Robin Welcomme

AbstractRivers contribute some 5 million tons of fish for human consumption, most of which comes from major floodplains. Fisheries in the tropics are over-exploited and show signs of fishing-down. Several trends are damaging river fisheries. Floodplains are being drained for irrigated agriculture, leading to modifications to their form and function. River channels are also being modified. The modifications result in the loss of some fish guilds. The growing demand for water is leading to increasing abstrac-tion and control of flow by dams. This means that there is often insufficient water available for the functioning of the system with lessened survival of fish species and lowering of production. Flood control to curb urban and farmland flooding is exac-erbating the separation of the plain from the river. Technical solutions are now well established and include the creation of protected floodplain beads in the developing landscape, and the agreement of environmental flows. Political and social application of these solutions is more difficult and calls for the application of mechanisms to incorporate fisheries interests into generalised river development plans. Key words: Floodplain fisheries; overfishing; environmental flows; river rehabilitation

1. Introduction

The world’s rivers have long supported fish-eries that contribute significantly to human diets and currently contribute an estimated 5Mt to world fish production. They also form major locations of biodiversity and being extremely varied in form and function (Ward et al. 2002), they support some of the most diverse assemblages of vertebrate spe-cies. The current situation must be viewed against the historical and future trends for landscape deve-lopment. Growing human expansion is exerting increasing pressures on the natural environment to satisfy needs for food, energy and living space which are being exacerbated by recent trends in energy and food pricing. These are leading to growing demands on water and fish resources lead-

ing to overfishing of river fish stocks on one hand and the increasing modification of the main chan-nel and floodplains of lowland rivers on the other. The large scale modification of aquatic environ-ments is a danger for the sustainability of river fisheries and biodiversity. Overfishing and envi-ronmental modification together account for the present dominance of r-selected, eurytopic species in fish assemblages, the apparent instabilities of the systems and the heightened overall production associated with the intermediate disturbance hypothesis (Zalewski, Welcomme 2001).This paper reviews the main trends and their impacts on fish-eries and fish assemblages. It also examines areas where the principles of ecohydrology can be applied to help national and international actions to mitigate the changes that are occurring.

Vol. 8No 2-4, 169-1822008

World prospects for floodplain fisheriesEcohydrological Processes and Sustainable Floodplain Management

Imperial College, South Kensington Campus, London SW7 2AZ

e-mail: Welcomme@dsl.pipex.com

DOI:10.2478/v10104-009-0013-0

R. Welcomme 170

2. Production potential of floodplain rivers

Global patterns of production

River fisheries are carried out either as dif-fuse, small-scale operations conducted on the small (lower order) streams and wetlands throughout the river basin or as more intensive commercially ori-ented activities on the larger lowland rivers includ-ing the major floodplain areas. In general national statistics only take into account fisheries concen-trated on identifiable centres where markets and fixed landings enable enumerators to work with ease. No distinction is made between river and lake fisheries so any figures are at best extrapola-tions based on circumstantial data. In a few river systems more scrupulous analyses have been undertaken, but even in these the more diffuse fisheries are often overlooked, resulting in a gen-eral underestimation of the overall contribution of river fisheries to global production (see for exam-ple Welcomme 1976; Coates 2002). Among the most serious of these oversights are the low level subsistence fisheries of rice fields that may pro-duce up to 150 kg ha-1 (Gregory, Guttman 2002). Furthermore there are many deficiencies in report-ing, so the assessments are often extrapolations from previous data sets and fail to reflect adequate-ly trends in catches. The difficulties of interpreting current data are indicated by the situation on the Mekong where the nominal freshwater catches reported to FAO by the riparian nations of the Mekong River (Cambodia, Lao, Thailand,

Vietnam) was 685 400 t in 2005, whereas the Mekong River Commission (2007) estimates the yield of the lower basin as about 2.2 Mt. Individually reported catches from the FAO Fishstat data base and reports from individual fish-eries give a rough estimate of about 5 Mt y-1 catch from the major floodplain fisheries of the World. This is approximately half of the total inland fish catch for 2005, the remainder being derived from lake and reservoir fisheries, many of which are enhanced by stocking. Because of the assumed general under-reporting of catches from rivers and associated wetlands, actual catches are probably far higher. The inland fish catch is highly biased by continent. Asia produced about 65% of the total in 2005 base mainly on large river fisheries as well as reservoirs and highly managed smaller lake sys-tems in China. Africa produced 25% about evenly divided between lake and river fisheries. Latin America contributed about 5% and Europe about 4% of the total.

Catches from individual rivers

Estimates of catches from many sources indicate that the catches from individual rivers, their floodplains, and riparian wetlands are inti-mately linked to the duration and extent of flooding during the flood phases of the hydro-logical cycle (see Table I) (Halls, Welcomme 2004; Welcomme et al. 2006a), Some authors have also identified links to the amount of water remaining in the river during the drawdown period (University of Michigan 1971; Quiros,

170

Climatic zone River Author

Temperate Europe Danube River (Romania)

Danube R.(Serbia) Botnariuc (1968) Stankovic, Jankovic (1971)

Danube R. (Slovakia) Holcik, Bastl (1977); Holcik, Kmet (1986); Holcik (1988)

Asia Amur R (Russia/China) Krykhtin (1975) South America La Plata system (Argentina) Quiros, Cutch (1989) Tropical Asia Mahakam R. Christensen (1993) Mekong (Dai fishery) Baran et al. (2001) Ganges/Brahmaputra delta, Bangladesh de Graaf, 2003 Africa Cross R. Moses (1987) Niger R. Welcomme (1976); Lae (1992) Shire R. Welcomme (1985) Lake Kariba (river driven reservoir) Karenge, Kolding (1994) Lake Turkana (river driven lake) Kolding (1993) South America Orinoco R. Novoa (1989) Pilcomayo R. Payne, Harvey (1989) Amazon R. Lambert et al. (1990) Australia River driven estuarine/ coastal Loneragan, Bunn (1999)

Table I. Some studies demonstrating positive correlations between flood intensity and fish catch.

World prospects for floodplain fisheries 171

Cutch 1989; Arthington et al. 2005). Fish pro-duction has declined where floodplains have been modified. For example, an estimated 40 000 t y-1 were lost when the middle Parana was dammed (Bonetto et al. 1969) by human activity and catches dropped in the Central Delta of the Niger by 50% to about 50 000 t y -1 during the Sahelian drought (Quensiere 1994). Pompeu, Godinho (2006) document the reduction in abundance and richness of fish from floodplain lagoons in the Sao Fransisco River after persist-ent failure of flooding. Studies in warmer temperate rivers [for example by Antipa (1910); Holcik (1988) and Schiemer, Weidbacher (1992) for the Danube; Amoros (1982) and Copp (1989) for the Rhone, Grift et al. (2003) for the Rhine and Lasne (2008) for the Loire] all show that floodplains and lateral connectivity to riparian wetlands are essential for refuge, feeding, breeding and nurseries for certain species of fish. They also show that changes to floodplain connectivity can alter the composition and abundance of the fish communities ( erný et al. 2003). The role of floodplains in the fish’s life cycle is less clear in cool temperate and semi-arid rivers. The reason for this is that in the warm temperate, sub-tropical and equatorial rivers flooding is usually coincident with the tempera-ture regimes so that fish on the flooded plains are presented with ideal spawning, growth and sur-vival potential. In cool temperate zones maximum floods often occur in winter and the growth advantage does not exist. Here fish do penetrate the floodplain but the reason may be more closely linked to shelter from excess current rather than reproduction and growth. The question of the role of lateral connectivity to riparian wetlands has been generally overlooked in European river studies, mainly because most of the former ripar-ian flood habitats have been destroyed. Conditions in semi-arid streams, such as those of Australia and South Africa are much more unpredictable and the streams may often fail to flood for several years in succession. Here too the role of the floodplain in the fish’s life his-tory is still being assessed (Jones, Stuart 2008). Catch information from a number of flood-plain fisheries indicates that catches from Latin American floodplains are about a mean of 28 kg ha–1 y-1 when the area in hectares is measured at maximum extent of flooding. Similar figures from Africa are 40-60 kg ha–1 y-1 and from Asia around 100-120 kg ha–1 y-1 (Welcomme 2001), although Sverdrup-Jensen (2002) and Baran et al. (2001) estimated yields as high as 230 kg ha–1 y-1 for the lower Mekong system. Catches from indi-vidual floodplains were listed by Welcomme (1985, 1995) on the basis of the work of a number of authors and remain the most detailed estimates made for many fisheries.

3. Factors influencing the future of floodplain fisheries

Floodplain fisheries are influenced by man-agement of the fishery on one hand and the state of the environment of the other. The state of manage-ment of river fisheries is generally poor, but the impacts of other human interventions on rivers pose a greater threat to floodplains. Many human uses directly modify the form and function of the floodplain and river channels. They also impact on the quantity and timing of river flows and indirect-ly affect the erosion deposition processes that are required to maintain the floodplain and the diver-sity of channel habitats. These processes impact the fishery by influencing the overall abundance of the fish community and the species composition of the fish assemblages present.

Fisheries management

Fisheries in large, lowland rivers typically involve large numbers of species occupying a wide variety of habitats. Oberdorff et al. (1995) found correlations between three independent variables and species richness in 292 rivers worldwide. Here basin area alone accounted for 61.7% of the variation implying an exponential relationship between the size of the river as rep-resented by its basin area and the number of spe-cies present. Because of the large number of spe-cies involved, their location within the river-floodplain system and the way in which these behave in response to flow regimes Welcomme et al. (2006b) proposed a system of behavioural guilds to assess the responses of different groups of fish to morphological and hydrological chang-es in river systems. This system will be used to describe the principal changes occurring in river-floodplain systems. Current levels of catch and effort involve substantial changes to the fish assemblages inhab-iting rivers. Fish assemblages of floodplain rivers respond to heavy fishing pressure by undergoing the fishing-down process (Welcomme 1999). This means the successive loss of the larger species and individuals from the fish assemblage. It does not involve any loss of production in the sense of absolute weight of catch but usually involves considerable shifts in relative abundance, food chain functioning and overall biodiversity. In extreme cases larger species may be actually eliminated for the fauna as local extinctions (Allan et al. 2005). Fishing-down also means that the population responds much more rapidly to changes in flood-ing between years. The extent to which this has occurred in any system is indicated by the time taken for a year class to enter the fishery. In the 1950s, some river catches were correlated with

R. Welcomme 172

floods as much as four to five years earlier, indi-cating that the fishery targeted older fish (See for example Krykhtin 1975). More recent catches are often correlated with floods of the same year, indicating that the fisheries are based mainly on very young fish, and that they may be at risk. For example catches were correlated with years y and y+1 in the Danube (Holcik, Bastl 1977) in the 1970s. By the 1990s catches of the major com-mercial species in Bangladesh consisted almost entirely 0+ age fish (Halls 1999). Sixty nine per-cent of the fish caught in the Niger River are 0+ fish (Lae et al. 2004). Mekong catches have come to be dominated by smaller short-lived species (van Zalinge et al. 2004). Personal experiences from the Oueme River in Africa showed a steady reduction in size of species with the disappear-ance on many of the larger ones. More recent observations by Laleye (unpubl.) confirm that most of the catch now consists of 0+ fish. In Latin America similar signs of fishing-down have been reported for areas around major urban centres such as Manaus, Iquitos and Leticia on the Amazon (Petrere et al. 2005) and several other Brazilian rivers (Agostinho et al. 2005), the lower Orinoco between Cuidad Bolivar and Cuidad Guyana (Novoa 1982) the Magdalena River, Colombia (Jimenez-Segura 2007). Experience in many countries has shown the management of river fisheries to be extremely dif-ficult. The diffuse nature of river fisheries, the large numbers of fishermen and the wide range of gears used have shown that the traditional technical measures and input and output controls are extreme-ly difficult to enforce (Welcomme 2001). One of the few means of effective management appears to be the introduction of co-management systems involving increased control of the fisheries by the fishermen themselves. Unfortunately, the increas-ing destabilization of traditional riparian communi-ties in response to the changing conditions along floodplains is leading to a corresponding fragmen-tation of fishing communities that increases the dif-ficulties of setting-up and maintaining such groups of fishermen (Dugan et al. 2007).

Physical modification

The World’s rivers have already been modi-fied to a considerable degree; Nilsson et al. (2005) for instance estimate that over half of large river systems have been modified by dams. Floodplain rivers are being modified in a number of ways to respond to human needs for living space, agricultural crops, energy and transporta-tion. The major tools in modifying rivers aim at controlling water flows in one way or another. Thus transversal dams retain water for later release for power generation and irrigation, or divert water from one river basin to another to

make up for shortfalls in supply. Longitudinal dams (levees) and polders control the lateral movement of water and separate the floodplain from the river channel, and river straightening and stabilization works associated with naviga-tion and drainage remove essential diversity from the river channels. One major problem is the changes that have occurred in siltation rates. These operate on the erosion-deposition cycles that maintain the river channels and floodplains. On one hand siltation has increased by about 20% through human activ-ities in recent decades (Millennium Ecosystem Assessment 2005). On the other hand, much of this siltation is not being deposited on floodplains but is being contained in main river channels or is being swept out to sea. This means that the nor-mal creation and aging process of natural flood-plain waterbodies and wetlands described by Botnariuc (1967) is halted, many riparian features such as dead arms silt up more rapidly and the diversity of substrates in the river channels is submerged in layers of silt. Changes occur within the floodplains and riparian wetlands as they are converted for agri-cultural use either as permanently dry land or as seasonally flooded rice fields. This involves fill-ing of permanent waterbodies, draining of marsh-es and siltation of channels. De-forestation is widespread, converting the dynamics of the floodplain from that of seasonally flooded wood-land to grassland savannah for agriculture and livestock grazing. The Millennium Ecosystem Assessment, (2005) estimated that between 56 and 65% of wet-lands (mostly floodplains) had been converted for agriculture in Europe and N. America by 1985. Similar figures for Asia were 27%; S. America 6% and Africa 2%. Certainly the situation has deterio-rated still further since 1985 and, with the increas-ing demand for grain crops worldwide it may be anticipated that the rate of conversion will acceler-ate. There is an especially shortfall in rice produc-tion at present which will probably force the trans-formation of more available bottom lands into rice-fields over the next few years. Furthermore, the growing concern over glo-bal climate change, the growing demand for cheap power and the perception that hydropower is “green” energy in that it does not release large quantities of carbon will probably result in a new wave of dam building across the World. This will probably take the form of new high dams, such as the Grand Inga project on the Congo River, the Ayourou project on the Niger River, the Khone Falls project on the Mekong, the Teotônio rapids dam on the Rio Madeira and a series of dams in the Amazon and Paraná rivers, although the development of increased numbers of small mini-power plants is equally likely.

172

World prospects for floodplain fisheries 173

Loss of floodplain by failure to flood because of upstream dams or because the floodplain has been poldered or otherwise isolated from the channels of the river, means the loss of those fish guilds associated with floodplains. Thus the len-tic guilds [Plesio-potamonic and paleopotamonic guilds] that inhabit various types of floodplain waterbody will disappear as the waterbodies are filled in for agriculture. Those guilds that inhabit the main river channel but rely on the floodplain for reproduction and nursery areas [Eupotamonic phytophilic and Para-potamonic guilds] will be severely depleted as access to the floodplain is denied. Many of the species in these guilds may be able to use parts of the river channel, such as dead arms, backwaters and local pools as surro-gate floodplain, to a certain extent. In channelized sections of river, even these habitats are lacking and threatened to a greater degree. Modifications to flow, floodplain connectivity and channel modification generally favour a range of eurytop-ic species and introduced species which can attain nuisance proportions in some rivers. Lotic guilds with long distance migratory behaviour [Eupotamonic pelagophilic and lithophilic guilds] disappear when their migratory pathways are disrupted by dams. Because of this, a more holistic approach to the river must be taken as many of the fish faunas associated with flood-plains for at least part of the year use upstream (non-floodplain) areas of the mainstem or tributary rivers for spawning. Failure to secure the connec-tivity of the river systems upstream of the flood-plain can be as damaging to these migratory spe-cies as modifications within the floodplain itself. Rice fields develop a characteristic fauna derived from eurytopic species adapted for air-breathing or tolerant to low dissolved oxygen conditions from the eurytopic riparian and paleo-potamonic guilds. Prominent among these are Oreochromis mossambicus, some cyprinids,

snakeheads, anabantids and clariid catfishes (see for exam-ple Khoa et al. 2005). These species contribute to the gener-al subsistence catches for rice fields but yields of certain of these species can be considera-ble increased from 150 kg ha–1 y-1 to some 380 kg ha–1 y-1 by extensive rice-fish culture using modifications of the rice field structure with ponds and retain-ing ditches. After a hiatus of about a decade in the develop-ment of rice fish culture it is now realised that this practice confers substantial benefits over the more intensively man-aged crops using pesticides and

will probably expand considerably in future.

Modification of flows Water availability and the resulting impacts on flows is currently one of the most serious problems for floodplain fisheries. Many human activities are placing demands on water. It is esti-mated by Clarke, King (2004) that global with-drawals of freshwater from rivers and lakes rose by 6.8 times in the 20th century and that they will rise by a further 31.8% by 2025 (Fig. 1). The majority of withdrawals (70%) are for agricul-ture, 20% for industry and 10% for domestic use. Given the pressures on food supplies worldwide, the increasing industrialisation of large sectors of the globe and rising living standards in many countries these estimates may prove conservative. Three main mechanisms are used for control of waters each of which has implications for flood-plains and their fisheries

Damming Dams are used to regulate flows so that they are available for power generation, irrigation or flood control. The impacts of dams on aquatic hab-itats have been summarized by World Commission on Dams (2000). Their major impact on lowland rivers is that they alter the quantity and timing of flows, storing water within the reservoir behind the dam for later release or abstraction. This means that there is often insufficient water available to flood the riparian wetlands downstream or, if flooding does occur, it may take place at times inappropriate for the needs of the fish communi-ties. In extreme cases the daily pulsed release of water from dams to respond to demands for elec-tricity results in alternation of low flows drying the channel and extreme flood pulses. This practice, known as hydropeaking is especially common in smaller rivers and is extremely detrimental to the fish populations present.

Fig. 1. Total annual water withdrawal (after Clarke, King 2006).

R. Welcomme 174

Abstractions Water may be abstracted directly from the river, especially for domestic water supply and small scale irrigation schemes. Individual abstractions may not be serious but cumulatively can cause severe deple-tion of the water especially at drawdown when the amount of water in the river channels is limited.

Water transfers The growing practice of moving large quanti-ties of water from one river basin to another can have serious consequences to both the donor river, whose floodplains may no longer flood, and the recipient stream where the new flow levels may exceed the capacity of the river/floodplains system.

174

Flow characte-ristic

Change Effect on floodplain Effect on fauna and flora

Timing Delay in arrival of flood

Flooding delayed or shortened

• Influences physical readiness of fish to mature, migrate and spawn

• Synchronises drift of larvae and movement to floodplains and backwaters

• Changes thermal coupling between flood and temperature

Continuity Interruption to flood

May result in incomplete flooding or several flood episodes

• Stranding of fish in temporary pools

• Failure of eggs and larvae to colonize floodplain

• Exposure of floodplain spawning substrates with stranding and desiccation of eggs

Smoothness Increased flashiness Decreased flashiness

May result in incomplete flooding or several flood episodes of short duration

• Exposure of floodplain spawning substrates with stranding and desiccation of eggs

• Fish populations in tributary

rivers may need a degree of flashiness as the spates may serve as triggers for migration and spawning.

Rapidity of change

Overly rapid rise in level Overly rapid fall in level Excessive flow in main channel

Rapid immersion of the floodplain Rapid drying of the floodplain surface and evacuation of floodplain pools

• Failure of floodplain vegetation • Submergence of nests and

spawning sites to too great a depth

• Increased stranding mortalities in

temporary waterbodies

• Can sweep drifting larvae past suitable habitat

Amplitude Decreased depth of flooding Decreased water levels in main channel

May cause failure of flooding of all or part of the floodplain

• Less space for reproduction, refuge and feeding of young and adult fish during flood.

• In extreme cases may prevent fish accessing the floodplain

• Smaller refuges for fish • Increased risk on anoxia • Greater mortality in main

channel through competition and predation

• Interruption of migration pathways

Duration Reduced time when floodplain submerged Increased time when fish are confined to main channel

Less time for development of floodplain vegetation

• Less time for growth of fish • Less time for fish to remain in

floodplain refugia • Greater exposure of fish to

negative condition on channels • Greater risk if desiccation of

channels • Greater exposure to fisheries

Table II. Effects of changes to flow on floodplains and flora and fauna.

World prospects for floodplain fisheries 175

4. Mitigation of detrimental effects

A range of actions are available to mitigate the impacts of human activities on floodplains. These consist of two essential components, firstly ensuring that flow conditions in the river are suf-ficient to maintain the floodplain and to fulfil the requirements of the fish and secondly ensuring that the physical structure of the floodplain and its associated rivers is maintained in a condition suitable for fish.

Environmental flows

Ecologically, lowland rivers are “Pulse regu-lated” in the sense of Odum (1969; 1971) and Odum et al. (1995). In pulse regulated systems the regular alternation in conditions is an essential part of the cycle and every part of the pulse is critical to the fish that live there. However, the flow regimes of rivers are very variable; Puckridge et al. (1998) for instance recognized 10 classes of rivers catego-rized by 11 independent measures including geog-raphy and climate. A number of methods have been used by Governments and Government agen-cies to assess the flows needed to secure the well-being of aquatic ecosystems and organisms. These methods were reviewed by Tharme (2003) although the science continues to advance as new models emerge and new types of environment included. Many of the earlier models were only applicable to smaller rivers and a limited range of fish species. It is only recently that methods suita-ble for large lowland rivers have began to emerge and flow methodologies applicable to the range of rivers types identified by Puckridge et al. (1998) have not yet emerged. Indeed many tropical coun-tries have as yet failed to come to grips with the need to assess environmental flows, although the earlier stages of the assessment process are being considered in some such as Brazil (Benetti et al. 2004) and India (Smakhtin, Anputhas 2006). Among the most promising methods for large f loodplain r ivers is the Building Block Methodology that originated in South Africa (King, Tharme 1994; Tharme, King 1998; King, Louw 1998; King et al. 2002), and has advanced through numerous applications linked to water resource projects to become the most commonly applied holistic methodology worldwide. This method starts from a base flow and adds individual com-ponents of flow that correspond to critical needs of the fish present (building blocks). To apply the method one needs to identify these critical compo-nents. Some of the flow related characteristics of flow regimes in lowland rivers were described by Poff et al. (1997) and Welcomme, Halls (2001, 2004) and are further summarized in Table II. In floodplain rivers flows are critical to various aspects of fish ecology:

- Flows influence biomass of the fish community through density-dependent interactions. They regulate the area of floodplain flooded and the duration of the flood. Through this they change the relative areas of different types of habitat in the channel, as well as accessibility to flood-plains, backwaters, and off-channel waterbod-ies. The main criteria are the amplitude and duration of the flow (Table II).

- Certain levels of flow are critical to trigger events such as migration and reproduction and are related to flow velocity and timing coupled with season, temperature regime, and sometimes day length and/or lunar phase. Critical flows usually depend on the timing and intensity of specific peaks in flooding, especially in tributary rivers (Table II).

- Flows maintain the availability of habitat and environmental quality, including temperature, dis-solved oxygen levels, sediment transport, and environmental support systems such as vegetation and food organisms. They also ensure that there is adequate water in the system to dilute nutrients and avoid excessive eutrophication and ensure connectivity at times of migration and the trans-port of drifting larvae downstream. Habitat flows may operate directly on fish, as for instance in flows that maintain aeration in stony substrates used by lithophilic species, or indirectly via habi-tat, food competition, and availability. The key features are volume of water, flow velocity, con-nectivity, and wetted or flooded area in a system (Table II).

- Exceptionally high- or low-water events may endanger fish, either because of excess velocity at high water or through desiccation and anoxia at low water. These may be expressed as extreme flows occurring as isolated peaks in an irregular hydrograph, or as floods or droughts of very long return periods. Catastrophic events occur even in highly controlled rivers when rainfall exceeds the control capacity of the system. Such flows can have profound impacts on fish communities and on the riverine environment, frequently acting as “reset events” and normally management should seek to avoid such levels of flow.

Not all species react in the same way to the different components of the flow regime. For example, the floodplain spawning Hydrocynus brevis and the main channel spawning H. for-skahlii in the River Niger (Dansoko et al. 1977) alternated in abundance depending on the hydro-logical regime in the river. Agostinho et al. (2004) found dependence on floods to be lower among sedentary species that have parental care (mostly plesiopotamonic and paleopotamonic guilds) than migratory fishes (mostly Eupotamonic pel-agophils). It is probable, therefore that different flood regimes will be required for the different types of river as defined by Puckridge et al. (1998) and for different orders of stream within

R. Welcomme 176

the same river system. Ideally too, variations in flood charac-teristics will be required between years in the same reach and same system in order to favour all species. Figs 2 and 3 illustrate some of the flow elements that are important to fish of various guilds that might be used as building blocks for lowland floodplain reaches and for upland tributary reaches. Two approaches to management of flows are needed to ensure that the flood regime meets the need of the various fish species. Abstraction flow man-agement concerns control of the amount of water taken directly from the river mainly by pumping and diversion. Abstractions mainly take place during the dry season for irri-gation of crops and primarily influence the amount of water remaining in the river at draw-down. Controls here affect minimum flows and should ensure that the main elements indicated in Figures 1 and 2 are adequate to ensure the sur-vival of fish at low water. Active flow manage-ment concerns the release of water from impoundments and concerns both high flow and low flow phases of the hydro-logical cycle. Effectively man-agers are called upon to con-struct artificial floods that are adequate to ensure the survival of functioning fish populations and to protect biodiversity.

Some considerations for guidelines for environ-mental flows in floodplain rivers Where projects and interventions in river basins are likely to alter the amount of water available and the tim-ing of its delivery, arrange-ments should be made to release the flows necessary for the maintenance of the flood-plain, the fish and the fishery. These flows should be calcu-

176

Fig. 2. Schematic flow curve in mainstream floodplain reaches of rivers and some elements that may affect fish that could be used as building blocks to designing flow releases.1. Trigger flows for migration and spawning of floodplain species2. Flows to ensure connectivity to floodplain and induce lateral migration3. Flood flows to determine amplitude and duration of flood: Should flood

near river waterbodies every year and waterbodies further removed from the river at less frequent intervals

4. Flows to ensure connectivity and safe return of fish to the river and flood-plain waterbodies

5. Flows to ensure adequate habitat and water quality in the dry season6. Trigger flows for migration and spawning of in-channel spawners and to

ensure connectivity for in-channel migrations7. Flows influencing survival and drift speed of larval fish in main channel

Fig. 3. Schematic flow curve in upland and tributary rivers and some elements that may affect fish that could be used as building blocks to designing flow releases.1. Channel maintenance flows – extreme high flows that modify channel

structure2. Habitat maintenance flow – flows that perform a specific function such as

cleaning spawning gravels3. Flows that provide longitudinal and lateral connectivity or serve as migra-

tion freshets.4. Water levels over spawning habitats5. Flows influencing larval survival and drift6. Low water flows sufficient to maintain sufficient water and water quality7. Trigger flows for spawning and migration

World prospects for floodplain fisheries 177

lated in terms of the total amount of water avail-able to the system, but should also respect certain norms with regard to timing and to the shape of the flood curve resulting from discharges and releases. Where water is abstracted from the river by pumping and diversions, adequate water should be left at low water to allow fish to per-form their normal dry season functions.

General- All major river systems should have an environ-

mental flows policy based on the needs to con-serve and sustainably use their fish fauna.

- A basin approach should be adopted to flow condition including the main river and the tribu-taries. These are likely to differ as the require-ment of the fish will be different.

- Flow regimes should be maintained as close as possible to the natural regimes for the class of river to which they belong.

Main Channel and floodplainFloodplain- Floods should be induced, preferably every year

but if not every year then at least with sufficient frequency as to allow all species to reproduce within their life spans.

- Flood curves should be as smooth as possible to avoid repeated advances and withdrawals of the water that strand and desiccate eggs adhering to marginal vegetation and nests.

- Rises and falls in level should be relatively slow to avoid over-rapid submergence of nesting sites, failure of vegetation to adapt and grow in the ris-ing phase and stranding of fish during the falling phase.

- High short floods should be alternated with lower but longer ones to favour all groups of species.

- Floods should be adjusted to connect near-chan-nel waterbodies (plesiopotamon) every year and those further away (paleopotamon) every 3-5 years

- Appropriate flows should be incorporated to con-nect floodplain water bodies at times critical to fish migration

- Flows should be sufficient and so timed as to act as triggers for migration and spawning in sensi-tive species

- Where possible flood releases should be timed to arrive after the wetting of the floodplains by local rainfall. This means that the water volume is used to maximum efficiency in flooding rather than in saturating the desiccated soils of the floodplain.

Channels- Flows should be so adjusted that larvae drifting

in the river channels should reach suitable flood-plains where they can complete their develop-ment. Excessive flows mean that the larvae would be carried beyond suitable habitats and

insufficient flows mean that they may not reach suitable areas (Fuentes 1998).

- Adequate dry season flows should be assured. The amount of water remaining in the river is as critical to the survival of the fish population as the flood. Prolonged period of drought will des-iccate the channel of the river and allow it to dry out into a series of de-oxygenated pools.

Tributaries- Flows in tributaries should be sufficient to allow

for migration at critical times- Flows should be so timed as to act as triggers

for migration and induce spawning in sensitive species

- Flows should be so adjusted that larvae drifting in the river channels should reach the main channels of the river where they can complete their migration and development.

- Flows should be sufficient and so timed as to maintain suitable spawning substrates for lithophilic species.

Flow regulation into poldered areas In poldered floodplains control of the water regime within the polder is most commonly used in the interests of agriculture, usually rice culture, but can also be used to advantage by the fishery. For example, Halls (2005), showed that the oper-ation of the sluice gates to polders on floodplains in Bangladesh usually exclude valuable migratory rheophilic species and favour local limnophilic ones. Regulation of sluice gates giving access to poldered areas can control the quantity of fish entering and leaving the enclosed system by improving access to the more valuable migratory, floodplain spawning species.

River and floodplain conservation and rehabilitation

The functioning of floodplain rivers for fish depends on both the form and function of the main river channels and tributary rivers and on the floodplain itself. The physical structure of the environment is closely linked to the hydrology and consideration of both aspects should be combined when developing a conservation or rehabilitation strategy. Given the pressures on rivers and their riparian wetlands for food production and living space, the ideal of maintaining a pristine, wilder-ness condition is impossible to meet, especially as many of the rivers have already been substantially modified. Once altered, occupation of the flood-plain for housing and agriculture makes it extreme-ly difficult to reverse the process. It is thus better to conserve the individual features of existing floodplains and river channels than to have to undergo the social, legal and economic costs of trying to rehabilitate modified systems.

R. Welcomme 178

The association between various groupings of fish and channel-floodplain structure has been noted in temperate rivers such as the Rhone (Amoros et al. 1982) and the Danube (Schiemer, Waidbacher 1992), and tropical ones (as exam-ples Arrington, Winemiller 2004; Pouilly, Rodriguez 2004; Rodriguez, Lewis 1997; Chapman, Chapman 1993; Cordiviola de Yuan 1992). These studies indicate that fish assem-blages sort laterally across the river-floodplain system into guilds that depend on a range of fac-tors, including water depth, distance of water-bodies from the river channel, waterbody sub-strate and the type and amount of vegetation. It is, therefore, essential that certain elements of the system be maintained or restored.

Main river channel The river channels associated with the floodplain are essential as dry season habitats for many river fish species. They not only provide refuge from the often adverse conditions of the floodplain but also act as highways for migrating fish and for the transport of drifting fry. They also support faunas of their own that do not pen-etrate the floodplain but may interact with flood-plain faunas. Many of the larger predators show this behaviour as they mass in the river channels to feed on fish returning from the floodplain dur-ing the falling flood.

Lateral connectivity Lateral connectivity is essential if river fish that use the floodplain for breeding, refuge and growth are to access the floodplain water bodies and the inundated plain itself. Connectivity also ensures that waterbodies separated from the chan-nel fill with water. The degree that this happens depends on the amplitude (height) of the flood.

Floodplain water bodies In Europe, as in the tropics, the degree of connectivity of water bodies to the main river influences floral and faunal assemblages (Amoros, Bornette 2002). Floodplain water bod-ies near to the river or connected directly by channels are generally flooded every year where-as those further away or whose channels have become silted may only connect in years of exceptionally high flooding. The different types of water body tend to contain very different types of fish with those that are flooded less frequently having higher percentages of air breathing or low DO resistant species. Floodplain water bodies may retain water all year or may dry out during long or excessively hot dry seasons. They may also be in the form of seasonal wetlands that are populated mainly by air breathers such as lung-fishes, snakeheads or clariid catfishes or by annual fishes.

The Floodplain The floodplain itself consists of an area of low relief that is covered annually by sheet flood-ing to an extent that varies with the intensity of the flood and is left dry in the low water periods. In their natural state floodplains are usually cov-ered with flood resistant grasslands of by wood-lands or forests but have been systematically adapted by man for various forms of agriculture and grazing of domestic animals.

Some considerations for guidelines for envi-ronmental structure in floodplain rivers Rehabilitation and restoration strategies for large floodplain rivers are still in the early stages of development (Buijse et al. 2002). However, certain general principles may be developed for existing knowledge of floodplain function in the modified systems of the temperate zone and the less constrained systems of the tropic. For exam-ple it is not thought necessary to restore all the floodplain or river channel structure as this would be impractical in most rivers. The dynamics of floodplain fish populations suggests that there is an overproduction of young fish each year (Halls, Welcomme 2004). This provides the basis of the fishery but can also repopulate degraded reaches of river provided the fishery is not excessive. It is therefore suggested that only portions of the river be conserved or restored following the string of beads principle described by Cowx, Welcomme (1998). The following general principles should guide any conservation or rehabilitation projects.

Channels- The main river channel should be conserved in

a healthy state and not over channelized and leveed.

- Riparian structure should be conserved or restored including flowing side channels in braided systems.

- Periodically flowing side channels and dead arms should be conserved or restored as they are a habitat for certain guilds in their own right but also appear to be used as surrogate flood-plain by many species in rivers where the plains are no longer accessible (Schiemer et al. 2001).

- Connectivity should be maintained with upstream tributaries necessary for movements and breeding of migratory fish species

- Where the floodplain no longer floods because the river channel has become deeply incised, in-channel mechanisms such as submersible weirs should be considered.

Floodplains- Floodplain structure should be maintained and

floodplain water bodies should not be filled in. Old waterbodies that have been lost should be restored.

178

World prospects for floodplain fisheries 179

- New water bodies should be created and man-made water bodies associated with rivers, such as marinas, gravel and borrow pits, should be con-nected to the channel where original riparian struc-tures have been lost.

- Channels connecting floodplain water bodies to the river should be kept open and free of silt. Channels that have been sealed or lost should be restored

- Natural mixes of vegetation should be maintained or restored on the floodplain including grasslands and forests.

- Where rehabilitation of the floodplain is impossi-ble because of demands for agricultural uses, par-ticularly in rice growing areas, rice-fish aquacul-ture systems should be explored to maintain pro-duction.

- Embankments for road and rail that may disrupt flows over the floodplain should be equipped with adequate provisions for passage of water and fish.

- It is not necessary to restore all the floodplain as this would be impractical in most rivers. In any case the dynamics of floodplain fish populations suggests that there is an overproduction of young fish each year (Halls and Welcomme 2004). This provides the basis of the fishery but can also repopulate degraded reaches of river provided the fishery is not excessive. It is therefore suggested that only portions of the river be conserved or restored approach termed the string of beads prin-ciple by Cowx and Welcomme (1998).

5. Conclusions

The fisheries of the lowland, floodplain reaches of river systems are heavily exploited. In addition they are under considerable pressure from a range of other human uses that threaten to con-tinue the historic processes of river channel modi-fication and floodplain loss. Whereas the fisheries are resilient in terms of total catch, threats from physical modification and flow abstractions will lead to the disappearance of many of the major flood areas in the foreseeable future. The diversity of fish species and the fisheries will also decline and disappear. Scientific knowledge of the ecology of river fisheries is sufficient to mitigate for some aspects of this process and technical methodolo-gies exist to rehabilitate the river channels and associated riparian wetlands. The limiting factor is the social and political will and economic circum-stance to allow these processes to operate.

6. References

Ago stinho, A.A., Gomes, L.C., Verissimo, S., Okada, E.K. 2004. Flood regime, dam regulation and fish in the upper Parana River: effects on assmblage

attributes, reproduction and recruitment. Reviews in Fish Biology and Fisheries 14, 11-19.

Ago stinho, A.A., Thomaz, S.M., Gomez L.C. 2005. Conservation of the biodiversity of Brazil’s inland waters. Conservation Biology 19, 646-652.

All an, JD, Abell, R, Hogan, Z, Revenga, C, Taylor, B. Welcomme, R. L., Winemiller K, 2005. Overfishing in inland water. BioScience 55, 1041-1051.

Amo ros, C, Richardot-Coulet, M, Reygrobellet, J-L. 1982. Cartographie Polythematique Applique´ a la Gestion Ecologique des Eaux. Editions du Centre National de la Recherche Scientifique: Paris.

Amo ros, C., Bornette, G. 2002. Connectivity and biocom-plexity in waterbodies of riverine floodplains. Freshwater Biology 47, 761-776.

Ant ipa, G. 1910. Regiunea inundabbila a Dunarii. Starea ei actual si mijloacele de a o pune in valoare. Bucharest, Rumania, 318p.

Arr ington, D.A., Winemiller, K.O. 2004. Organization and maintenance of fish diversity in shallow waters of tropical floodplain rivers. In: Welcomme, R., Petr, T. [Eds] Proceedings of the Second International Symposium on the Management of Large Rivers for Fisheries Volume II. FAO Regional Office for Asia and the Pacific, Bangkok, Thailand. RAP Publication 2004/17, pp 25-36.

Art hington, A.H., Balcombe, S.R., Wilson, G.A., Thoms, M.C., Marshall, J. 2005. Spatial and temporal variation in fish assemblage structure in isolated waterholes dur-ing the 2001 dry season of an arid-zone river, Cooper Creek, Australia. Marine and Freshwater Research 56, 1–11

Bar an, E., van Zalinge, N., Ngor, P.B., Baird, I., Coates, D. 2001. Fish resources and hydrobiological modelling approaches in the Mekong Basin. ICLARM, Penang, Penang, Malaysia, and the Mekong River Commission, Phnom Penh, Cambodia.

Ben etti, A.D., Lanna, A.E., Cobalchini, M.S. 2004. Current practices for establishing environmental flows in Brazil. River Research and Applications 20, 427-444.

Bon etto, A.A., Cordiviola de Yaun, E., Pignalberi, C., Oliveros, O. 1969. Ciclos hidrologicos del rio Parana y las poblaciones de peces contenidas en la cuencas temporarias de su valle de inundación. Physis 78, 213-223.

Bot nariuc, N. 1967. ‘Some characteristic features of the floodplain ecosystems of the Danube’, Hydrobiologia 8, 39-49

Bot nariuc, N. 1968. ‘La conception de Gr. Antipa concer-nant la production biologique de la zone inondable du Danube, du point de vue des recherch.es actuelles’, Trav. Mus.Hist, Nat. Grigore Antipa 8, 60-68.

Bui jse, A.D., Coops, H., Staras, M., Jans, L.H., van Geest, G.V., Grift, R.E., Ibelings, B.W., Oosterberg, W., Roozen, F.C.J.M. 2002. Restoration strategies for river

R. Welcomme 180

floodplains in large lowland rivers of Europe. Freshwater Biology 47, 889-907.

er ný, J., Copp, G.H., Ková , R., Gozlan R., Vilizzi, L. 2003. Initial impacts of the Gab ikovo Hydroelectric scheme on the species richness and composition of 0+ fish assemblages in the Slovak floodplain, River Danube. River Research and Applications, 19, 749-766.

Cha pman, L.J., Chapman, C.A. 1993. Fish populations in tropical floodplain pools: a re-evaluation of Holden’s data on the river Sokoto. Ecology of Freshwater fish 2, 23-30.

Cla rke, R., King, J. 2004. The Atlas of water, Earthscan Ltd., London.

Coa tes, D. 2002. Inland capture fishery statistics of Southeast Asia: current status and information needs. FAO Regional Office for Asia and the Pacific, Bangkok.

Cop p, G.H. 1989. The habitat diversity and fish reproduc-tive function of floodplain ecosystems. Env. Biol. Fish. 26, 1-27

Cor diviola de Yuan, E. 1992. Fish populations of lentic environments of the Parana River, Hydrobiologia 237, 159-73.

Cow x, I.G., Welcomme, R.L. 1998. Rehabilitation of riv-ers for fish, Blackwells.

Chr istensen, M.S. 1993. The artisanal fishery of the Mahakam River floodplain in East Kalimantan, Indonesia. III. Actual and estimated yields, their rela-tionship to water levels and management option. Journal of Applied Ichthyology 9, 202-209.

Dan soko, F.D., Breman, H., Daget, J. 1976. Influence de la secheresse sur les populations d’Hydrocynus dans le Delta Central du Niger. Cahier ORSTOM (Hydrobiologie). 10, 71-76

De Graaf, G. 2003. Dynamics of floodplain fisheries in Bangladesh, results of 8 years fisheries monitoring in the compartmentalization Pilot Project. Fisheries Management and Ecology 10, 191–199

Dug an, P., Sugunan, V.V., Welcomme, R.L., Béné, C. 2007. Inland fisheries and aquaculture In: Molden D. [Ed.]. Water for Food Water for Life: A comprehen-sive assessment of water management in agriculture, Earthscan, London, pp 459-484.

Fue ntes, C.M. 1998. Deriva de larvas de sábalo, Prochilodus lineatus, y otras especies de peces de interés comercial en el río Paraná Inferior. Tesis Doctoral. Facultad de Ciencias Exactas y Naturales. Universidad de Buenos Aires. Diciembre de 1998.

Gre gory, R., Guttman, H. 2002 The ricefield catch and rural food security. In: Edwards, P., Little, D.C., Demaine, H. [Eds]. Rural aquaculture. CAB, Oxford, pp. 1–13.

Gri ft, R.E., Buijse, A.D., Van Densen, W.L.T., Machiels, M.A.M., Kranebarg, J., Klein Breteler, J.G.P., Backx, J.J.G.M. 2003. Suitable habitats for 0-group fish in

rehabilitated floodplains along the lower River Rhine. River Research and Applications 19, 353-374.

Hal ls, A., Welcomme, R.L. 2004. Dynamics of river fish populations in response to hydrological conditions: A simulation study. River Research and Applications 20, 985-1000.

Hal ls, A.S. 2005 The use of sluice gates for stock enhancement and diversification of livelihoods. (Fisheries assessment report R8210).MRAG, London

Hol cik, J., Bastl, I. 1977. Predicting fish yield in the Czechoslovakian section of the Danube river based on hydrological regime. Int. rev. ges. Hydrobiol 62, 523-32.

Hol cik, J., Kmet, T. 1986. Simple models of the popula-tion dynamics of some fish species from the lower reaches of the Danube. Folia Zoologica 35(2), 183-91

Hol cik, J. 1988. Influence of hydrological regime and water temperature on the activity and population density of fishes in the anabranches of the Danube. Prace Ust.Rybar.Hydrobiol. 6, 33-58.

Jim enez-Segura, L.F. 2007.Ictioplancton y reprodiccion de los peces en la Cuenca media del Rio Magdalena (sector de Purto Berrio, Antioquia). Dissertation, Instituto de Biologia de la Universidad de Atioquia. 265p.

Jon es, M.J., Stuart, I.G. 2008. Regulated flooplains – a trap for unwary fish. Fisheries Management and Ecology 15, 71-79.

Kar enge, L., Kolding, J. 1994. Inshore fish population changes in Lake Kariba, Zimbabwe. In: Pitcher, T.J., Hart, P.J.B. [Eds]. Impact of Species Changes in African Lakes. Chapman and Hall, London.

Kho a Nguyen, S., Lorenzen, K., Garaway, C., Chamsinhg, B., Siebert, D., Randone, M. 2005. Impacts of irriga-tion on fisheries in rain-fed rice-farming landscapes, Journal of Applied Ecology, 42, 892-900(9)

Kin g, J.M., Louw, M.D. 1998. Instream flow assess-ments for regulated rivers in South Africa using the Building Block Methodology. Aquatic Ecosystem Health and Management 1, 109-124.

Kin g, J.M., Tharme, R.E. 1994. Assessment of the instream flow incremental methodology and initial development of alternative instream flow methodol-ogies for South Africa. Water Research Commission Report No. 295/1/94. Water Research Commission, Pretoria. 590 pp.

Kin g, J.M., Brown, C.A., Sabet, H. 2002. A scenario-based holistic approach to environmental flow assessments for regulated rivers. Rivers Research and Application Vol. 18.

Kol ding, J. 1993. Population dynamics and life-history styles of Nile tilapia, Oreochromis niloticus, in Ferguson’s Gulf, Lake Turkana, Kenya. Environmental Biology of Fishes 37, 25-46.

180

World prospects for floodplain fisheries 181

Kry khtin, K.L. 1975. Causes of periodic fluctuations in the abundance of the non-anadromous fishes of the Amur River. J Ichthyol 15, 826–829

Lae , R., Williams, S., Malam Massou, A., Morand, Mikollasek, O. 2004. Review of the present state of the environment, fish stocks and fisheries of the River Niger (West Africa) In: Welcomme, R., Petr, T. [Eds]. Proceedings of the Second International Symposium on the Management of Large Rivers for Fisheries Volume I. FAO Regional Office for Asia and the Pacific, Bangkok, Thailand. RAP Publication 2004/16, pp.

Lae , R. 1992. Influence de l’hydrobiologie sur l’evolution des pecheries du delta Centrale du Niger, de 1966 a 1989. Aquatic living Resources 5, 115-126.

Lam bert de Brito Ribeiro, M.C., Petrere, M. 1990. Fisheries ecology and management of the jaraqui (Semaprochilodus taeniurus, S. insignis) in central Amazonia. Regulated rivers 5, 195-215.

Las ne, E., Acou, A., Vila-Gispert, A., Laffaille, P. 2008. European eel distribution and body condition in a river floodplain: effect of longitudinal and lateral connectiv-ity. Ecology of Freshwater Fish 17.

Lon eragan, N.R., Bunn, S.E. 1999. River flows and estua-rine ecosystems: implications for coastal fisheries from a review and a case study of the Logan River, southeast Queensland. Australian Journal of Ecology 24(4), 431-40.

Mek ong River Commission 2007. Consumption and the yield of fish and other aquatic animals from the lower Mekong basin MRC technical paper no. 16

Mil lennium Ecosystem Assessment 2005. Ecosystems and human well-being: wetlands and water synthesis. World Resources Institute, Washington, DC.

Mos es, B.S. 1987. The influence of flood regime on fish catch and fish communities of the Cross River flood-plain ecosystem, Nigeria. Environmental Biology of Fishes 18 (1), 51-65.

Nil sson, C., Reidy, C., Dynesius, M., Revenga, C. 2005. Fragmentation and flow regulation of the world’s large river systems. Science 308, 405–408

Nov oa, D.R. 1982. Los recursos pesqueros del Rio Orinoco y su explotación Arte. Caracas. VE. 386 p.

Nov oa, D.R. 1989. Variaciones interanuals de la abundan-cia del coporo (Prochilodus mariae) en el Orinoco medio y su relacion con el reclutamiento. Sociedad Venezolana de ciencias naturales, 111-122.

Obe rdorff, T., Guegan, J-F., Hugueny, B. 1995. Global scale patterns of fish species richness in rivers. Ecography 18, 345-352

Odu m, E.P. 1969. The strategy of ecosystem develop-ment. Science 164, 262-270.

Odu m, E.P. 1971. Fundamentals of Ecology (3rd edition). Saunders, Philadelphia.

Odu m, W.E., Odum, E.P., Odum, H.T. 1995. Nature’s pulsing paradigm. Estuaries 18, 547-555.

Pay ne, A.I., Harvey, M.J. 1989. An assessment of the Prochilodus platensis Holmberg population in the Pilcomayo river fishery, Bolivia using scale-based and computer-assisted methods. Aquaculture and Fisheries Management 20(3), 233-48.

Pet rere, M. Jr., Borges Barthem, R., Agudelo Cordoba, E., Corrales Gomez, B. 2005. Review of the large catfish fisheries in the upper Amazon and the stock depletion of piraíba (Brachyplatystoma filamentosum Lichtenstein). Reviews in Fish Biology and Fisheries 14, 403–414

Pof f, N.L., Allan, J.D., Bain, M.B., Karr, J.R., Prestegaard, K.L., Richte,r B.D., Sparks, R.E, Stromberg, J.C. 1997. The natural flow regime a paradigm for river conservation and restoration. BioScience 47, 769–784

Pom peu, P.S., Godinho, H.P. 2006. Effects of extended absence of flooding on the fish assemblages of three floodplain lagoons in the middle Sao Francisco River, Brazil. Neotropical Ichthyology 4, 417-433.

Pou illy, M, Rodriguez, M.A. 2004. Determinism of fish assemblage structure in neotropical floodplain lakes: Influence of internal and landscape lake conditions. In: Welcomme, R., Petr, T. [Eds]. Proceedings of the Second International Symposium on the Management of Large Rivers for Fisheries Volume II. FAO Regional Office for Asia and the Pacific, Bangkok, Thailand. RAP Publication 2004/17, pp. 243-265.

Puc kridge, J.T., Sheldon, F., Walker, K.F., Boulton, A.J. 1998. Flow variability and the ecology of large rivers. Marine and Freshwater Research 49, 55-72.

Que nsiere, J. [Ed.]. 1994. La peche dans le delta central du Niger. Paris, ORSTOM.

Qui ros, R., Cutch, S. 1989. The fisheries and limnology of the lower Plata basin. In: Dodge, D.P. [Ed.]. Proceedings of the international large rivers symposium. Canadian Special Publication on Fish and Aquatic Science 106, 429–443

Rod riguez, M.A., Lewis, W.M. Jr. 1997. Structure of fish assemblages along environmental gradients in flood-plain lakes of the Orinoco River. Ecological Monographs, 67, 109-128.

Sch iemer, F., Weidbacher, H. 1992. Strategies for the con-servation of the Danubian fish fauna. In: Boon, P.J., Calow, P., Petts, G.E. [Eds]. River Conservation and Management. John Wiley, Chichester, UK. pp.364-84.

Sch iemer, F., Keckeis, H., Winkler, G., Flore, L. 2001. Large rivers: the relevance of ecotonal structure and hydrological properties on fish faunas. Archiv fur Hydrobiologie Supplement 135, 487-508.

Sma khtin, V., Anputhas, M. 2006. An assessment of envi-ronmental flow requirements of Indian river basins. Colombo, Sri Lanka: International Water Management Institute. IWMI Research Report 107, 42p.

Sta nkovic, von S., Jankovic, D. 1971. Mechanismus der fischproduktion im gebeit des mittleren Donaulaufes. Archiv fur Hydrobiologie Supplement 36, 4, 299-305.

R. Welcomme 182

Sve rdrup-Jensen, S. 2002. Fisheries in the Lower Mekong Basin: status and perspectives. MRC Technical Paper No. 6. Mekong River Commission, Phnom Penh, Cambodia

Tha rme, R.E., King, J.M. 1998. Development of the Building Block Methodology for instream flow assess-ments, and supporting research on the effects of differ-ent magnitude flows on riverine ecosystems. Water Research Commission Report No. 576/1/98.

Tha rme, R.E. 2003. A global perspective on environmen-tal flow assessment: emerging trends in the develop-ment and application of environmental flow method-ologies. River Research and Application 19 (5-6): (due September 2003).

Uni versity of Michigan 1971. The fisheries of the Kafue River flats, Zambia, in relation to the Kafue Gorge dam. (Report prepared for UN/FAO: FI:SF/ZAM 11, technical report 1) University of Michigan, Ann Arbor, 161 pp

Van Zalinge, N., Degen, P.O., Ponsri, C., Nuov., Jensen, J.G., Nguyen, V.H., Choulamany, X. 2004. The Mekong River System In: Welcomme, R., Petr, T. [Eds]. Proceedings of the Second International Symposium on the Management of Large Rivers for Fisheries Volume I. FAO Regional Office for Asia and the Pacific, Bangkok, Thailand. RAP Publication 2004/16, pp.

War d, J.V., Tockner, K., Arscott, D.B., Claret, C. 2002. Riverine landscape diversity. Freshwater Biology 47, 517-539.

Wel comme, R.L. 1976. Some general and theoretical considerations on the fish yield of African rivers. Journal of Fish.Biology 8, 351-64

Wel comme, R.L. 1985. River fisheries. FAO Fisheries Technical Paper 262, FAO, Rome.

Wel comme, R.L. 1995. Relationships between fisheries and the integrity of river systems. Regulated rivers: Research and management 11, 121-136

Wel comme, R.L. 1999. A review of a model for qualita-tive evaluation of exploitation levels in multi-species fisheries. Journal of Fisheries Ecology and Management 6, 1-20

Wel comme, R.L. 2001. Inland Fisheries: Ecology and Management. Oxford, Fishing News Books.

Wel comme, R.L., Halls, A. 2001. Some considerations of the effects of differences in flood patterns on fish populations. Ecohydrology and Hydrobiology, 1, 313-323

Wel comme, R.L., Halls, A. 2004. Dependence of Tropical River Fisheries on Flow. In Welcomme, R.L., Petr, T. [Eds]. Proceedings of the Second International Symposium on the Management of Large Rivers for Fisheries Vols I and II FAO Regional Office for Asia and the Pacific, Bangkok, Thailand. RAP Publication 2004/17, 267-284p.

Wel comme, R.L., Bene, C., Brown, C.A., Arthington, A., Dugan, P., King, J.M., Sugunan, V. 2006a Predicting the Water Requirements of River Fisheries. In: Verhoeven, J.T.A., Beltman, B., Bobbink, R., Whigham, D.F. [Eds]. Wetlands and Natural Resource Management. Ecological Studies, Springer-Verlag Berlin Heidelberg 190, 123-154

Wel comme, R.L., Winemiller, K.O., Cowx, I.G. 2006b. Fish environmental guilds as a tool for assessment of ecological condition of rivers. River Research and Applications 22, 377–396

Wor ld Commission on Dams 2000. Dams and develop-ment. Earthscan Publications Ltd, London.

Zal ewski, M., Welcomme, R.L. 2001. Restoration of sus-tainability of physically degraded fish habitats – The model of intermediate restoration. Ecohydrology and Hydrobiology 1, 275-278.

182