed and may therefore be used for mark-recapture exercises. Some attempts have been made at obtaining absolute estimates of abundance from seines but there may be considerable escapage particularly of large fish during hauling and for this reason trammel nets are often set parallel to the line of the haul.
(c) Purse Seine - Documentation: EIFAC/SC-I-Symp. 24, 51, 53
Purse seines can be successfully operated in large, deep lakes for the capture of pelagic species. The method is, however, expensive and usually confined to commercial scale operations.
(d) Gill nets - Documentation: EIFAC/SC-I-Symp. 1, 4, 6, 9, 14, 23, 24, 30, 37, 41
Because of their cheapness and simplicity of operation, gill nets are widely used for sampling fish populations. However, the high selectivity of the gear may create problems if it is not taken into account. If a fleet of gill nets, containing a variety of mesh sizes is used, fish of a wide range of lengths can be caught, and the different selectivity curves may be combined with other catch data to give estimates for a number of parameters. Gill nets are a passive gear, and their fishing efficiency depends largely on the behaviour of the fish. Catching efficiency thus depends to a great degree on the type of netting material used; its visibility to the fish and its resistence to penetration. Because of these problems great care is needed in drawing conclusions from catches with this gear, and in any case only relative information can be obtained.
(e) Entangling Nets - Documentation: EIFAC/SC-I-Symp. 9, 31, 73
Trammel nets and frame nets are of interest for sampling fish populations, and trammel nets particularly have been used for this purpose for a number of years. Because they are relatively unselective they are suitable for sampling a wide range of fish populations. Entangling nets can be used as either passive or active gear although operating costs are higher than those of gill nets. Frame nets, a new gear for Europe, offer possibilities for the future although they have only been used on a limited scale until now.
| Convener: | P. Lamarque |
| Panel Members: | E. Halsband |
| W.G. Hartley | |
| A. Lelek | |
| P. Sharkey |
Documentation: EIFAC/SC-I-Symp. 65 [Panel Review] 3, 4, 10, 13, 38, 39, 46, 47, 50, 68, 69, 71, 74
(a) Description
Electric fishing should be considered as an auxiliary to traditional fishing gears, which considerably extends their effectiveness. Among the various possible combinations, electrified dipnets, trawls, townets, and guiding screens are known. Electricity attracts fish either from open waters, or from concealment in vegetation or the bottom, so that they may be caught by dipnets or trawls. Fish may equally be guided into traps, or immobilized while they are washed into a stopnet by water flow.
(b) Advantages and Possible Applications
The yield from traditional gear can be considerably increased by electrification. For example, the electrification of a trawl can improve its performance between four and thirty times compared with catches by the non-electrified gear. The electrification of fyke nets set in a river effectively increases their size by a factor of three. Electric screens guiding migratory fish into traps can attain 80 percent efficiency.
The electrification of fishing gear allows fish to be caught which cannot be otherwise captured, or only captured at particular times of the year. Because of the increased efficiency of the gear, or the smaller dimensions needed for the equivalent yield, the cost of fishing is usually considerably reduced.
If the current is properly regulated, fish are caught unharmed and may be returned to the water after examination or tagging.
(c) Disadvantages and Limits of Utility
Electrified gear shows a certain selectivity both for species and for fish of different size within the same species, due to their greater or lesser susceptibility to the field. Larger fish are usually more easily captured than smaller ones.
The catchability of fish tends to diminish with successive electric shocks. This makes it necessary to exercise extreme caution when a constant fishing efficiency is required, as for example in de Lury estimations.
Fishing efficiency can also vary with certain parameters of the environment, such as conductivity, temperature, water turbidity or clarity, water flow, etc.
The use of electrified dipnets is restricted to waters under 2–3 m in depth, but electric trawling is practicable in all waters normally trawled, irrespective of depth.
At the existing state of generator design, the use of electrified hand nets is limited to waters of maximum conductivity 10 000 us/cm, and that of trawls to waters of up to 3 000 us/cm. This technological problem is, however, capable of solution by using new square wave pulsed currents.
For optimum results, the technique and type of current employed should be adapted to the local conditions, which include the limitations of the apparatus available, the electric field and their action on the fish.
(d) Ease of Interpretation of Results
The validity of the results of sampling depends largely on the size of the sample with respect to the size of the stock. This is, among other things, a function of fishing efficiency which depends on the effectiveness of the machine and the duration of fishing.
The electrified dipnet has for some time made it possible to estimate the population by the Petersen and de Lury methods, mainly in trout streams where the efficiency of capture (proportion of fish caught relative to stock) frequently lies between 30 and 80 percent. This order of magnitude can also be obtained with other species which inhabit waters fishable by the electrified dipnet.
The efficiency of an electrified trawl, recently used in a 15 ha lake 2 m deep, reached 6 percent as estimated by the Petersen method. This efficiency increased +20 percent for a 16 hour fishing period, permitting an estimation of the stock to be made within confidence limits of ± 10 percent.
| Convener: | A.V. Holden |
| Panel Members: | P. Bauman |
| G. Hall | |
| W.D. Davies | |
| O. Sumari |
Documentation: EIFAC/SC-I-Symp.72 [Panel Review], 2, 18, 26, 35, 40, 58, 62
(a) Mode of Operation
Although about 40 chemicals have been used for the destruction of fish populations, only two (rotenone and antimycin) appear to have been employed in fish sampling tehniques. Antimycin is more toxic than rotenone, has a greater variability in sensitivity among species, but is not avoided by fish. These properties could make it more useful to fishery biologists than rotenone, but the technique requires more development and most experience has been obtained with rotenone (in the form of a derris extract or emulsion).
Poisons are used in areas delineated by barrier nets, across rivers or coves in a lake, or set out from a lake shore. In rivers, it is usually necessary to treat the water below the downstream net with potassium permanganate to destroy the toxicant and prevent any mortality outside the area. The concentrations of both toxicant and detoxifying chemical must be calculated carefully.
(b) Advantages and Possible Applications
The following information can be obtained by the method.
Estimation of standing stock
The accuracy will depend on the efficiency of collection of dead fish from the surface and bottom (using scuba divers). Fry and juveniles are most difficult to collect, and this may result in a serious underestimate of numbers, though a smaller error in biomass. A reasonable estimate of the standing stock in a lake can be obtained by cove or open-shore sampling.
Appraisal of species composition
If a sufficient concentration of poison is used all species are usually killed, and estimates of the relative proportions (by numbers or weight) can be calculated.
Presence of species
Useful in producing a faunal list, or checking on the survival of introduced species.
Relative year-class strengths
Gives information on year groups, success of reproduction, survival of first-year class, decline of a dominant age-group and on mortality rates.
Balance of groups of species
Can indicate balance between forage species and carnivores, or game fish and coarse or trash fish.
Check on other methods
May be used to check information from netting or trapping techniques.
When no prior information is available on the existence of fish, poisoning will provide some information quickly, although it should be remembered that seasonal behaviour may result in a species being absent from shallow water areas at the time of poisoning.
The cost of the technique may be low relative to that of other methods and should be considered for initial surveys, although the use of a detoxifying agent significantly increases the cost. Poisons may be unsuitable in waters used for public supply purposes, and their use may be discouraged by wildlife authorities concerned with the welfare of birds or other species.
(c) Disadvantages and Limits of Utility
Among the disadvantages should be mentioned the toxicity of rotenone and antimycin to several species of invertebrates, and of antimycin to fish eggs, although such disadvantages are relatively unimportant when the area poisoned is only a small fraction of the entire water body. Poisons are rarely species-selective, and not size-selective. They are not very suitable in deep or fast-flowing water. Fish retrieval is difficult in turbid water or weed beds and juveniles may be lost in sediment. The introduction of marked fish prior to poisoning may enable a correction to be made for fish not retrieved.
The method cannot be used to obtain information on stomach contents, as some fish disgorge their stomachs when poisoned, while large fish may eat the dying smaller fish before being affected themselves. In warm waters kills are rapid, but accelerated decomposition of fish on the bottom may lead to an underestimate of the total number of fish killed.
(d) Need for Further Work
Little work has been done on the use of poisons for sampling, but the technique seems capable of further development, both in the identification of chemicals with species-selective properties and in the limitation of poisons to pre-determined depths. More efficient collection methods and the calculation of percentage retrieval for each age-group by mark-and-recapture techniques, should also be investigated to improve the accuracy of population estimation.
| Convener: | O.A. Mathisen |
| Panel Members: | H. Braithwaite |
| K. Johannesson | |
| S.T. Forbes |
Documentation: EIFAC/SC-I-Symp. 76 [Panel Review], 17, 27, 28, 34, 37, 54, 66, 67
Acoustic methods have developed to a point where quantitative estimates of ichthyomass can be made with quite good precision and accuracy in waters of more than 10 m depth, provided that the fish echoes are sufficiently distinct from those of the bottom (pelagic and semi-pelagic stocks). Quite detailed distributional maps may be made of the abundance of open water stocks in the larger lakes using this method. In shallow waters of both lakes and rivers, methodology is less well defined, but sounders operating in a horizontal mode present possibilities for future development. For migrating fish, systems processing all signals received from individual fish must in most cases be based upon characteristics of the fish itself, such as its direction of movement, or possibly its tail beat, using the döppler effect.
Signal processing methods include visual inspection and measurements from standard echograms, electronic integration of signals corrected for losses associated with different fish depths, and computer analysis of signals as recorded on magnetic tape. The latter permits separate estimates for various depth strata and for several levels of amplitude.
Direct calculation of the strengths of echoes to be expected for different quantities of fish has not yet proved satisfactory as a basis for calibration. A comparison of counts of individual targets with integrator readings may be compared, in certain circumstances, with biomass estimates obtained by trawling or other non-acoustic gears. In this connexion, it may prove possible to calibrate simple echo-sounders against more complex systems for routine surveys. At present the high capital costs of the equipment needed to produce biomass estimates tend to restrict the use of this gear to calibration of other methods of estimating abundance on occasional reference surveys. However, disregarding capital investment, acoustic surveys typically only cost about 15 percent of a trawl survey in the marine environment and are three times as efficient. The volume sampled may be 50 times greater with the acoustic survey and since variance is approximately proportional to volume sampled, the variance of estimates should be improved to the same extent. It may also be expected that the cost of acoustic surveys will drop substantially with advances in miniaturization, better knowledge of sampling requirements for given levels of precision and the use of shore based computers rather than on board integrators to process the data acquired.
| Convener: | W.G. Hartley |
| Panel Members: | D. Simpson |
| L. Stewart | |
| P. Sharkey |
Documentation: EIFAC/SC-I-Symp. 77 [Panel Review], 11, 12, 15, 16, 17, 21, 28, 36, 58, 64, 67, 70
Automatic counters may be used to advantage to provide total counts of passing individuals where the species to be sampled are forced by habit or by artificial structures to pass through confined channels. Fish may be counted using the electrical bridge principle with a group of three electrodes, but only if the hydraulic conditions at the site are constant or if they change steadily. It is necessary to check the accuracy of the record from such counters by visual means, or by television records or flash photography triggered by the passage of fish. In this way too individuals may be separated by species.
Inaccuracy of counting does not arise from difficulties in detecting fish but from failure to provide conditions compatible with fish behaviour, and from electrical interference. In general, high accuracy of counting is not as important as stability in the system, as erratic counts cannot be corrected for error.
It is essential that the electrodes always be submerged and this may involve design difficulties in the case of deep channel counters, or free water sites where waves occur. A tube counter, with electrodes in the tunnel, is the most consistent and reliable arrangement in this respect, and can also give approximate information on fish size. Tubes are, however, liable to blockage and fish may in some cases be delayed in the tube thus giving false results by the multiple recording of the same fish. Successful counting has also been achieved at Crump gauging weirs where the depth of flow is relatively constant and such that the fish are induced to swim close to electrodes attached to the face. Sonic counters are also being developed for use in larger channels but are significantly more complex in design.
Shielding from electrical noise is essential for reliable operation and areas of high electrical background noise level should be avoided although it is usually possible to shield an area sufficiently large for the counter in bad conditions.
Automatic counters have proved particularly effective in evaluating diurnal changes in the passage of fish and in reducing errors resulting from fatigue during long periods of visual monitoring. Multiple installation along river or stream systems are to be particularly recommended both to improve accuracy and, if such information is desired, to increase the chance of detecting the effects of changes in flow or water quality or the movement of fish.
| Convener: | C.J. McGrath |
| Panel Members: | M. Leopold |
| D. Piggins | |
| C.P. Ruggles | |
| I.R.H. Allan |
Documentation: EIFAC/SC-I-Symp. 78 [Panel Review], 2, 4, 7, 11, 13, 14, 19, 20, 21, 24, 25, 32, 48, 56, 57, 58, 68
(a) Fish counting installations in fish passes
The principal advantage of fish passes is that all fish wishing to move upstream must go through the installation. The entire run is thereby made available for counting by the means described in Panel 4b (Section 5.4.2). All migratory fish may be monitored in this way and where the numbers of fish are not too great the fish pass may be so designed that individual fish can be removed and inspected individually with little or no damage. A very high degree of accuracy of count can be achieved which can vary from totality for rivers in which the number of migrants is small to a somewhat lesser proportion where the number of fish to be counted is large. Labour requirements are generally very low in suitably designed installations.
(b) Traps for anadromous fish and their young
Where site conditions permit the construction of a fish barrier across the full width of the river a complete count of the upstream and downstream migration can be obtained. Where only part of the channel can be covered by the barrier, it is obviously possible to sample only a portion of the run and great care is needed in extrapolating the results to obtain estimates of the total size of the run. Floating trap units have also been used successfully to sample runs of downstream migrants.
The use of devices employing fish fences extending across the full width of the river channel is restricted to smaller river systems and the tributaries of large rivers because of the cost involved. However, all diadromous fish can be sampled in systems using full width traps and the fish are relatively unharmed and may thus be used for mark-recapture experiments.
(c) Self-contained portable traps
Methods of catching fish involving the use of portable traps with or without wings of netting are used commercially in many parts of the world. They are cheap to make and modest in manpower requirement; however, they are susceptible to interference from the public and may be objected to by sport fishermen. The gear can be used in almost all conditions of weather and depth, and indeed in places where no other gear is usable. They therefore constitute a valuable method for sampling fish, for instance in weed grown areas. Many lake-dwelling fish can be caught in such traps and they can also be used in rivers. They can be highly selective for species and even for sex where differences occur in the movement patterns. As most traps are constructed of wire or yarn netting, they are selective for size depending on the mesh size used. The reliability of estimates of total population using trap catches is questioned by many scientists, mainly because the biological variables controlling movement into traps is poorly understood. However, it is the experience in Poland that wing nets provide a very satisfactory measure of the density of all stocks in lakes not only as regards relative estimation of the density but also in respect of absolute numbers.
(d) Conclusions
When considering the role of traps for the survey, monitoring and appraisal of fishery resources emphasis should be placed on:
The need for adequate preliminary planning of the experiment;
Avoiding creating an artificial situation whereby the presence of the trap interferes excessively with the behaviour of the fish;
Considerable care in handling the fish and release of upstream migrants, particularly where valuable sport fishing interests exist on a trapped river;
The development of a sampling programme that will permit avoidance of interference with total population;
Proper understanding of the limitations of portable trapping units for assessment work while recognizing their usefulness as a supplementary method particularly where site conditions do not permit the use of other methods.
| Convener: | K. Tiews |
| Panel Members: | W. Nellen |
| T. Bagenal |
Documentation: EIFAC/SC-I-Symp. 59 [Panel Review], 18, 24, 28, 35, 48, 53, 57, 58
Eggs, larvae and juveniles pose special sampling problems but may be considered important as there may be differential responses to pollutants at different phases in the life history. Furthermore, it has been estimated that 50 percent of the production of ichthyomass is during the juvenile stages. While some of the methods discussed in Sections 5.2 to 5.5, such as poisoning and electrofishing, can be adapted for use in catching juveniles and even larvae, special netting devices are more commonly employed. Few methods so far utilized appear to have general applicability for quantitative sampling.
Among the methods described for larvaè and juveniles are several nets designed to rise to the surface, capturing the fish from a known volume of water or area of surface. It appears that this technique usually reduces net avoidance. Dip nets, purse seines, and tow nets have equally been successful both with and without light attraction. Larval traps seem not to be consistent in performance, their use being recommended by some workers and not by others. Devices which take advantage of special patterns of behaviour such as seeking cover or congregation under lights are often successful.
Few freshwater species produce pelagic eggs and a wide range of habitats are chosen for spawning by different species of fish. In most instances, therefore, sampling methods for estimating abundance must be tailored to the particular habits of the species in question. Several methods are available to separate eggs from bottom materials by hydrostatic or hydraulic devices permitting quantitative collection. The abundance of adhesive eggs can sometimes be estimated by a two stage process in which the quantity of substrate is first estimated and secondly the quantity of eggs per unit of substrate. Alternatively, spawn collectors can sometimes be devised which are placed in the habitat prior to spawning. Direct observation and counting is frequently the best choice though divers may be required.
It is generally agreed that further research on identification of eggs and larvae or on their quantitative sampling is greatly needed, owing both to the special difficulties encountered and the importance of knowledge of these stages to the evaluation of the effects of environmental changes on fish populations.
| Convener: | J.E. Thorpe |
| Panel Members: | C. Kipling |
| H. Koops | |
| V. Steiner | |
| K. Jensen |
Documentation: EIFAC/SC-I-Symp. 79 [Panel Review], 1, 2, 7, 22, 24, 25, 42, 43, 44, 45, 52, 58, 60, 68
The mark-recapture method for estimating the size (N) of a population of fishes consists of releasing a known number of marked individuals and recording recaptures in a subsequent sample. It is assumed that the fraction of the marked population recovered in the sample is equal to a fraction of the total population sampled.
Two general conditions for the validity of estimates are that:
Either the marking or the subsequent sampling should be at random with respect to the population being estimated;
Marking should not alter the behaviour or the life expectancy of the fish in any way.
In practice these ideals are rarely realized and supplementary experiments are necessary to establish the degree of departure from these conditions. Sufficient sample size may also be difficult to achieve in one sampling operation so various methods of multiple sampling have been devised. Estimation theory is well developed making it possible to specify statistically efficient estimators for a wide variety of modifications, and in special circumstances to judge their precision. A variety of marks are available and the investigator should take care to select the marking method best adapted to his objectives.
The advantages of the method are that the precision of estimates may be predetermined, and sample sizes selected accordingly. The recapture gear is calibrated simultaneously as an instrument measuring abundance, and simpler, quicker and cheaper determinations of relative abundance are possible thereafter. Information on growth movements and mortality is also obtained.
The method is, however, additional to one or more methods of capture and is, therefore, more expensive that these. Several supplementary experiments are necessary to check for non-randomness, differential mortality, recruitment, tag-losses and incomplete reporting of tags. It is also restricted to cohorts of populations of single species.
Owing to the comparatively high cost of well designed experiments, the method is perhaps most useful as a means of validating the gears used for sampling and estimating their efficiency.
| Convener: | R. Lloyd |
| Panel Members: | T. Northcote |
| K. Tiews | |
| E.D. Le Cren | |
| J.A. Gulland |
(a) Methods of catching or sampling fish which were not included in Panels 2–7
Several important methods of catching and sampling fish populations had not been included in the papers submitted to the Symposium and were therefore not covered in the Panel Reviews; the Symposium participants put forward the following methods as worthy of consideration for this purpose:
Netting methods
stow nets, which are used with a single otter board in rivers where the water current is strong
lift nets
push nets (small trawls attached forward of the bow of the boat, with the cod end accessible in a well amidships)
verandah traps for trapping leaping fish
various types of stake nets.
Angling techniques
rod and line
long lining
spearing, harpooning, shooting.
Visual methods
direct observation by scuba divers
aerial photography
underwater photography television.
Other methods
use of explosives to stun fish, sometimes in conjunction with netting and sonar techniques
sampling salmonid redds by hydraulic flushing or total freezing
suction methods for sampling shoaling fish.
It was agreed that participants would inform the EIFAC Secretariat of any other methods about which they had information on the respective merits and disadvantages (see Section 4, Recommendation 74/5).
(b) Synthesis of information presented to the Symposium
During a general discussion on the ways by which the information collected by the Symposium could be collated into a simplified form, it was suggested that one method would be to prepare a set of tables which would assist fishery scientists in selecting the most effective gear, gears or techniques for the specific problem to be solved. It was agreed, however, that the number of variables associated with the physical parameters of rivers and lakes, and with the species to be sampled, would make the preparation of such tables difficult although if such a system could be devised, it would be of great value.
Accordingly it was recommended that a Working Party should be set up to prepare a digest of all the various techniques covered by the Symposium to form a manual similar to those prepared by FAO in other fields. It was also considered that the employment of a consultant would be valuable in this respect. Such a manual should not aim solely at indicating a single technique which could be used in a specific situation but a whole range of techniques which could be used singly or in combination (see Section 4, Recommendation 74/7).
(c) Publication of the Symposium papers
It is felt that the individual papers presented to the Symposium formed a valuable collection of information which should be published in some form or another. It was pointed out by the Secretariat that this could be best achieved if the papers were first reduced in size and/or number by an editorial panel, in order to reduce the cost and size of the publication (see Section 4, Recommendation 74/6).
(d) Future research needs
It was clear from the papers presented that some types of aquatic systems, such as small streams, small lakes, and open waters of large lakes presented few problems and sampling techniques for migratory salmonids have been well defined. However, as the size of the water body increases, there is a concomitant increase in ecological complexity as found in the marginal areas of large lakes and in large rivers. In general, it was concluded that there was a lack of suitable methods for quantitative sampling of such areas, particularly where the substrate was obstructed by vegetation, boulders or other hazards, or where current velocity was excessive. There was, therefore, a need to expand existing methods to sample such situations and to seek new techniques perhaps novel in their approach. Such a research programme woudd necessarily be accompanied by a more thorough investigation into the general ecology of these types of habitat.
In applying methods such as sampling of catches made by commercial and recreational fishermen, including the recovery of marked fish, there remain problems in obtaining the cooperation of the individuals involved, and a greater emphasis should be placed on educating the public on the need for fishery research; also more account should be taken of the social and economic aspects of both sport and commercial fisheries and fishermen in order to obtain the maximum assistance from these groups. This may also help to facilitate the provision of necessary financial assistance for the new research programme outlined above. It is unlikely that these problems are capable of rapid solution. Progress is most likely to be achieved as a result of cumulative experience using methods of observation.
