The FAO Fisheries Department has long concerned itself with problems of inland fisheries. In the past the main areas of interest have been natural lakes and the many reservoirs which were created in Africa and Asia in the sixties and early seventies. Little attention has been paid to rivers and it is only recently that a general awareness has developed of the importance of river systems to fisheries throughout the world. While rivers were relatively unaltered, the fish they produced was taken for granted, but with increasing competition for water for agricultural, industrial and domestic uses, fisheries usually have been given a secondary role among the range of users. The river systems themselves are being increasingly modified both in their form and in their flood characteristics. As a consequence, fish populations are being altered and productive capacities lowered. FAO is now called upon to advise countries on the ways in which such fisheries can be managed, both on their own and in harmony with such activities as irrigation, flood control or power generation. In trying to fulfil this task, we are limited by the comparative lack of knowledge about the systems under consideration. A meeting of experts was convened in an attempt to define the methodologies for the assessment of fishery resources in rivers and the use of these methodologies in evaluating the impacts of fisheries and other uses of the aquatic environment on the fishery.
The following took part in the consultation:
| Prof. J.B.E. Awachie Head, Department of Zoology and Leader Hydrobiology/Fishery Research Unit University of Nigeria Nsukka, NIGERIA Mr. P. Bayley INPA Caixa Postal 478 Manaus, A.M. 69.000, BRAZIL Dr. J. Holčik Laboratory of Fishery Research Slovak Academy of Agricultural Sciences Drienova 5 Bratislava, CZECHOSLOVAKIA Dr. V.R. Pantulu Chief, Environment Unit Mekong Secretariat c/o ESCAP, United Nations Building Bangkok 2, THAILAND Mr. T. Scudder California Institute of Technology Division of Humanities and Social Sciences 1201 East California Blvd. Pasadena, California 91109, U.S.A. |
Dr. W. Q.-B. West Regional Fisheries Officer FAO Regional Office for Africa P.O. Box 1628 Accra, GHANA Dr. R.L. Welcomme (Technical Secretary) Senior Fishery Resources Officer FAO, Rome Dr. H.F. Henderson Chief, Inland Water Resources and Aquaculture Service FAO, Rome Dr. J. Kapetsky Fishery Resources Officer FAO, Rome Mr. G.P. Bazigos Fishery Resources Officer FAO, Rome Mr. M. Ben-Yami Fishery Industry Officer FAO, Rome |
The meeting was opened by Dr. H. Kasahara, Director of the Fishery Resources and Environment Division, who welcomed the delegates and gave a short background statement describing the interest of FAO in river fisheries.
The Provisional Agenda was adopted without modification and it was decided that the sessions would be chaired by separate individuals according to their specialization in terms of the subject under discussion.
(The limits of limnological theory and approaches as applied to river floodplain systems and their fish production - see page 23)
Any classification of rivers must take into account the characteristics of the flood. Some large rivers are of the reservoir type, with even flows throughout the year, but most show some seasonal fluctuations in water level. It is rivers in this latter category which support major fisheries and which, by reason of their instability, differ widely from lakes in their ecology.
Large flood river systems are generally thought of as having two major components - the floodplain and the river channel, which have sometimes been treated separately by some hydrobiologists. This division seems to be artificial and the view that the two components are interdependent parts of a wider complex system was generally accepted. Within these two components there are, of course, sub-systems which are extremely diverse, for example, in the Mekong and the Ganges rivers, where there are long and deep depressions within the main channel which form transient chains of lakes in the dry season. Similarly, part of the floodplain may be occupied by large lakes such as the Grand Lac of the Mekong, which are more or less permanent features of the geography. In more developed river systems such as the Danube, Thames and others, hydraulic works tend to alter the characteristics of the ecosystems, usually in such a way as to convert flood rivers into reservoir rivers. Some division can also be drawn between forest rivers as opposed to savanna rivers. It is clear that there is, as yet, insufficient information to allow good comparisons to be made between the biology and productivity of each of these types. An essential feature of any river system is the nature of the input upon which all subsequent productivity patterns are based. In the upper reaches these are mainly of allochthonous origin, but on the floodplains of the potamon reaches the major inputs are silt and dissolved nutrients. It is not yet clear as to how these inputs are related to energy flow and productivity patterns within floodplain systems and further studies are needed both on savanna and forest floodplains as to the nutrient pathways and energy flow patterns. Such aspects are of particular importance when considering devices for the management of rivers which involve the prevention of silt reaching the floodplain, or cropping by the gross removal of nutrients. The contribution of other forms of nutrient input, such as windborne dust or dung from land animals, is even less completely understood.
It is obvious from most floodplains that the bulk of primary production is located in higher vegetation. In forest floodplains this is apt to take the form of long-lived structures such as trees and vines, whereas in savanna floodplains the major plant forms are floating or emergent annual macrophytes. These differences may be fundamental to the energy balance, as in the case of the savanna plain the annual cycle of growth and death of the vegetation makes the recycling of nutrients very much faster. In many forest floodplains the accumulated biomass may be very high even though the nutrient content of the water is low and such systems may be considered to be accumulations of nutrients (i.e., nutrient pools or sinks) which have often taken considerable periods to build up. Cropping of such systems can reduce the overall productivity to the stage where long recovery times are necessary, although at low levels of cropping such a reduction may not be immediately apparent1. Other forms of primary production appear to be less important, although their true significance remains to be assessed. During the dry season blooms of phytoplankton,and the zooplankton which depend on them, appear to be dense. However, it is quite possible that the actual concentration and productivity of plankton is much greater during the flood season because of the dispersion of the organisms throughout a much greater volume of water and because a large proportion of the organisms present are liable to be eaten by young fish. These aspects of primary productivity are again very poorly understood and many more studies are required to define the role of the various basic foodstuffs present in the river system.
Feeding: Although a range of adaptations for various feeding habits are present in fish species of floodplains, the actual diet appears to be very flexible in most fishes. Some true specialists such as mud-eating or detritivore species do exist and often contribute a considerable part of the biomass, but in other fishes specializations seem to be more directed at maintenance feeding during the severe periods of drawdown. Most fish stop feeding almost completely during low water in tropical systems, where the cut-off may be very severe and corresponds to a “physiological winter”. Such behaviour mimics the true winter in temperate systems where the drop in temperature slows the metabolic process. A considerable part of feeding takes place on the floodplain, which therefore plays an indispensable role in the nutrition and growth of most components of the fish community.
Migration: Migration is an essential part of the biology of the fish in flood-river systems and the phenomenon is so marked that many fisheries are based almost entirely on such movements. Migration may be either longitudinal within the river channels or lateral from the river channel onto the floodplain. Some true anadromous species exist in rivers, i.e. Hilsa canagutsa of the Mekong or the sturgeons of the Danube but many freshwater species show potamodrous habits, undergoing very long migrations within the river system. Migrations are often thought to be solely for breeding but are equally important for feeding and in many systems there is no other explanation for the return downstream after breeding. Other movements are associated with avoidance of unfavourable conditions such as overly-strong current, deoxygenated conditions or overcrowding.
Breeding: The breeding of river fishes is strongly influenced by flood regime. Many species breed during the rising water, but this is by no means the only pattern. Some species breed prior to the flood during low water, or at peak floods, but breeding during the period of falling water is comparatively rare. Differences in breeding strategy are probably linked to the length of migrations to be undertaken as well as to the nature of the river system. Breeding strategies other than migration to the headwaters are very diversified and are aimed at securing favourable conditions for the hatching, survival and growth of the larvae. Floodplain fish accordingly show almost all conceivable variations of breeding patterns2.
Sub-Populations and Races: It is becoming increasingly evident that the composition of fish communities is considerably more complex than has been previously thought. Many species seem to be divided into races, which are distinguished either geographically or by behavioural differences such as the timing of their breeding period. The division of stocks into sub-populations or races means that some caution should be exercised in exploiting species because what initially may appear to be a homogeneous stock, may, in fact, be a number of sub-populations of which some elements are more vulnerable to the fishery than others.
1 Macrophytes are only rarely directly used by fishes but their function as a support for periphyton and perizoon is probably critical, as in their conversion to detritus after death.
2 “Whitefish” usually undertake long migrations toward the headwaters of the river and produce large numbers of eggs, whereas “blackfish” are more likely to show elaborate reproductive behaviour including nest building, mouth brooding or internal fertilization coupled with parental care.
Conclusion: The basic nature of the biology of river fish communities makes them extremely sensitive to changes in flood regimes because they are dependent on the seasonal flood to inundate the ground needed for feeding and reproduction. The nature of the flood regime also conditions fisheries to a great extent and this is discussed more fully in Section 3.
The Flood Curve: In flood rivers the characteristics of the flood regime have been shown to control the fish communities. Examination of the biology of these communities shows that, in general, the flood curve should have certain characteristics for the maintenance of good fish production and species composition. The curve itself may be divided into 5 parts as shown in Fig. 1, each of which may influence the potential catch from a system.
(a) The rising limb: Once the flood has started to rise, increase in depth should be moderately steady as an intermittent rise with alternate spates and recessions is detrimental to the breeding of many species. Similarly, hesitations in the rise of the water can allow deoxygenated conditions to occur in the littoral zone with consequent mortality to the young fish.
Fig. 1 A typical flood curve
(b) The flood peak: While preservation of flooded conditions is advantageous for the growth of young fish, it is apparent from experience in several systems that if a flood persists for too long water quality may deteriorate and fish kills sometimes become serious. Also any further primary production by macrophytes will probably only have marginal benefits to fish production. This has implications in floodplain areas which are to be submerged by bunding or by dams.
(c) The period of falling water: This is especially critical. It only contributes slightly to the growth of the fish and as such cannot be fully considered biologically as part of the duration of the flood. However, over-rapid decline is liable to increase stranding mortalities although these may well depend on the type of system and the ratio of permanent water area to total flooded area as well as on the degree of connexion of floodplain water bodies to the main river channel.
(d) Low water: It is of obvious advantage to the fish community to maintain the maximum amount of water in the system at low water, provided the floodplain itself is left dry. This implies that any further falls in water level once the bankfull stage has been passed, are detrimental to the fish. Also the low water period should be kept reasonably short although there is some benefit to having the terrestrial system remain dry for sufficient time for decomposition of terrestrial vegetation to occur. Alterations in the form and timing of the flood curve by dams in many rivers have shown that changes in biomass, yield and species composition of fish communities downstream (and upstream in the reservoir) rapidly ensue.
Production: Studies of parameters of river fish populations show that patterns of mortality and growth within one year depend on the flood curve and do not follow the more normally used exponential models that are applied to the more stable lacustrine or marine fish populations. The combination of “floodplain” models for growth and mortality gives a curve of the form shown in Fig. 2, which indicates that there is a considerable excess production during any one year, an unknown part of which is exploitable by the fishery.
Fig. 2 Within-year production curve obtained from “floodplain” models of growth and mortality. Normal production curve is shown for comparison.
(The assessment of fish stock and prediction of catch in large rivers, see page
The methodologies concerned with assessment of fish stocks can be divided into those concerning commercial catch data, experimental fishing data, and the use of environmental parameters.
Commercial Catch Data:
(a) In Backiel's method there is a problem of defining both the production biomass ratio and the natural mortality loss per annum with sufficient accuracy. Backiel's approach is very similar to Gulland's model for an unexploited stock in which biomass and instantaneous mortality were multiplied together with a constant to provide a rough estimate of future yield. Such simple but crude approaches may provide useful first cut estimates of potential yield, providing that reliable commercial catch data are available.
(b) The biostatistical approach initiated by Derghavin depends on a large volume of data concerning ageing of the catches and mortality estimates. The original application of the model was in a stepwise process, but there are continuous versions of the model, a variant of which is Virtual Population Analysis. The problem with the application of this method in multi-species fisheries is that many species are involved with different parameters. Obtaining data on all the species can be too expensive or in the case of some tropical species, impossible at present. In conclusion, the biostatistical approach is possibly better suited to assessment of single stocks under relatively stable environmental conditions unless we find significantly large groups of species which have similar dynamic characteristics and vulnerability to gear.
Experimental Fishing Methods:
(a) Area catch methods
Single sample methods: One of the most generally useful methods in floodplain lakes is that of chemo-fishing, usually with Rotenone in a blocked-off area so as to remove as much biomass as possible. This has the advantage of giving accurate point estimates, but the disadvantage of being costly and time consuming when large numbers of samples are required. The variance of biomass estimates obtained with this method is often quite considerable. In addition, there may be bias due to more active fish species escaping the encircling net before the poison is introduced, and there are difficulties in recuperating all fish that are killed. This can be offset to a certain extent by mark and recapture within the enclosure and by SCUBA observations.
Other area catch methods include the calculation of catch within the area swept by such moving gears as seines or trawls. Seines are used intensively in river systems but so far trawls are limited in their use. These methods can only provide reliable assessments if the catchabilities can be estimated, which can often be an expensive procedure.
Repeated methods: The most commonly used methods in temperate streams are those of De Lury and Leslie, although Leslie's method is essentially a variant of De Lury's. The major assumption in these methods is the constant catchability of the fish through fishing repeatedly in a restricted area. In practice, such methods have been used with great success in contained lagoons such as those of the Paraná, but for other systems interference has been present in the form of migrating shoals of fish which have raised the catches of fish late in the series thereby preventing proper extrapolation. Repeated electro-fishing in blocked-off areas of salmonid streams has proved successful but its applicability in large rivers is limited by species and size diversity, making calibration very difficult.
The two-catch method of Le Cren-Seber: This is limited to bodies of water where the catches make an impression on the stock.
(b) Catch per unit effort (c.p.u.e.)
Catch per unit effort is often used as an index of relative abundance, particularly with passive gears, but also with active gears whose catchability is unknown. The index may be biased by different catchabilities in different zones of the habitat. Also c.p.u.e. can be extremely variable, although variance can be reduced to some extent by using commercial catch statistics providing that effort can be satisfactorily defined.
Another problem when comparing long-term series of data is the learned response of fish species to gear.
(c) Mark and recapture
The use of mark recapture methods in extensive systems involves a large financial investment and the fulfilment of numerous conditions. The most important of these is the ability to mark large quantities of fish without producing a greater than normal natural mortality, particularly in river floodplain systems where there is a high annual mortality in any case. Another condition is the existence of an extension service or some other authority whereby the maximum chance of recovering marks and rewarding recaptures can be made.
Mark and recapture methods in large rivers at least in developing countries, seem to be generally not feasible for stock assessment although they may be of great value in defining migrations, certain dynamic parameters, and geographical limits of stocks.
(d) Mark and recapture for gear calibration
Probably the best use of limited mark-recapture experiments is for the valuation of other fishing methods. This is feasible in many situations, such as Rotenoning or checking seine net efficiencies, where released fish are marked and released into an enclosing net.
(e) Fish counters and fish counting
In certain situations it is possible to count individual fish as they migrate past a fixed point. This requires sophisticated technical services and high investment and at present is generally only useful for high-value stocks, such as salmonids.
Physical counting of such factors as the numbers of fish on spawning beds is only possible under certain conditions, as found in some temperate rivers.
The above grouping of methods is designed to provide estimates of potential catch and should ideally be combined with assessments of the existing fishery into an integrated sampling programme as indicated in Fig. 3. Many experimental fishing methods are supplemented by the data coming from commercial fisheries and for that reason a standard approach involving both catch assessment surveys for the evaluation of the commercial fishery and experimental fishing for theoretical determinations of potential, should be adopted wherever possible.
Long-term evaluations of catch made over a number of years have given disappointing results often because the characteristics of the fishing gear change inadvertently through the time series. It is, therefore, desirable to establish some form of standards for experimental fishing gears in order to ensure continuity and comparatibility of samples in time and between different systems.
Use of Environmental Parameters:
An index of production using environmental parameters has been evolved in Europe for the particular set of rivers found there (Leger-Huets method). This index involves such considerations as general productivity of the system as indicated either by vegetation cover or by benthic biomass as well as such morphological considerations as the width of the river or such edaphic considerations as the alkalinity or the temperature of the water. The index is further modified by the type of fish community present in the river. It gives good results for the European rivers for which it has been evolved. However, its value to other systems outside the temperate region has not yet been explored. An alternative approach simply using the catch of river systems as a correlation of the main channel length or basin areas has been adopted in Africa and the relationships gained from this experience have successfully been applied to other tropical systems. When the relationship is analysed in terms of the floodplain area of the various systems, it is apparent that the main determinant of catch is the maximum area flooded in any one year, and on the basis of this, values of 40–60 kg/ha are advanced as a normal “rule of thumb” production from tropical river systems. In many cases, it is doubtful whether greater accuracy than estimates obtained from such environmental methods is necessary and their quickness and simplicity permits advice to be given in situations where long delays cannot be tolerated.
Fig. 3 Suggested combinations of methods for stock assessments and yield predictions
The assessment of fisheries under the particular conditions prevailing in most rivers is difficult because the fisheries are highly dispersed and are rarely concentrated on particular landings. However, a well tried methodology has been elaborated by FAO in a series of projects (see page ). This methodology has sufficient flexibility to permit its adaptation to most conceivable cases, although there may be limitations in such highly dispersed fisheries as exist on some West African rivers. From the experience gained in these projects, it is clear that fisheries can be assessed by properly designed surveys with considerable saving of manpower and costs. It was pointed out that while such surveys may well be carried out under the aegis of a project, many countries have internal problems of financing and for this reason it is conceivable that external aid may continue to be necessary at least in the initial stages of planning and execution of such surveys. Once a base plan is established, this need only be altered when major shifts in river use or socio-economic patterns in the country occur, which are liable to be reflected in the fishery.
Prediction of yields is required for fisheries in three contexts:
(i) For estimating probable yields to be expected from new fisheries or fisheries which have not been previously investigated;
(ii) For prediction of the effects of environmental changes produced by natural catastrophes or human intervention;
(iii) For the year-to-year prediction of yields for economic planning.
Each of these needs a different degree of precision and any evaluation of the predictive capacity of the various models used can only be made in terms of the precision required for any particular sector. For instance, for estimates of the first type a 50% accuracy may be acceptable, as at this level it is possible to project gross investments such as roads, fish harbours, etc. A somewhat greater precision, say about 25%, is probably desirable for the prediction of effects of environmental changes and an even greater degree, possibly 15% or less, would be required for such managerial activities as issuing of fishermen's licences, marketing licences, or fish import permits. Estimates can be made in three main ways:
(a) Biostatistics: As discussed on page 6, biostatistics are apt to be overly-cumbersome in most river fisheries and are therefore of somewhat limited application. They are used with a certain degree of success in conjunction with environmental parameters in the case of such high value fisheries as those for salmonoids in temperate rivers.
(b) Fractional commercial: This approach has not actually been attempted in rivers but has been applied with some success to prediction of catches in some lakes. In this method the catch of fish in any one year is assumed to be a correlate of some fraction of the catch in preceding years.
(c) Environmental parameters: The environmental parameters described on page 7 have also been used to predict the yield of rivers in which the catch or the potential is unknown from other known rivers of the set. Year-to-year data has been gathered in at least five rivers: the Niger, Kafue and the Shire in Africa, the Danube in Europe and the Amur in Asia, and each of these sets of data clearly shows a correlation between the catch in year “y” and the flood regimes in one or two of the preceding years (y-1, y-2, etc.).
The lag between the year of good flood and the response in the fishery by increased catch appears to be a function of the growth and mortality rates of the fish and the time between hatching and recruitment to the fishery. Such relationships have been used to predict catches in future years and Holčik claims that for certain reaches of the Danube successful and accurate predictions can be made. However, work on the African rivers using series regression to which an increasing number of terms are added has shown that no increase in accuracy of prediction is obtained as larger numbers of terms are added to the equation. This implies that, with our present state of knowledge, and using limited regression formulae, we cannot make predictions that are sufficiently accurate to fulfil the requirements for year-to-year managerial decision making, although the accuracy of ± 25% that is obtained is adequate for both other purposes.
Because of the importance of prediction for a variety of purposes, further attempts should be made to set up predictive models. For this, long series of detailed data are needed from a variety of floodplains. It is, therefore, to be recommended that research institutions involved in work on river fisheries should attempt to collect such data on a continuing basis. This is particularly applicable to those floodplains such as the Kafue, Central Delta of the Niger or the Mekong, where series of data already exist, but also for rivers such as the Danube, where the annual flood cycle is already modified by damming.
(On fishing and fisheries management in large tropical African rivers with particular reference to Nigeria, see page 37)
Rivers tend to evolve through several developmental stages as pressure on the various fluvial and adjacent terrestrial resources increases. These stages can be summarized as in Table 1. Research, development and management strategies should be closely linked to the developmental stage of the river, as suggested in the fourth column of the table.
One of the most characteristic distinctions between industrial fisheries and small-scale artisanal fisheries is that industrial fisheries can be viewed solely within a fisheries context and managed by general fisheries principles, whereas small-scale fisheries can only be viewed in total socio-economic context of the basin in which they are practised. Normally all river fisheries fall in this latter category. Management can be applied in two ways. Firstly, there is scientific management involving initial deep study which leads to an understanding of the system permitting the identification of areas for control upon which decisions can be based. This type of management is reliable but is very costly in time and money and cannot therefore be applied in many of the cases with which we are confronted in fisheries in developing countries.
Secondly, as an alternative to scientific management, experimental management involves the formulation of a hypothesis about the fishery, based on whatever information can be readily obtained. The application of a management strategy based on this hypothesis and the subsequent evaluation of the results over a period of time. This type of management is very common in traditional fisheries and is being increasingly accepted as a legitimate practice in those cases where inadequate resources are available for a more analytical approach. Obviously such empirical methods are to be controlled closely and require adequate monitoring systems for their evaluation.
Traditionally, management has been effected largely through direct measures acting upon the fish resources and fishermen themselves. But in the broader context of regional or river basin development, it will often be better to take indirect action to improve the fishery, such as building a road or a market, or organizing transport coooperatives, etc.
The management of the fishery depends very much on the type of objectives selected and within this there are often tensions between fisheries for selected species of especial commercial value and for maximum protein production. The more it is thought desirable to control the fishery for the maximum production of certain species, the more controls are necessary and as a corollary, the greater the size of the species selected, the higher the costs of such control. River fishery communities react to fishing pressure by maintaining a certain yield level, but at the same time by changing their composition toward smaller and faster-growing species. Should the simple production of the maximum amount of protein be an acceptable objective, it is probably better to leave the fishery relatively uncontrolled, so long as it lies within those limits of exploitation that do not overly degrade the quality or quantity of the fish stock.
| Stage | Floodplain use | Fisheries | Research and management activities |
|---|---|---|---|
| Unmodified Floodplain shows most characteristic features, flood regime unhindered by direct human interventions but indirect effects of activities elsewhere in the river basin may be apparent. (Sepik, Niger, Sudd) |
In wild state often forested, supports game, later used for grazing cattle. Vegetation modified by burning. Seasonal occupation by nomadic fishermen, hunters and pastoralists. |
Fish stocks largely in original condition of diversity but size structure may be modified by fishing. Moderate to heavy fishery in both river channels and standing waters. Area available for fisheries, whole plain. |
Exploratory fishing for description of composition of fish stock. Identification of major resources. Studies on biology of individual species and their geographical and seasonal distribution. Studies on local fishing methods and introduction of appropriate additional techniques. Establishment of simple regulatory measures for protection of major stocks. Improve access and marketing network. |
| INTRODUCTION OF SYSTEMATIC AGRICULTURE | |||
| Slightly modified Some drainage channels for more rapid and efficient removal of flood waters. Smaller floodplain depressions filled or regularized. Flood still largely unaltered in timing and duration (Senegal, Ouémé) |
Floodplain largely cleared of forest, extensive drawdown agriculture, some floating rice in suitable depressions. Some areas reserved for grazing and zonation of floodplain for different uses often highly developed. Settlement on levees and higher ground or on artificial islands and stilt villages. |
Fish stock largely unaltered although larger species may be becoming rarer and size structure heavily biased toward smaller individuals. Some depressions may be dammed as holding ponds, or for extensive aquaculture, or fish holes may be excavated. Area available for fisheries, whole plain. |
Population dynamics of major elements of community to give refined estimates of potential yields. Continue studies on biology to identify possible subpopulations and to describe ecological interactions between species. Monitoring of fishery to detect potential over fishing of major stocks coupled with intensification of regulatory measures to protect fish stock. Investigation of simple forms of extensive aquaculture. Improvement and concentration of fish landings and preservation techniques. |
| INTRODUCTION OF FLOOD CONTROL MEASURES | |||
| Extensively modified Drainage and irrigation common, some flood control through dams and levees which contain main channel. Depressions usually filled or regularized. Flood often modified in timing and duration. (Chao Phrya, Mekong) |
Flood agriculture (usually rice) and intensive dry season agriculture. Moderately extensive occupation of the dryer areas of the plain for habitation - beginnings of urbanization. Much of plain still subject to flooding. |
Some modification to fish stock with disappearance of larger species. Wild fisheries often very intense in main river channels, with some new fisheries in reservoirs. Rice fish culture in suitable areas. Drain in ponds and some intensive fish culture in regularized depressions. River area available for fisheries restricted. |
Examination of general dynamics of fish community to judge reaction to various sources of loading. Intensification of monitoring of fisheries with increased control of catching methods by licensing and legislation. Examine impacts of other activities in the river basin on the fishery and endeavour to ensure that suitable conditions are maintained. Investigation of intensive aquaculture methods. Development of reservoir fisheries and seek alternative employment to reduce fishing pressure on main river. |
| CONSTRUCTION OF MAJOR UPSTREAM DAMS | |||
| Completely modified Flood control by large upstream dams and by levees. Main channel sometimes channelized. Floodplain largely dry although still subject to occasional catastrophic floods. River often reduced to chain of reservoirs (Mississippi) |
Urbanization, intensive use of plain for agriculture, industry and habitation. |
Fish stock changed by loss of some species through pollution and channelization, sometimes by introduction of exotic species. Some sport fisheries in main channels or in few lakes retained on floodplain. Some intensive aquaculture in specially constructed pond. River area available for fisheries small but reservoirs may be developed for fisheries |
Investigation of pollution and other management impacts to establish criteria for maintenance of fish stock. Regulation of discharge and effluent according to these criteria. Contemplate introduction of new elements to fish community or stocking to support threatened species. Study access problems to fishery to resolve conflicting demands of sport and commercial fishermen. Intensify development of aquaculture and reservoir fisheries. |
Several techniques are used for management as summarized in Fig. 4.
Most of these techniques have been found to be impracticable under the conditions of spatial diffusion and high population prevailing in most river systems. Perhaps the most reliable of the above techniques is that involving the control of fishing season and here, fortunately, in flood rivers the flood itself imposes a closed season during the period of rising and high water. Nevertheless, it may be felt necessary to impose further controls especially during the end of the dry season when fish are congregated in much reduced water areas and aggregations of pre-spawning fish are especially vulnerable to certain fishing methods. Each of these management strategies can be superimposed from above or below. In most established fisheries traditional controls which evolve over many centuries have used these techniques within a social self-regulatory context but this has since disappeared. It is consequently necessary to reintroduce these controls by legislation from above or by education and popular support from below.
Fishing gear:
Because the fish communities in large rivers are highly complex and show great differences in seasonal distribution as well as in the size and ecology of their component species no one gear is adequate to fish a whole community. In traditional fisheries throughout the world a wide range of methods have been developed, many of which are highly specific to particular species or size classes. There is, therefore, a tendency for professional fishermen to use a succession of gears which are adapted to particular phases in the annual cycle. Generally, static gears are set during the rising and falling flood for the capture of migrating fish and the active gears tend to be favoured at times of low water when the fish are less active. Experience with river fisheries has shown that this diversified pattern of exploitation is highly efficient and should be encouraged in development programmes. New gears or improved methods of fishing should be considered most carefully in the light of existing fishing systems before they are introduced. The other basic article of equipment for fishermen is their fishing craft. A distinction can be drawn between two types of boat: the boats used for fishing, especially on highly overgrown floodplains, and the considerably larger craft needed for the transport of catch from the fishing site to the market. In general, traditional designs of dug-out canoes are more than adequate for the first of these purposes but in many cases improved designs are required for transport. Similarly, motorization of craft is often desirable for the transport boats, which tend to be restricted to the main river channels, but is rarely profitable for the smaller canoes which operate on the floodplain system itself.
Strategies for development:
Because in most instances the level of catching technology appears adequate, further development of river fisheries lies in the improvement of the handling, collection, processing, storage, transporting and marketing of the fish caught.
The choice of techniques used for the development of capture fisheries depends on which measures are practical for any combination of geographic and socio-economic factors. Sometimes, the most important development and management measures cannot be applied either for lack of general response or for the impossibility of enforcement.
Three main strategies can be identified for development. These should not be considered as alternatives as they may be applied side by side, or combined into intermediate variants, depending on the specific conditions of the country, region, development stage and institutional level.
Fig. 4 Diagram of relationships between main techniques for fishery management
These main strategies are:
(i) Public service development and activation strategy (development from above)
(ii) Beach-head strategy (development from below)
(iii) Profit-oriented enterprise strategy.
(i) Development through public services is applicable only where adequate government or municipal institutions and extension services exist or where there is a good chance of having them established or improved. The government or the responsible development agencies may thus execute their policies through these services and institutions, activating them at all hierarchical levels down to the fishermen. The government, central or regional, or foreign development agencies would provide funds, training and advice at rather a high level so that assistance, initiative and know-how could reach the fishermen through these services.
(ii) Beach-head strategies are applicable where, for various reasons, activity through the existing administration does not seem feasible or possible. In some cases, the existing institutions may be a liability rather than an asset; for example, where corruption, petrified bureaucracy or elements resistent to the development policy dominate the administration. This strategy is equally recommended where substantial central organizations do not exist or are in a foetal stage and establishment of a useful service is either unfeasible or would require a very long time. Under such conditions, beach-head type projects which are financially, institutionally and logistically independent from the intermediate bureaucracy may be initiated and directly supported by development agencies, if necessary, with expatriate experts participating directly at the fishermen's level. Successful beach-head projects would expand through radiation of both the developmental ideology and direct technical assistance throughout neighbouring communities.
(iii) Profit-oriented enterprises may be used where dormant financial resources are available and profitable commercial development is evident. This applies in particular to fisheries where underexploited resources may abound in the immediate riverine or marine area or where conditions permit establishment of aquaculture projects. It applies equally where new land areas are opened for farming and where rural industrialization or mining can become elements in integrated rural development. Such development may include small canning or reduction plants installed by local national or international interests which can, if necessary, be supported technologically and technically by external aid. It is essential to involve the local rural population in such developments at all levels.
Any of these strategies or some mix of them can be applied in most types of national or regional political situations. For example, in a strictly socialist country the profit-oriented strategy can be applied through cooperative, collective or government owned farms and enterprises, while under a strictly anti-socialist regime the same strategy can be applied by encouraging private enterprise and banks' capital investment. In more liberal regimes private enterprise, government owned enterprises, and cooperatives can all be encouraged and assisted in the process of development. The same, obviously, applies to other strategies.
Community fishery centres:
Because the present socio-economic situation prevailing in river fisheries cannot be improved without bettering fish technology, and because this is best done at the community level, the concept of community fishery centres has been developed. This proposes an integrated approach to the development of artisanal fisheries through establishment of groups of community fishery centres. Long-term technological and extension support can be provided to groups of three or more centres from a centrally situated fisheries extension unit. Typically, such centres would be a complex of modules selected to suit local conditions. This approach can be adapted to any political, social, economic and institutional organization. It may be shore-based, constructed on an “artificial island” or on floating rafts or powered barges in the form of a mobile “trading station”. By having the centres in groups or fleets, the cost of the technical supporting expertise per fisherman and per ton of fish is reduced. A centre can become a component or even a nucleus for integrated rural development. Its modules may comprise capture, landing, processing, transporting, supply and marketing services. A thorough socio-economic and fishery study on the prospective site of every centre is a condition for its rational technical design, suitable institutional set-up and successful operation.
For the specific purposes of riverine fisheries, mobile collecting/trading stations (motorized barges, vessels, etc.) which incorporate the needed technological modules can play the role of community fishery centres. A smaller or larger fleet of such vessels would represent a community fishery centre group, while the fisheries extension unit support may be provided by both leading crew members of those vessels and by a stationary unit at the catch landing sites. Where after several years of development such a system successfully replaces the previously existing one, new possibilities of management may occur through, for example, controlled fish collection/purchase from the fishermen and controlled supply of equipment. The new social spirit encouraged by such centres would facilitate the application of popularly supported management measures.
Natural production of rivers can be exceeded by various forms of husbandry and extensive or intensive aquaculture. These are widespread on floodplains and various techniques are available for the different stages of the developmental sequences of flood rivers. A catalogue of such culture systems would include the following:
(i) Drain-in ponds: These are simple ditches or depressions excavated into the surface of the floodplain which serve to concentrate the fish as the waters recede. They may be stocked with fry or fed artificially, but natural production of over 1 ton/ha are often achieved and such supplementary management seems not to produce much improvement.
(ii) Akadjas: These are parks of branches or floating vegetation which are installed in rivers or large floodplain lakes and serve as collectors for fish. They are somewhat more than simple refuge traps as the populations within them increase their breeding and growth independently of the conditions in the surrounding waters.
(iii) Blocking of natural floodplain depressions: The blocking of existing canals from natural floodplain depressions to retain water in the depression for a longer period of the dry season is extremely common throughout the world. This form of husbandry tends to be practised in Africa and Asia to conserve stocks of fish for harvesting during the time when fish from other sources have become scarce. However, systematic studies of this method in Nigeria show that by feeding and stocking, such depression cultures can produce up to 500 kg/ha/year and as such a method is potentially very valuable.
(iv) Rice-fish culture: The culture of fish in rice fields and in the canals associated with them is a particular feature of S.E. Asia, where it has formed the basis for the rural economy for many centuries. Considerable yields exceeding 1 t/ha can be achieved by this method provided the rice is single cropped. However, intensification of rice culture through double cropping and insecticide applications has proved detrimental and the amount of fish produced by this method is at present diminishing on a world scale.
(v) Intensive aquaculture in ponds: Once the flood is completely controlled and the former floodplain is converted to irrigated agriculture, intensive aquaculture in pond farms can be practised as has been shown in the Chao Phrya river in Thailand and in the Danube in Romania and the U.S.S.R. Here, series of ponds are associated with the irrigation system and the plentiful supply of agricultural waste materials made available by the intensification of agriculture in the surrounding regions allows intensified feeding of the fish crop.
(vi) Cage culture in rivers: Once flow is stabilized within the main river channels or even before such stabilization occurs, fish may be cultured in floating cages. This type of culture usually applies to species of high commercial value which are caught as fry and reared to marketable size in such structures.
(Fishery problems associated with multiple uses of large rivers, page 47, and An alternative scenario for river basin development in African woodland savannas, page 53)
Conflicts tend to occur at all levels within the fishery itself. Such conflicts as those between local fishermen and immigrant fishermen, commercial fishermen and artisanal fishermen, sports fishermen and commercial fishermen, or even between the users of different types of fishing gear are common in all river fisheries. Such tensions may be resolved either from below within the fishing community or from above through legislation. The role of management in this situation should be to resolve the conflicts as quickly as possible so as to minimize the loss to the fishery, and also to avoid the introduction of new conflicts by failure to clarify its objectives. In most cases, however, the objectives for which the fishery is pursued are dictated by political or economical situations which lie outside the control of the pure fishery administrator.
It is expected that large-scale alteration of river basins will accelerate during the next two decades, both because of the energy crisis, which will require the construction of an increasing number of hydro-electric power facilities, and the increasing demand for grain. In such projects, water management considerations have dominated in the past at the expense of the integrated development of the human, land and water resources of river basins. Consequently the river basin developments have favoured the urban industrial sector over the rural sector. The over-emphasis on this type of water management can be expected to continue in the future. Nevertheless, the time is ripe to attempt a more balanced development of river basin resources including fisheries, crop agriculture, livestock, tourism and wildlife in contrast to the existing stress on power generation, flood control and irrigation. In part, this is because of the increasing emphasis on the need for development to reach the rural core, but also because many case histories now show the dangers, both economic and ecological, of too narrow an approach to river basin development.
To increase the role of fisheries in river basin development projects, it is necessary, however, to be realistic. To meet the needs of the engineers and others responsible for model building, fisheries experts must put values on the cubic metres of water needed for fisheries utilization, so that there is sufficient information to assess the trade-offs between different development strategies. Another reason why the time is ripe to push for utilization of a wider range of resources is that in recent years we have achieved a better understanding of the impacts of reservoirs and dams on fish populations and on both fishing and farming systems. There are a number of important case histories showing both the right direction and the wrong direction to proceed typified by the Mekong and Kainji experiences (page 21).
The Kainji case is a classic example of the serious costs which can arise from a narrow focus on river basin development. The Mekong case is an example of how multi-purpose planning can incorporate fisheries and other considerations.
This Mekong study, carried out in connexion with the Pa Mong Dam, is of special importance. It is the best and perhaps the only example of an optimized study, which assesses the implications of different water management strategies for different resources and puts considerable emphasis on biological productivity in terms of both fisheries and agriculture.
It is expected that the use of controlled floods in the future will show sufficiently high economic rates of return to justify some use of reservoir water for downstream development purposes. While, here again, the primacy in many cases of hydro-electric power generation must be kept in mind. The opportunity for developing compromise systems for water use appear particularly high at this moment. In two cases, the Shire River project and the Pongolo river project, down-river controlled floods have been tried and have had favourable effects on the fish populations.
At the scale of planning required for action at the basin level centralized planning “from above” cannot be avoided. It is characteristic of such planning that it focuses on the constraints and weaknesses of the existing systems, infrastructures and relationships. If the strength of such systems is to be taken into account, it is essential that localized “planning from below” be dove-tailed with the centralized planning. It is therefore of highest priority that the authorities and planning agencies responsible for development found their plans and actions on an accurate understanding of the existing base. This means that broad-based but carefully designed integrated surveys and inventories of the existing ecological-sociological structure be made to provide baseline knowledge before changes are planned. These, in turn, provide the framework for more detailed specific surveys and studies, designed to answer specific developmental questions about sub-components such as - should a fishery be motorized?
Experience has shown that such surveys should extend to the whole river basin, especially downstream of proposed project activities, as the major impact of almost any action will extend downstream. It was also noted that fishery communities on floodplains and rivers are characteristically more permanent, more seasonal in activity and more often of subsistence type compared to other fisheries; and least organized to exert political and economic “muscle” on their own. On the other hand, such fishermen are more apt to have existing water rights and related traditional social structures suited to the existing conditions and are hence most easily damaged by changes to the river regime.
An example of the potential impacts of fisheries on development is the far-reaching effect of the initial peak “surplus” of fish characteristic of new reservoirs. At the Kariba Dam, where the most data are available, the very high rates of return in the first years of fishing on the reservoir provided capital for financing the local initiatives in farming, trading and other activities now much more important in the region than fishing. The consultation urged that every effort be made to take advantage of the abundant fish stocks available in the initial period of a new reservoir's life to provide such general development stimuli. While an injection of capital into local communities also occurs along with every major construction project, none of the other sources appear to be so well concentrated as to produce the “finding buried treasure” effect that good fishermen have enjoyed at Kariba, Volta and Kainji Lakes.
The main conclusion that emerges from the discussions is the great need to set up, on an experimental basis, an integrated trans-disciplinary model. This should be capable of predicting the effects of a variety of development strategies and development activities upon the rural populations inhabiting river margins and floodplains, as well as the overall impacts of such schemes on national programmes. A major focus for their work could well be the application of systems analysis to some key river basins. To bring this about, the following sequence of activities was recommended:
(a) In order to command the attention of river basin and water management planners, it will be necessary for fisheries, agriculture, wildlife and other disciplines concerned with rural development to join forces. Within FAO this can probably be best accomplished through the Working Group on Rural Development and the IDWG on Natural Resources and the Human Environment. It is thus proposed that the first step should be to present the findings and recommendations of this Group to these Working Groups. Joint attempts can then be made, interdepartmentally, to define alternatives for river basin development, to determine the kinds of information and data that are likely to be required for the development of the systems model and to act as the liaison group within FAO for approaches to the International Association for Systems Analysis in Vienna (IASA) and the various river basin group authorities. Still within FAO, attention should be given to the development of various case histories for presentation to planning groups, to engineers and to relevant Government agencies, explaining both the successes and the failures arising out of major river-basin projects completed or underway throughout the world. Consideration should also be given to the development, standardization and dissemination of methodologies for resource, economic and sociological evaluation of existing systems of river and floodplain uses and of proposed changes.
(b) Following upon the completion of a substantial part of (a) above, a general report should be prepared on these activities aimed at presenting the problems of river and floodplain development to such groups as IASA on the one hand and the planners and engineers engaged in major river basin projects on the other. At this time, a general operational manual should be assembled which would:
(i) Set out the initial problems that should be considered in relation to major development projects;
(ii) Provide guidelines for developing the information systems and organizational structures necessary to ensure effective optimization of all the potential impacts of river basin projects.
(c) When such materials have been assembled and the cooperation of IASA has been obtained, further consideration should be given to the selection of specific rivers to be used for experimental or pilot studies in the mode of systems analysis. It was noted that the necessary resources to fund the required studies would be most likely to become available for projects based on international rivers.
(d) Having selected suitable basins for study and having established necessary working relations with the relevant authorities, regional working parties should be held to bring together personnel from the authorities, from IASA, from FAO and other UN agencies and from other potentially cooperating institutions. The objectives of these regional working parties should be to establish appropriate organizational structures required to carry out the development of appropriate analytical models and their incorporation into the existing planning frameworks.
(e) At this stage, consideration should be given to the establishment of regional institutes for river basin development research linked either directly or indirectly with existing regional research institutes, such as the CGIAR group. These research institutes should continue an experimental approach to the improvement of river basin development planning along the lines already initiated, and provide guidance to existing national research units as well as to the national and inter-governmental planning agencies.
(f) Finally, these regional centres should undertake an evaluation and long-term monitoring of the results of the pilot and experimental programmes outlined above.
A series of related actions was also recommended, which should be undertaken parallel with the above, but which could be implemented independently of the other sequence.
(i) A step-wise approach extending over two biennia was proposed for obtaining data-relating fisheries yields on floodplain rivers to the amounts of water provided and to other management practices. This should involve firstly a series of meetings at the regional level associated with the appropriate regional fisheries bodies and ultimately to culminate in a major symposium which would draw together leading experts in the field from all over the world. In order to generate the data needed for such meetings, it was suggested that FAO use its good offices to approach sources of funding for the support of institutions in developing countries which could carry out the necessary research.
(ii) It was recommended that the Fisheries Department engage in a series of discussions with other units in FAO and in other UN agencies to attempt to identify specific areas where practical compromises can be achieved in river basin development schemes. Such areas might include the amount and timing of release of water from dams to recharge lagoons on downstream floodplains for aquaculture and irrigation uses, flow required on spillways and canal structures to reduce the incidence of onchocerciasis, etc.
(iii) That FAO assemble a roster of experts working in the various disciplines concerned with floodplain and river basin development, coordinating such rosters with other specialized rosters which have been developed by bilateral groups, the World Bank and other agencies of the UN.
(iv) FAO should serve as a centre for collecting and disseminating information and data on floodplain development projects and their execution.
(v) A Working Group or Groups be assembled to formulate appropriate research programmes to be used as models for national research activities associated with river basins and for presentation to UN universities and other potentially interested groups.
The Working Group accepted to continue to act in an advisory capacity by correspondence and eventually to serve as nuclei for future activities in their respective regions.
