Paper 13 Water management for aquaculture and fisheries; irrigation, irritation or integration?

James F. Muir

Institute of Aquaculture, University of Stirling

Summary: Aquaculture and fisheries represent a valuable component in food supply and have a useful role in many development initiatives. The sector is dependent for its potential on suitable water resource management, and there are important interactions to be considered within the wider context of planning, physical development and environmental management. This paper describes some relationships between aquaculture and fishery production, and the water resources which support them, and outlines practical implications for developers and managers.

Introduction

The supply of food and other products through aquaculture and fisheries constitutes one of the most direct and yet highly dependent goals in water resource management. From the local through to the global level, the continuity and interdependence of water resources, their subtle and intimate connections with harvested species, and the many levels of society and economy which depend on these, mean that almost any intervention must have consequences for human well-being. While aquatic ecosystems and the species they support may be tolerant of substantial external change, and may continue to be productive under even the most constrained of circumstances, we see increasing evidence of damage in the wake of the land-use and water resource effects of population growth, and economic and physical development. However, armed with the knowledge of the potential which could be realised, and given the opportunity to incorporate aquatic production goals into water resource management, there are clear prospects for improving the situation. This paper considers the relationships between aquaculture, fisheries and the water resources which support them, and outlines the practical implications for developers and managers.

Fisheries and aquaculture: resources and development

Table 1 summarises the recent status of aquaculture and fisheries production; this is based on aggregated national catch and production statistics and includes food and industrial processed fish (e.g. for fishmeal), crustacea, molluscs and seaweeds, but includes only variable amounts of artisanal fisheries and non-food production, and eg aquatic plants. Most prognoses of fishery supply suggest that capture fisheries are unlikely to grow by more than 1 or 2% per year and will ultimately be limited by physical and biological capacity, by deteriorating environments, and by increasing resource and energy costs of exploitation. Aquaculture by contrast has been widely promoted as compensating and ultimately supplementing traditional fisheries. Unlike fisheries, inputs, production processes and quality of output can be controlled to some extent, and ownership, care and some degree of environmental responsibility can be more readily established. By removing stocks from natural constraints of survival and productivity, and by husbandry and management, production need be limited only by the availability of inputs, land, water, seed, fertiliser and feeds. In 1988, aquaculture production accounted for some 14% by weight of total fisheries production, more than doubled over the last 15 years. By the year 2000, farmed aquatic production might account for some 20 million tonnes per year, some 16-18% by weight, and perhaps as much as 45% by value.

Regional production, and the relative importance of fish in national consumption, and in particular that of quality dietary components, is summarised in Table 2, which indicates that fish products represent a small but important role, and that LDC areas demonstrate a shortfall in supply, partly reflected in the smaller share in total fishery supply. Though not shown here, and in spite of changes in territorial rights in marine fisheries, much of the recorded catch in LDC areas, particularly in Africa, the Pacific and Latin America, is also fished by international fleets, many of which are of developed country origin.

Table 1 Fisheries and aquaculture production

Table 2 Regional production and importance of fish in protein supply (1984-86)

As Table 3 indicates, the bulk of fisheries production comes from marine sources, of which the majority originates from traditional 'open ocean' catches. Thus in Thailand over the 1974-80 period, about 696 of shrimp production was from aquaculture, 22% from small-scale (inshore) fishery and the remainder from larger-scale industrial fisheries, while in Indonesia in 1977 coastal (mangrove) dependent fisheries were about 3% of total marine capture (FAO, 1985). While the balance in this region has more recently changed towards coastal aquaculture, the relative dominance of open-water fishing still exists. As this resource is also ultimately dependent on water management decisions, it is more appropriate here to focus on those resources which are most directly involved in water management strategies; i.e. inland and coastal fisheries and aquaculture. The main focus here is therefore on the more explicitly managed water resources and the more directly controllable fishery outputs.

The role and use of water

The availability and quality of water resources is clearly a key factor in determining where and how fisheries and aquaculture may be supported or developed. As a medium, water provides direct life support, nutrient exchange and food production, as well as providing environmental and behavioural context, and the conditions in which stocks can be controlled, protected or exploited. The relationships involved, although at times complex, depend particularly on the habitat involved, the species and life-cyde characteristics, and the nature of the production system. Table 4 summarises some of the important points.

Table 3 Fishery sector production 1989 by environment/habitat

A fishery or aquaculture system may be vulnerable at any life-cycle stage and to any of these processes, and the absence of satisfactory conditions in only one respect may jeopardise the entire production potential. Nevertheless, many systems can be surprisingly robust, and may respond to external influence by, for example, changing the relative importance or distribution of species mix, rather than an absolute fall in output. However, much quality and value may be lost in the process when simple measures may have permitted or even expanded the original potential. In some cases aquaculture techniques can be used to circumvent or constraints in 'natural' systems, but the existence and mechanisms of these constraints must be recognised and defined before action can be considered.

Water resources and their potential

For the purpose of fisheries and aquaculture production, distinctions can be made between:

· inland water resources; streams, rivers, floodplains, lakes, reservoirs, ponds, groundwater, irrigation supplies

· coastal water resources; ponds, lagoons, estuaries, mangrove areas, inshore reef zones.

Unlike fisheries, production statistics for aquaculture versus resource volume (Table 5) demonstrate low relative utilization of marine resources. Although volume definitions such as these do not account for mean residence time, and e.g. nutrient cycling and accumulation, they serve to illustrate the relative 'loadings'. Surface waters are most intensively used, and of this much is abstracted from rivers or streams, though some is in conjunction with irrigation. A UK survey in 1982 showed the majority of aquaculture to be based on river water, although spring and well supplies were also used, particularly for hatching and growing juvenile fish (Solbe, 1982). In Scotland there has been an increased use of freshwater lochs (lakes) for salmonid cage culture. Cage culture of freshwater and marine fish has expanded widely in recent years; lakes and reservoirs, either for cage culture or 'culture-based fisheries' are increasingly involved (Beveridge, 1987). Groundwater is less important globally, though may be significant in e.g. catfish and trout farming in USA, carp hatcheries in Bangladesh, shrimp farms in Taiwan, trout, seabass and turbot aquaculture in Europe. In the Philippines, the majority of ponds use irrigation water (39%) some use pumped groundwater (32%) or surface runoff and springs (29%). By contrast, industrial or municipal wastewaters are rarely used. Although the intensive aquaculture of high-value species has more scope for using a wider range of water resources than low-input extensive or semi-intensive systems, the general focus on readily available resources increases potential for conflicts with other users.

Table 4 Water resources, habitats and life-cycle requirements

Table 5 Aquaculture production/water resources

Key characteristics of accessible resources with respect to fisheries and aquaculture are as follows:

Surface freshwaters: Surface freshwaters including streams, rivers and temporary bodies such as floodplains; can be particularly vulnerable to human influence, including physical obstruction, impoundment, containment, pollution; support wide range of species including anadromous/ catadromous migratory fish; often subject to complex seasonal/water level interactions, with dramatic short-term changes in productivity, as related to water quality, flow regimes, diversity of flowing water habitat. Conventional aquaculture potential primarily relates to pond, tank and raceway culture; also cages and enclosures in rivers, and hatchery/fingerling production for stocking temporary water bodies. Water quality is determined largely by topography and solubility of catchment, climate and the nature and extent of any pollution sources. Pollutant and natural solute concentrations tend to increase from source; pollution of remote catchments by acid rain and radiation products emphasises pervasive effects of human influence. Amplitudes of short- and long-term flow are important in suitability for aquaculture; e.g. development may be constrained by minimum flows, rather than mean flows.

Static freshwaters: Static freshwaters, such as ponds, lakes, reservoirs, normally have more stable physical and chemical characteristics. Related to flushing time, controlled fertilisation allows highly fertile pond ecosystems, successfully used in many tropical and subtropical regions. These systems are also of considerable interest for cage culture, aquaculture-based fisheries, and water supply or storage. There can however be problems from waste accumulation from intensive cage aquaculture (Beveridge, 1984; Phillips et al.,1985). Water bodies are characterised by factors such as overall shape, average and distributed depth, permanence of volume, their nutrient characteristics and trophic states, incident solar and wind energy, the degree and effects of mixing and turnover of water, seasonal effects, as well as their importance for other social or economic objectives. Overall they are complex systems, heavily biologically mediated in their potential. For fisheries-management based aquaculture, the priorities are to identify the biological capacity and its constraints and to identify the forms of aquaculture most suitable to assist in effective management and development. For more intensive aquaculture, the main priorities are to ensure adequate exchange of water and sufficient 'processing capacity' to handle the additional nutrient loads generated within specified criteria for lake condition. Additionally there should be sufficient space for physical and service facilities involved, and there should be sufficient 'activity capacity' to accept the needs of other users.

Groundwater supplies: Groundwater supplies (artesian or other wells, springs and groundwater-fed ponds) are of interest for aquaculture, particularly for hatcheries, and usually have more stable chemical characteristics than surface waters, although they may contain undesirable levels of dissolved salts and metal ions, low levels of dissolved oxygen and high levels of toxic gases (carbon dioxide, hydrogen sulphide). There is increasing concern about contamination of groundwater supplies by persistent pollutants which may also prove detrimental to aquaculture (there is also concern about aquaculture contributing to this; e.g. nitrogen from operations in Bavaria; Solbe, 1987; Taiwanese shrimp ponds).

Coastal water supplies: Coastal water supplies may be relatively stable, though estuarine coastal water, influenced by large and/or periodic freshwater effluxes is more variable, with a risk from pollution, salinity and temperature fluctuations. Additional problems of marine water include fouling organisms and blooms of toxic algae. Coastal areas may be critically important for intermediate life-cycle stages of marine species and often support extensive artisanal fisheries. Inland areas though accessible to brackish waters are increasingly reclaimed for agriculture and may be extensively damaged by pumping or retention of saline water for aquaculture. Aquaculture can be developed by placing the culture system directly into a suitably sheltered or protected area of the sea, e.g. cage culture, suspended mussel culture or sea-bed ranching of clams; by channelling into ponds, e.g. shrimp farming, or by pumping ashore for more intensive tank-based systems. This last is rarely economic because of the high capital cost of tank construction, and operating costs of pumping.

Industrial sources: Aquaculture has been successful in using water e.g. warm cooling water from nuclear power stations and municipal tap-water supplies. Heated effluents are frequently supersaturated with nitrogen and other gases and other industrial and municipal waters may be contaminated with various inorganic and organic compounds which may require pretreatment or culture of appropriately tolerant organisms.

Wastewaters: Wastewaters hold considerable potential in the longer term, and approaches for wastewater use are being developed widely throughout the world, particularly in association with increased integration of municipal-level wastewater treatment and disposal. Possibilities include the direct use of fish culture in association with wastewater treatment, e.g. as a component in secondary or tertiary lagooning systems - with some limit to the extent of use, e.g. with specific processing or 'polishing' e.g. in clean reservoirs, or to use as feed or fishmeal resource, or to use e.g. for broodstock, where pathogen transfer to succeeding generations could be avoided. There is also more potential to use wastewaters in irrigation schemes with integration of fish culture in supply channels and/or intermediate storage ponds - subject to similar public health/ acceptability constraints. Wastewaters may be used to recharge reservoirs, e.g. for irrigation purposes, stock could be transferred to 'cleaning reservoirs' prior to harvest. Although technical means are available for carrying out these approaches, there remain problems of public perception and acceptability, even if wastewater is treated, let alone using it for direct production of food. However it can be noted that adventitious sewage enrichment already contributes to fish production in many water bodies.

Water quality criteria

A primary constraint for the use of water is its quality; equally the processes of aquatic production, particularly the more intensive forms, may bring about quality changes which may affect the system itself or its suitability for other uses. Although a water supply can be a source of toxins it is also important for receiving and dispersing excretory products and other waste materials which are harmful if accumulated. Water-borne organisms such as phytoplankton and bacteria also control quality in slow-flushing systems; water itself may detoxify wastes e.g. the pH of most freshwater and sea water allows excreted ammonia to be rapidly converted to nontoxic ammonium, whereas terrestrial animals use energy-intensive mechanisms to remove ammonia. Quality criteria are commonly divided between external (exogenous) and internal (endogenous) factors:

· External: inherent feature of water supply, usually from outside the site, independent of aquaculture or fishery activities. Cost-effective control is difficult but may be carried out by: choice of site, chemical treatment (e.g. Iiming, flocculation, oxidation, chlorination), physical treatment (e.g. screening, filtration, settling ponds for solids), reusing the water supply

· Internal: produced by stock itself, or by activities associated with fishing or culture (e.g. harvesting, feeding or disease treatment). Particularly in intensive farms with heavy feeding and high stock densities, where internal loading of waste food and excretion may be as important for stock health as external loading, it becomes crucial to match stocking and feeding with species tolerance to: ammonia, nitrite, suspended solids, biochemical oxygen demand, anaerobic conditions, treatment chemicals.

The effects of water quality depend on the species and life-cycle stage (e.g. Iarvae, fry, fingerling, grow-out etc.). As conditions deteriorate (one or more criteria reach limits), stocks become stressed. It is important to consider both lethal and sub-lethal effects, and factors such as handling, crowding and disease treatments. The extreme response to high levels of stress is mortality. Below this level, stress can:

- affect stock behaviour (e.g. predator avoidance, feeding behaviour)
- reduce growth/maximum size/food conversion efficiency
- impair reproduction (e.g. low egg fecundity, spawning success)
- reduce tolerance to disease
- reduce the ability to tolerate further stress.

The last two are crucial, because they imply that disease can follow any non-lethal but stressful pollution and that a combination of two or more factors, for example handling and high ammonia levels, can be more damaging than (a possibly more serious) exposure to one factor alone.

Specific water demand

Water use and production systems

In fisheries, where production is more commonly considered to be incidental to the presence and quality of water, there is little concept of specific water demand, i.e. allocated to fishery production as opposed to other use. However, where, for example, river levels are managed for flood control or water supply, where lakes or reservoirs are controlled for power generation or irrigation, or where coastal areas may be affected by land reclamation or by salinity control, the overall relationships between volume, flow and fishery biomass may need to be entered into management decisions.

At a simple level, these may be defined by productivity relationships; e.g. production, per time period in kilograms per unit area, or per standardised unit of river length. While annual production is most common, a shorter time period may be more appropriate for seasonal or periodic water supplies. As noted earlier, availability, flow, level and/or quality characteristics may be particularly important at specific life-cycle stages, and production may be more closely related to these critical points than to longer-period features. Published productivity levels range with fertility and ecosystem characteristics but are typically of the order of 50-500 kg/ha/year in rivers, lakes and coastal water areas, and of the order of 1-5 kg/ha/year in open sea. However it is also important to consider the aggregation characteristics of fisheries, where apparent production may be greatly in excess of this in localised areas, due e.g. to seasonal feeding, migratory activity, and that of the wider water body correspondingly less.

In aquaculture, relationships with water demand are usually more clearly defined, as according to production system. This may be defined by the species (e.g. because of environmental or feeding requirements, or market value), which will in turn define the sites required. In other cases, a certain site may define which systems are feasible and may in turn suggest particular species. Alternatively a specific system will support certain species and require particular sites. Systems are usually defined according to two main criteria:

· Type of system: i.e. ponds, lagoons, tanks, raceways, cages, enclosures, rafts, ropes, trestles etc.

· Intensity of production. typically described as extensive, semi-intensive, intensive, etc.

Typical system characteristics are given in Table 6. These will clearly have a significant bearing on capital and operating requirements, and hence the project costs, and a variety of capital and operating input mixes can be defined.

Table 6 Aquaculture systems; outline design characteristics

Water demand in aquaculture systems

Results from a range of production systems are shown in Table 7. Quantities per tonne of production vary widely, determined primarily through managed water exchange associated with intensity of production (i.e stocking density and use of feeds and fertilisers), but also through physical factors such as seepage and evaporation. It must of course be clarified that these figures do not (with the exception of static, high-evaporation systems) represent actual consumption, but account the quantities of water which have to be made available to bring about specified forms of production. The main implications are then the opportunity costs and the quality changes involved.

The table shows that the lowest water requirements are for the air-breathing walking catfish Clarias batrachus in Thailand. Above this, various tropical and subtropical pond systems have comparatively low unit water demand. The highest demands are for salmonids (and carp) in intensive flowthrough systems, with water being the only vehicle for supplying oxygen and removing metabolites. In ponds, seepage typically ranges from 2-25 mm/day, depending on soil type, pond surface area and wall construction. Losses through evaporation may reach 25 mm/day, though in the subtropics are more typically around 5 mm/day. With total losses of 10-20 mm/day, each hectare of pond will 'consume' 100-200 m³ water per day. In Israel, total requirements for ponds are estimated by Hepher and Pruginin (1981) to vary between 35 and 60 000 m³/ha/year to maintain an average depth of 1.5 m throughout the 240 day growing season.

Table 7 Water requirements per tonne of aquaculture production

Aquaculture has a comparatively high water demand (Table 8), may in relative terms add only moderate value from nominal use of water resources, and might therefore have a low priority when competing with industrial or agricultural users. However value versus water-quality change may be higher than for some competing users - see Table 9 (Muir and Beveridge, 1987).

Table 8 Typical water requirements of industry and agriculture compared with selected aquaculture systems

Table 9 Production loading from aquaculture and other industrial and agricultural sources

Water quality changes

As fisheries are by definition associated with the natural productivity of water bodies, there are no specific water quality changes involved in its output, though there may be localised effects of fishing activity, such as sediment disturbance, and the more serious matters of poisoning or the use of explosives. There may also be water quality implications of measures associated with attempts to conserve or enhance fisheries, such as reservoir or river-flow manipulations. By contrast, aquaculture is often associated with water quality changes, which tend to increase with intensification. These have been sufficiently well studied to allow prediction of the nature and scale of potential impacts in most situations (e.g. Hakanson et al., 1988). Impacts on the quality of receiving waters and on the associated sediments are of most concern, particularly in relation to the mass input of nutrients. There is also a variety of other effects, such as chemical treatments and genetic pollution, as discussed below.

Nutrients: Aquaculture effluents comprise waste products (urine and faeces) and uneaten feed, containing nitrogen and phosphorus in both soluble and insoluble forms. If in sufficient quantities relative to the volume of the receiving waters, this may lead to hypernutrification and stimulate undesired growth of phytoplankton, seaweeds and aquatic plants. However, due e.g. to bioavailability and potential limiting status of these nutrients, effects vary widely with the characteristics of the ecosystem.

Sedimentation: The solid fraction of effluents may deposit as sediment in close vicinity of the farm site. This may result in physical changes in nearby sediment, and in some cases, highly localised changes in the community of sediment-dwelling animals and plants. The chemical composition of sediments will also change, becoming enriched with organic material and nutrients (Kupka-Hansen et al., 1991). Elsewhere, aquaculture installations will themselves act as controllers of exogenous sediment; e.g. trapping sediment in coastal supply sluices or in valley-floor ponds, and in return discharging some or all of these during periods of flushing or drainage.

Salinity changes: Where saline water is being pumped from coastal wells, problems may arise through salinity intrusion in the aquifer, affecting the aquaculture operation itself, and contaminating the aquifer for other users. The surface discharge of saline wastes from these and from coastal pumped farms may also have a substantial and longer-term effect on soil salinisation, and its potential for e.g. conventional crop production.

Chemical treatments: The use of disease chemicals, including compounds ranging from antibiotics such as oxytetracycline to parasite treatments such as malachite green and organophosphate pesticides (reviewed in NCC, 1989 and 1990), is widespread, particularly in intensive farming operations, and requires careful, compound-specific regulation and monitoring.

Escapes of stock: Fish cages may become damaged (e.g. by storms) and ponds or tanks may flood, resulting in the release of large numbers of stock. Smaller numbers may also be periodically lost to the local environment. If the species is indigenous the impact may be minimal, though if the scale of release is large relative to the size of the local population, 'genetic pollution', or dilution of the local gene pool may occur. If the species is imported, it may establish within the local environment and possibly spread to other areas with unforeseen consequences for the native aquatic communities.

Engineering and other implications

The previous sections have identified the main water management issues involving fisheries and aquaculture production and have commented on a range of constraints and effects. The relationships between the fisheries sector and other development activities, particularly those involving engineering and other forms of physical change, are further described in Table 10.

Table 10 Engineering, land use and fisheries; basic water management inter-relationships

Table 10 continued

Social, legal and other issues

The use of water for aquaculture or fisheries is often regulated either specifically, or as part of the right for fishing or agricultural purposes (van Houtte et al., 1989). Until recently there have been few constraints in practice. However, where water is scarce conflicts can arise, and thus in Israel, where water is now priced, it has been found to be more profitable to use water for crop irrigation than aquaculture. Fish ponds have been rebuilt to form irrigation or dualpurpose fish culture/irrigation reservoirs and fish culture has been restricted to only part of the year (Hepher, 1985; Sarig, 1989). In many areas in the developed world, aquaculture, especially if involving river abstraction for pond farming, or the use of lake and coastal areas for cage sites, is being constrained by real or perceived effects on other users. In many traditional societies, water rights are carefully and often sparingly controlled, and access to these for nontraditional activities may be very difficult.

In Cyprus, problems of water allocation for aquaculture have recently emerged (Muir and Baird, 1992). Its dry Eastern Mediterranean climate means that water is crucial to the society and economy, and many traditional rights attach to water use and management. So strong are these, and the politically important rural societies' attitudes to water, that they distort rational allocation of what is in every respect a scarce resource. In spite of a recognised need to conserve, policies do little to constrain use; charges, particularly for irrigation and other agricultural purposes are substantially below costs of supply. Fisheries and aquaculture have little or no traditional status and are therefore almost completely excluded from resource allocation. Groundwater resources are heavily used and localised declines in water-tables are common. Although major groundwater development has taken place, with substantial investment in conveyor and recharge projects, much of this has been immediately absorbed into intensive agricultural production and domestic or tourist water supply. Problems of salinisation of coastal zones have been reduced by the various recharge schemes but supply is precariously balanced with little for aquaculture unless integrated, e.g. with irrigation. Reservoir storage capacity has reached 297 million m³ out of a total 450 million m³ net surface runoff. For irrigation supply reservoirs, cage-based trout farms have been established and there is a programme of stocking and restocking, mainly for sport fishing. However, capacity is now heavily absorbed, as around 90% of reservoir volume is stocked, 25-30% is used for aquaculture and reservoirs were until recently drawn down to extreme low levels. More reservoirs may be converted to potable supply and fish farming may not be able to continue. Although coastal seawater aquaculture has developed, waste nutrients were being blamed for a troublesome growth of filamentous algae which caused serious disturbance to coastal areas and major concern for tourist interests. Though the problem was more likely a response to the far greater enrichment from intensive agriculture and domestic and tourist wastes (Baird and Muir, unpublished), such was the public concern, that aquaculture was placed under an effective moratorium.

The changing nature of aquaculture may also affect its impacts and acceptability. Until the late 1960s, Taiwan's aquaculture was based heavily on traditional Chinese integrated polyculture in fresh and brackish water. Due to industrial development, consumer buying power increased, moving demand from traditional low-cost products such as carp and milkfish to more intensively produced quality species like eel, grouper, and shrimp. Aquaculture was based on small family units of 1-3 ha; shortage and very high cost of land have meant that most systems have intensified, requiring constant management - uneconomic if families were not involved and labour had to be hired. In 1981 aquaculture took 11% of Taiwan's total water consumption, some 21 million m³, mostly from shallow (about 8 m deep) boreholes, giving unpolluted and temperature stable water (25-26°C). In the last five years uncontrolled development for intensive shrimp farming stretched this resource, and overpumping has led to land subsidence, sometimes exceeding 2 m. Pollution problems and disease have also affected the industry and contributed to the collapse of production from 80 000 t in 1987 to 20 000 t in 1988. Increased competition from mainland China and throughout the rest of Asia has created further uncertainty. However, recent government control over licensing water use and effluent discharge, combined with better management practices, may help the industry as it evolves in the future. Meanwhile, this experience provides a good cautionary example for the planning and development of aquaculture (Phillips et al., 1988).

The role of local and international development and investment projects might also be noted. Though many agencies understand fisheries and aquaculture to make a contribution to basic development aims, investment has often tended to support projects aimed at export production and foreign currency earnings - valid perhaps for structural readjustment - or has been imprecisely targeted to potential beneficiaries. Heavy commercial investment (local and international) has also tended to focus on the more obviously profitable areas of aquaculture, often at considerable cost to local resources and environments. Thus by the late 1980s, penaeid shrimp farming was Ecuador's most important export after oil and bananas, and in Bangladesh shrimp exports became second only to jute. In both areas, however, environmental and social issues cause increasing concern. Rather than banishing such development, or denying its positive impacts, there is perhaps a more pressing need to improve efficiency, control impact and reduce conflict.

Improving efficiency

There is an increasing trend towards intensifying fishery and aquaculture production. This has resulted in different approaches - and with most species (e.g. salmonids, channel catfish and shrimps) resulting in increased water use. Intensive, high flow-rate systems are rare in developing countries; most are in urban areas or where water is scarce; high water use is common only where water is plentiful and/or there are no laws or costs to restrict use. There would appear to be ample scope for improving efficiency of water use; an EIFAC survey of European freshwater farms (Alabaster, 1982), showed water consumption per tonne to vary more than 100 fold. Where conservation of water is necessary or desirable and cost penalties are enforced, water use can usually be reduced. Thus in Israel intensified culture of common carp and tilapia has accompanied a decrease in water use (Hepher, 1985; Sarig, 1988), primarily because of the high cost of water, and the priority given to water conservation. Species tolerating poor water quality (e.g. catfish) can also be produced with little increased water use.

Where possible, there may be distinct advantages for fisheries and aquaculture to be integrated with other industries or with agriculture. There may also be benefits where for example aquaculture can add value to a water resource, e.g. with fish production in sewage fed ponds, to improve water quality and generate income from a waste material (Edwards et al., 1987; Little and Muir, 1987), or where improved fishery yields can add discernably to the benefit streams of a capital project. There may be opportunities for environmental enhancement or rehabilitation; in Scandinavia researchers are attempting to improve water quality in acidified freshwater lakes through selective 'pollution' by cage culture (Solbe, 1987).

For fisheries the main inputs towards improving efficiency lie in improving the utilisation of water bodies, controlling stock yields and economic value, possibly using hatchery stock, nursery techniques or other aquaculture-based methods to supply or supplement production. Such approaches would be based on the use of traditional fisheries management techniques, together with assessments of local social and economic characteristics to identify target groups, use rights and custom, and to define cost-effective and sustainable management methods. For aquaculture, the main aims would involve more flexible use of available permanent or temporary water resources, improved nutrient transfer from inputs to product, and practices which minimised external impact. This can be achieved in a range of ways; by modifying construction and design, amending operation and management, or by applying technical treatment.

Ideally, aquaculture operations should be conceived from the outset to minimise impact, though for a variety of reasons - lack of incentive, lack of awareness of problems and/or site or operational characteristics - many problems are dealt with retrospectively, through modification. Design and construction factors include: at the general level - designing the system to maximise payoff between productivity and environmental effect, i.e. to get the best, most efficient performance with as little as possible environmental 'cost'; at the specific level - careful site selection to ensure minimum impact on sensitive areas, whether visually, physically in terms of cover/erosion, or biologically in terms of ecosystems and/or individual species; where necessary, providing adequate mixing, dilution and dispersal; if needed, providing waste and impact management measures; designing to minimise releases, contact with external water supplies, transfer to other stock, disease contamination; limiting construction disturbance, keeping top-cover, avoiding discharge of materials into water courses.

In operation and management, control can be exerted at each step in the production cycle; e.g. with better formulated, more digestible and water-stable feeds and better feeding methods. There is considerable economic pressure to minimise feed losses - this was in part responsible for a more than 50% drop in effluent solids loadings per unit production, observed from UK land-based farms between 1980 and 1987 (Solbe, 1987). Other factors include: care over incoming stock and genetic consequences; stock density/system management control; feed control to optimise efficiency rather than growth; use of fallow sites; controlling the use of chemicals and the transfer of resistance, predator control; changing ways of disposal of mortalities; restricting numbers of different stocks and, hence, the possibility of disease transfer.

Water treatment may be used to improve stocking densities and reduce requirements. This already exists in extensive and semi-intensive ponds which are managed to promote phytoplankton growth as fish feed organisms or as a carbon source at the base of the food chain. Phytoplankton have the added benefit of producing oxygen by photosynthesis and assimilating waste metabolites. However, as stocking densities increase, more precise water quality control is needed to ensure adequate supplies of dissolved oxygen and efficient dilution and assimilation of metabolites and aerobic metabolism of organic wastes (Hepher, 1985). The tendency towards increasing water flows and decreasing water retention time in more intensive systems results in a loss of many of the internal water quality control systems and, inevitably, leads to even greater water demand. Flow regimes and production plans can be improved by design to maximise stocking per unit flow. Systems can also be intensified by varying degrees of water treatment, from simple aeration through to recycling using biological filtration, solids removal and oxygenation. As well as improving water economy, such systems can provide controlled environments, e.g. for temperature and/or salinity. They involve additional costs, though this may be justified by improved performance or the production of valuable species or stocks.

Water-use efficiency of existing systems can be improved with a number of means, including reducing seepage losses and intensifying use and re-use of water. Water loss from ponds can be reduced by better siting and management, and further minimised by increasing depth. Thus Sarig (1988) reports water losses from ponds to be 6-10 times higher (35 000-60 000 m³/ha/year) than those from deeper dual-purpose reservoirs (-6000 m³/ha/year). Seepage should also be reduced to control adverse production effects of high water exchange on alkalinity. Concretelined ponds as developed in Taiwan and Israel (Hepher, 1985) can help in intensifying water use, though waste output may be greater as there is much less opportunity for biotransformation. In comparison with more intensive flow-through systems, simple pond farming could also be proposed as being more water efficient, although large land resources may be required because of relatively low areal production. There may also be significant management problems with large volumes of water and low densities of stock. These may be partly overcome by culturing fish in cages in large ponds, as is now developing in several regions (e.g. Hungary and Israel). Water can also be conserved by culturing fish species with improved tolerance to low dissolved oxygen and high levels of waste metabolites (e.g. tilapias).

Routine water conservation can also be implemented. Deepening ponds to increase storage has been recommended to conserve seasonal water resources (Hepher and Pruginin, 1981). Loss of water during harvest can be reduced by avoiding draining, although this may cause longer-term deterioration of water quality. Integrating agricultural water with aquaculture can benefit both (Little and Muir, 1987). In Israel, integration has been forced on fish producers because of the high value of irrigation water, and more than 20% of fish ponds have been turned over to integrated fish/irrigation schemes (Sari", 1988). This has involved a shift to more intensively managed ponds or to cage and pen culture in larger irrigation reservoirs. There is also scope for increasing the fertility of irrigation water, to improve yields or maintain existing yields with lower levels of fertilisation. Fish-pond water has been used as a fertiliser for maize (Edwards, 1980) and for rice and vegetables in the Philippines, and pond sediments and water have been widely used as fertiliser for many years in China. In irrigated and rainfed rice fields in tropical regions, fish culture also offers potential for increasing output value. Water costs for fish culture are very low as the infrastructure and supply is already available, and management of soil and water is already present. Economic benefits through production of fish and increased rice yields can be significant. Fish culture can also improve the general availability of water resources. Stocking of irrigation canals with macrophyte feeders such as grass carp (Ctenopharyngodon idella), java carp (Puntius gonionotus) and Tilapia zilli has also been used in several regions for removing machrophytes and improving water flow to traditional agricultural crops (Little and Muir, 1987).

Cage and pen culture do not require water abstraction and so offer good scope to improve water resource use. Cages have been used successfully in irrigation reservoirs (Beveridge and Phillips, 1988) and in a wide range of lakes, reservoirs and running waters (Beveridge, 1987). Problems associated with environmental impact are a continuing concern, although greater understanding of the interactions between cage culture, open stocking levels, environment and water quality should assist optimal use. Aquaculture can also be integrated with drinking-water supply; cages of filter feeding bighead carp have been used to remove phytoplankton from potable water reservoirs in Singapore (Beveridge, 1984). Integration of open water stocking or extensive cage culture with intensive aquaculture also offers scope to minimise or control eutrophication.

In coastal areas, saline water extraction and discharge can have considerable negative impact, and improved water use can be an important objective. The effects of saline intrusion or deposition can be reduced by positioning wells and screens carefully and limiting multiple well development, and by controlling extent, locations and/or timing of deposition. In freshwater zones it may be appropriate to recharge water from the aquaculture operation or other sources, using e.g. percolation pits or trenches. As well as recharging the groundwater resource, this can be useful for cleaning and decontaminating mildly affected supplies. Particular care has to be taken to avoid excessive nutrient loading and clogging of substrate pores, and to avoid direct contamination and immediate deoxygenation problems; it is useful to keep recharge areas well apart.

Goals for water management

Prospects for development

What are the prospects and constraints for maintaining and developing the fishery sector? Fisheries and aquaculture development of even the simplest form needs resources, management and reasonable social and economic stability. Areas where malnutrition is related to drought and crop failure are unlikely to have resources for aquaculture but may be able to support very important seasonal fisheries. In Africa's case, problems of production are partly bound up with internal political, social and economic changes, and with the external pressures of changing trade conditions. Asia by contrast, has a long history and association with fishery and aquaculture, and vigorous economic growth in relative political stability. In many areas, the two activities might be complementary; traditional dependence on capture allows consumption at significantly lower prices than are common for aquaculture, while the latter supplies more developed urban and export markets.

Where aquaculture has traditionally produced lower-cost supplies - e.g. the inland water carp culture of China, SE Asia and India, and the brackishwater milkfish production in Indonesia and the Philippines, products are either increasing in price beyond the means of the poorer rural communities, or are being supplanted by other higher-value cash crops. Within a given range of aquatic resource potential, lower-value products such as seaweeds, molluscs and local food fish may lose out to crops for export or for the more prosperous sectors of the society. While lower-value production from simple traditional methods continues to be significant in volume terms, its importance may become eclipsed. This section considers some of the implications for water demand.

The potential for water demand

The needs for water in the fishery and aquaculture sector have already been commented upon. As an illustration of the longer-term implications and based on aquaculture production only, Table 11 summarises the water involved in meeting expected production levels for the year 2000.

This usage would obviously be subject to the provisos already discussed but the table clearly indicates the scale of requirement, and the implications for water resource needs, and for the improved management of those resources involved. Additional allocation for aquaculture implies an equivalent change in availability or quality for other uses, whether for fisheries and environmental management, for agriculture, industry or potable supply. As suggested above, most of this would occur in the area of semi-intensive aquaculture, in freshwater, and would involve the management of surface-water supplies.

Themes for development

It is obvious that aquaculture and fisheries require to be seen within the wider views of e.g. aquatic resource use, land-use development, rural, regional, general economic development. Though it has frequently been claimed that aquaculture is very 'complementary', profitably using resources which are otherwise unused, all forms of aquaculture require the organisation and mobilisation of resources, whose use represents a removal from use elsewhere - whether productive, recreational, aesthetic. Capture fisheries, based on natural aquatic habitats and their productivity, are also traditionally considered non-demanding of resources and may even respond positively to additional 'waste' inputs. However, the opportunity costs of maintaining or re-establishing fishery production, and the importance of the communities this serves, might increasingly be brought into benefit assessment. Both in generalised planning and in specific design, it is important to be aware of these resource implications, and to consider the capacity of various locations and systems to supply resources. Increasingly, questions of aquaculture and fisheries in development will feature in the wider technical context, involving the understanding, management and rehabilitation in terms of individual ecosystems or ecotopes such as lake and river systems, lagoons, mangroves and coastal reef, in terms of understanding the characteristics and potential of ranges of production systems, and in terms of better characterisation of water resources, including domestic, agricultural and industrial 'waste' water. These components would then form part of a set of larger, higher-order multiple-linked systems and management approaches, such as:

· Watershed management - particularly for inland fisheries - involving issues such as water retention, local albedo effects and soil stabilisation; sediment collection, maintenance of nutrients and productivity; groundwater use, re-use and quality management; multiple use of reservoirs, lakes, and the development of multi-use ecological and economic models; incorporation in larger-scale projects for flood control, irrigation, water supply and wastewater treatment.

· Coastal area management - for brackish-water and marine aquaculture, coastal and lagoon fisheries - involving issues such as water exchange, physical development, sedimentation and salinisation, groundwater use, nutrient and chemical management.

Table 11 Projected water use by major species/production system

There is considerable scope for greater integration between fisheries, aquaculture and water resource management. Where social and other conditions permit the management and regulation of fisheries, or where existing natural resources are threatened by other activities, aquaculture techniques - broodstock selection, hatchery production - may become more important in supporting or enhancing fisheries. 'Biomanipulation', using aquaculture-based knowledge or techniques to enhance productivity of natural water bodies may widen opportunities for lower-cost food production. These techniques, rather than competing with other forms of aquatic production, should provide better means for managing aquatic resources, whether in terms of understanding natural populations of phytoplankton or fishery stocks, supplementing production with habitat structures and controlled releases of nursery stock, or manipulating environments for restoration or enhancement. This understanding should be broad and open-minded, and would comprise many different aspects of traditional planning and engineering, together with sizeable elements of biology, sociology and economics. We might aim for a similar understanding to that of agricultural systems, of the means to define the possibilities, what resources they require, and how systems can be successfully set up to meet our objectives. In spite of a difference of several thousand years' experience, there is no reason not to attempt similar approaches in aquatic resource use, and to develop the kind of practical framework in which we can research and develop, plan and manage.

References

ALABASTER, J. S. (1982) Report of the EIFAC Workshop on fish-fann effluents. Silkeborg, Denmark, 26-28 May 1981. EIFAC Technical Paper 41, FAO, Rome.

BAIRD, D. J. and J. F. MUIR, (Unpublished) The impact of the Telia Agua Marine Fish Farm on the Coastal Environment at Liopetri at Cyprus. UNDP/Cyprus Government unpublished report.

BEVERIDGE, M. C. M. (1984) Cage and penfishfarming. Carrying capacity models and environmental impact. FAO Fisheries Technical Paper No. 255.unpublished

BEVERIDGE, M. C. M. (1987) Cage Aguaculture. Fishing News Books, Surrey, England.

BEVERIDGE, M. C. M. and PHILLIPS, M. J. (1988) Aquaculture in Reservoirs. In: Proceedings of the IDRC Workshop on Tropical Reservoir Management, Kathmandu, Nepal, 21-28 November 1987. S. S. de Silva (ed) IDRC, Ottawa, Canada.

BOYD, C. E. (1982) Water Quality Management for Pond Fish Culture. Developments in Aquaculture and Fisheries Science, Vol. 9. Elsevier, Amsterdam, The Netherlands.

EDWARDS, P. (1980) Food Potential of Aguatic Macrophytes. ICLARM Studies Review 5. ICLARM, Manila, Philippines.

EDWARDS, P., POLPRASERT, C. and KOK, L.W. (1987) Resource Recovery and Health Aspects of Sanitation. Asian Institute of Technology Research Report, No. 205.

FAO (1991) FAO Fisheries Circular (Provisional) (815 Rev. 3. Provisional) Rome, Italy.

HAKANSON, L., ERVIK, A., MAKINEN, T. and MOLLER, B. (1988) Basic Concepts Concerning the Environmental Effects of Marine Fish farms. Nord 1988:90, Nordic Council of Ministers, Copenhagen.

HEPHER, B. (1985) Aquaculture intensification under land and water limitations. Geojournal, 10, 253-259.

HEPHER, B. and PRUGININ, Y. (1981) Commercial Fishfarming. John Wiley & Sons, New York.

KUPKA-HANSEN, P., PITTMAN, K. and ERVIK, A. (1991) Organic waste from marine fish farms - effect on the seabed. In: Marine Aquaculture and Environment. MAKINEN, T. (ed) Nord 1991:22, Nordic Council of Ministers, Copenhagen.

LITTLE, D. and MUIR, J. F. (1987) A Guide to Integrated Warm Water Aquaculture.. Institute of Aquaculture, University of Stirling, Stirling. (Chapter 7, Irrigation/Water Resources)

MUIR, J. F. and BAIRD, D. J. (1991) Environmental Management of Aquaculture Development. Technical Report, TCP/CYP/9152, December. 1991.

MUIR, J. F. and BEVERIDGE, M. C. M. (1987) Water resources and aquaculture development. Arch. Hydrobiol. Beih. 28, 321-324.

NCC (1989) Fish-fanning and the Safeguard of the Natural Marine Environment of Scotland. Nature Conservancy Council, Edinburgh.

NCC (1990) Fish-fanning and the Scottish Freshwater Environment. Nature Conservancy Council, Edinburgh.

NEW, M. B. (1991) Turn of of millennium aquaculture: navigating troubled waters or riding the crest of the wave? Journal of the World Aquaculture Society, 22(3), 28-49.

PHILLIPS, M. J., BEVERIDGE, M. C. M. and CLARKE, R. M. (1988) Impact of aquaculture on water resources. In: Aquaculture and Water Quality. Advances in World Aquaculture, Vol.3, BRUNE, D.E. and TOMASSO, J. R. (eds), World Aquaculture Society.

PHILLIPS, M. J., BEVERIDGE, M. C. M. and ROSS, L. G. (1985) The environmental impact of salmonid cage culture on inland fisheries: present status and future trends. Journal of Fish Biology, 27 (Supplement A): 123-137.

SARIG, S. (1988) Is an increase in yields in intensive commercial fish farming always accompanied by an increase in profitability? The Israeli Journal of Aquaculture, 40, 122-124.

SOLBE, J. F. de L. G. (1982) Fish farm effluents: a United Kingdom survey. In: Report of the EIFAC Workshop on Fish-Fann Effluents. ALABASTER, J. S. (ed), Silkeborg, Denmark, 26-28 May 1981. EIFAC Technical Paper (41).

SOLBE, J. F. de L. G. (1987) Report of the European Inland Fisheries Advisory Commission Working Party on Fish Fann Effluents. Water Research Centre, Medmenham, UK.

Van HOUTTE, A. R., BONUCCI, N. and EDESON, W. (1989) A Preliminary Review of Selected Legislation Governing Aquaculture. ADCP/REP/89/42, UNDP/FAO, Rome.
WILSON, E. M. (1984) Engineering Hydrology. 3rd edn. Macmillan, London.

Discussion

In the discussion the need to understand the political economy of fisheries and aquaculture was emphasised. In Bangladesh, for example, freshwater fisheries have been neglected and are subject to interference from other water interventions especially flood control and irrigation. An openly conducted National Water Plan exercise was very helpful in bringing out these conflicts, giving fisheries greater prominence and priority within government. One of the reasons why the importance of fisheries to different sectors of the community has not been emphasised has been the difficulty in measuring benefits in ways compatible with conventional cost-benefit analysis. With the recent attention on the environment there is better recognition of the importance of exploiting the diverse opportunities in aquatic systems and stronger economic arguments for fisheries investments. The question of drawing drinking-water supplies from water used for aquaculture depends on the intensity of use; highly intensive aquaculture will give poor quality water. In the case of fish cultured in wastewater, information about how toxins and pathogens move through fish and therefore the safety risks to consumers is available. The highly intensive form of food production in which fish ponds, vegetable production and pig keeping are integrated was discussed. Integrated fish farming can typically produce 2-8 tons/ha/year of fish provided it fits in to the physical and organisational aspects of the farming system. However, there are difficulties of unutilised waste matter accumulating in the pond.