Based on a report by Henk van der Mheen Aquaculturist, ALCOM1

With Special Reference to Zambia and Tanzania




Irrigation systems differ in water source (river, well, spring, borehole), capture method (open intake, gate valve, pump), delivery (open canal, pipes, buckets, pumps), distribution system (flood, furrow, sprinkler, drip), and the allocation scheduling of the water. These differences influence the flexibility of the systems and the freedom of the individual farmer to make management decisions.

Management complexity increases and flexibility decreases with the size of the scheme, as more stringent management is applied. Large-scale systems face greater management complexity in delivering water to meet crop requirements over a large area. This is compounded if farmers choose diverse cropping patterns with staggered planting dates. To avoid this, relatively uniform cropping patterns are often imposed in larger schemes. Crops are grown in blocks and selection of these crops and planting time are organized for all the farmers with plots within that block in such a way that optimal use can be made of the available water. The specific blocks are only supplied with water when those crops require irrigation. This detailed organization of the water supply is especially essential for large schemes with a complex canal system, and where the water supply is being controlled either by pumping or release from a reservoir. The imposition of uniform cropping patterns reduces flexibility at the farm/farmer level.

Farmers operate farming systems which incorporate a variety of activities. These systems are flexible and change according to markets,

available resources, experiences, etc. Flexibility to react to uncertainty and changing conditions may be so highly valued that farmers avoid obligations and organizational bonds in order to keep individual room to maneuver. Irrigation schemes where farmers are left with little freedom to maneuver are therefore not popular among farmers. In addition, large formal irrigation schemes are often hampered by operational and maintenance problems, mainly because the schemes were designed from a technical perspective without the involvement of the users. The vast majority of the formal irrigation schemes in Zambia is actually not in operation (IFAD, 1993).

It is now increasingly recognized that farmers should be given considerable freedom of crop choice, even under formal irrigation schemes. This is especially relevant during economic changes such as changing national or world market prices, development of market structures, abolishment of subsidies, etc.

This implies that the cropping pattern is no longer predictable and the design of the irrigation system can no longer be adapted to a particular crop or set of crops but must allow diversification. This is difficult to accommodate in large schemes with wide canals, where, for the purpose of optimal water use, along one feeder canal are expected to irrigate at the same set times.

1van der Mheen, H. 1999. Adoption of integrated aquaculture and irrigation. A study conducted in Zambia and Tanzania. TCP/RAF/7825, ALCOM Working Paper No. 23. Rome, FAO. pp.21.


Flexibility of time allocation is important for farmers, especially for women. Gravity flow canal irrigation requires the presence of farmers at specific times. For optimal use of farmers' labor, water should be available when needed, by means of effective and problem-free delivery systems. Farmers value assured and predictable water supplies; unpredictable supplies are reflected in low input, low production strategies, and low application of fertilizer (Chambers, 1988). Small-scale systems, where farmers have much more direct control over the water supply, are characterized by intensive systems of diverse cropping patterns and planting dates. Diversification of crops is a way to optimize the use of available resources, cope with fluctuating prices and adverse weather conditions, and provide a wider range of products for home consumption.

The majority of farm households maintain a significant, if varying, degree of autonomy from the market as characterized by the share of farm output which is consumed within the household rather than sold. Risk avoidance, rather than profit maximization, dominates household decision-making.


Integration of aquaculture and small-scale irrigation aims at increasing overall production through improved efficiency of water and land use, use of labor or other inputs, and the diversification of farm production for sale and home consumption. Integration can also improve the efficiency of the dissemination of knowledge and information.

Water Use

The main concern of farmers in irrigation is the reliability of the water source, and the predictability of its availability. In many cases, the construction of small ponds can improve the reliability of the water source. Where groundwater is the source, the construction of ponds to collect this water increases the supply directly available to the farmer. In many places in Zambia and Tanzania, farmers have constructed fish ponds supplied with groundwater that also function as water storage reservoirs. The farmer has access to a large quantity of water when he needs it and the pond refills from the water table when the farmer does not irrigate.

Water storage close to the fields facilitates the planning of application and also allows control over the water flow by farmers themselves. Water flow must not be too large, too small or too variable in order to insure ease and efficiency of capture, control and application by farmers. Farmers in Mozambique constructed storage ponds next to their fields in order to cope with water allocation schedules and water shortages during the day. These ponds were at the same time used for fish production. At an irrigation scheme in northern Senegal, a distribution system with reservoirs was developed to provide irrigators with individual freedom and minimize organizational processes of water allocation (Ubels and Horst, 1993). This was especially appreciated by women who, due to other activities, were unable to be present at their plots at fixed times.

Chambers (1988) discussed flexibility of time allocation as an important option to deal with labor shortages. The construction of storage reservoirs near the fields greatly enhances this flexibility and improves irrigation. He presented an example of small individual storage reservoirs which are filled during the night and are used by day for irrigation.

Collection ponds and small storage reservoirs can be used for the production of fish if they have a dead volume. In cases where the feeder canal provides enough head, these fish ponds can be constructed to permit irrigation by gravity from the pond, thus integrating water use for fish farming and irrigation.


Fish farming can also be integrated with irrigation to further diversify farm crops for home consumption and for sale.
Small-scale aquaculture provides fish for home consumption. Fish are not only consumed when the whole pond is harvested but also throughout the year through intermittent harvesting. Fish have an advantage over other livestock in that they can be harvested and consumed when desired, while animals such as chickens, goats, and cattle are usually only consumed on special occasions. Therefore, fish from small-scale aquaculture contribute to improved food security, especially transitory food insecurity.


Apart from providing the household with an additional crop for consumption, fish can also be sold locally fairly easily, and for relatively high prices. This may be especially important for more remote locations where marketing of vegetables is difficult and expensive. An additional advantage with the marketing of fish is that fish does not have to be marketed during a certain season, or at a certain point in time. Fish does not ripen in the sense that it has to be harvested before it decays. Although a pond may have its highest production after a certain production cycle, fish can be left in a pond until the prices are high enough or until a farmer needs the money. Contrary to most horticulture crops, a fish farmer is never forced to sell his crop, unless his pond dries up. Normally, fish can be marketed locally; production levels from small-scale aquaculture are not high enough to force farmers to market fish further afield.

Development Approach and Information Dissemination

Many development programmes lose the confidence of local communities because they come with rigidly defined official agendas and technical packages with specific boundaries of intervention.

In the context of a participatory intervention strategy, a programme can be more successful if it has a mechanism to take into consideration the total experiences and aspirations of target groups. The focus should be on the entire family livelihood system rather than on a distinct subsystem or a solution to only a very specific part of the problem. Fish farming, as well as other technologies, should be seen as one aspect of rural development, not an isolated technology. It must be integrated into a holistic approach to rural development. This requires a wider use of interdisciplinary approaches to rural development.

Fish farming and irrigation have many commonalties. They require similar resources and inputs and consequently compete for resources. But they can also complement one another and thus increase the efficiency of resource use. Yet, extension services for irrigation and aquaculture are often separate, and as a consequence the information and advice for one may conflict or interfere with the other. The integration of extension systems would clearly encourage an integrated approach to resource use.


Many considerations influence the decision to adopt or reject an innovation. The prevailing conditions of a farming household, including farming practices and perceived needs, influence adoption. The resources required have to be available and information on the innovation must be accessible to the farmer in useful form. The latter depends on the communication channels used and on the extent of the change agents' efforts to promote the innovation. The characteristics of the innovation itself will also influence the decision for its adoption or rejection. It is very difficult to predict the adoption rate of an innovation. However, evaluation of its characteristics and its potential acceptability during the prediffusion stage could, to a certain extent, help assess the potential for successful adoption.

Rogers (1983) suggested five characteristics of an innovation that affect the rate at which it is diffused and adopted, these are: relative advantage, compatibility, complexity, triability and observability.

The relative advantage is the degree to which an innovation is perceived as being better than the idea it supersedes. The advantages of fish farming have been discussed above, fish farming produces an extra crop, and fish ponds can be integrated with the water distribution system and as such improve the water delivery for the farmer.

Compatibility is the degree to which an innovation is perceived as consistent with the existing values, past experiences, and needs of potential adopters. Fish farming is an agriculture activity, and farmers who are used to irrigation and management of water resources do not perceive fish farming as an alien activity. A local lack of animal protein or a need for crop diversification provides an entry point for aquaculture and increases the compatibility of the innovation. Where farmers feel that water allocation within irrigation schemes is a serious problem, ponds can reduce this problem. This increases the compatibility of fish farming. However, the rationale for having ponds, i.e. the shortage or irregular supply of water, decreases its compatibility. Previous experience with water shortage makes adoption of fish farming less likely.


Complexity is the degree to which an innovation is perceived as relatively difficult to understand and use. In general, more complex, or less well understood innovations are more difficult to adopt. Fish farming in small ponds can be introduced in a way that it is easy to understand, but the management of the fish pond is a more complex matter, and the adoption of proper management techniques requires much more external assistance. The adoption of fish ponds integrated in the water management system is more complex than the adoption of fish ponds that are operated independently from irrigated crops.

Triability is the degree to which an innovation may be experimented with on a limited basis. Adoption becomes much easier if farmers can try an innovation on a small scale. The triability of ponds integrated into the water distribution system and those operated independently could be considered the same, although the former would require a minimum size and depth to effectively improve the water distribution system.

Observability is the degree to which the results of an innovation are visible to others. The rate of adoption increases with visibility. Fish ponds are of course very visible; the construction process as well as the completed ponds are easy for others to observe.

The influence of the characteristics of the two types of fish ponds (integrated vs. independently operated) is summarized in Table1.

Integrated fish ponds have an extra advantage over independent ponds in that they can improve water distribution. However, they are less compatible with existing activities and experiences of the farmers, and are more complicated to understand. This deters adoption. Triability is also slightly more difficult with integrated ponds, while observability can be assumed to be equal, or slightly greater for integrated ponds. Benefits in the form of fish and income seem to be less evident than the benefit of an independent source of water.

Observed Adoption

Fish farming was adopted at 13 of 17 sites in Zambia and Tanzania where farmers were practising irrigation.

Perceived needs of farmers were identified at the thirteen sites, and are presented in Table 2 together with the relative rate of adoption. One could argue that the relative advantage of aquaculture is greatest where it meets farmers' needs for protein, diversification and improved water use, and lowest where only one of these needs is perceived. This would explain the low rate of adoption in Shantumbo and Ngwerere. At these sites farmers had access to water, and to nearby markets that offered good prices for vegetables. In addition, the physical conditions at the sites were not very favorable for the construction of ponds (Table 2).

13-tab1.jpg (40570 byte)


14-tab1.jpg (51745 byte)


At Mbembeshi there was an additional need to improve water allocation, but the area offered few locations for pond construction due to poor soils. Conditions were good at Malolo and fish farming could meet the needs for diversification and protein supply. However, the adoption of fish farming was hampered by the location of the sites and the negative influence of some local officials on the introduction of fish farming (Wetengere et al. 1998).

At Chipapa and Chipungwe, all perceived needs could be met by adoption of pond culture, but adoption was slow nevertheless. Farmers did not construct fish ponds integrated into the water allocation system due to the unreliability of the water supply. Farmers who did start fish farming at these sites constructed ponds with water sources independent from the irrigation water. Ponds were constructed immediately down stream from a dam, and were filled with seepage water from the dam. This option was less complex and did not interfere with irrigated plots. The ponds increased the amount of irrigation water in Chipungwe, because farmers used the water from the ponds to irrigate crops using buckets. However, only a few ponds could be built in this manner.

In Mgeta, the rate of adoption of fish farming was moderate but steady, and the practice continued to spread to nearby villages. Land and water are both intensively used and the market for vegetables and crops is relatively well developed. Fish ponds had to be located near

the homestead for fear of theft and thus competed with other high value crops that were normally planted there. Since land was a limiting factor for increasing production, not all farmers were able to adopt fish farming. Fish ponds were rarely used for storage of irrigation water; water was allocated to the ponds when it was not in high demand for irrigation.

Sites with sufficient water and land and with a clearly felt need for protein and diversification showed fast adoption. Farmers at Tangeni, Mfumbwe, Kibwaya, Chalata, Irumi and Musofu had access to land and water; water was available in sufficient quantity and there was no competition from other activities or other farmers. Here fish farming met a felt need and could easily be adopted without interfering with other farming activities. Therefore, adoption was fast. Adoption was especially fast in Musofu where most farmers had access to land with water, but where markets were difficult to reach and prices for vegetables barely met the costs for production and transport. There was also a severe shortage of animal protein.

The topography of the area proved to be of much less importance than anticipated. In Tangeni, ponds were constructed on 50 percent slopes (1:1), which are normally considered unsuitable, and the high construction labor was not a major concern. In some cases, diversion canals were constructed over long distances in order to bring water to the ponds.



The above observations were gathered in the course of a very brief field survey of selected sites in Tanzania and Zambia to examine the adoption of fish farming and the appropriateness of the integration of fish farming and irrigation. The limited duration of the survey only allowed a brief description of what happened at the sites. For a more scientific in-depth study, a larger sample of sites is required encompassing a wider range of irrigation schemes, including larger schemes with clear systems for water allocation and scheduling, as well as schemes situated in drier areas of southern Africa.

Fully integrated techniques are always more difficult to adopt since they interfere with established activities and therefore pose a risk. Step by step introduction towards full integration, if technically possible, may increase the adoption rate. Fully integrated techniques most likely will require more technical assistance and training to facilitate adoption.

At many of the surveyed sites, water was not a restricting factor and there was no special incentive for direct integration of water use. The farmers did not pay for the use of water and, therefore, inefficient use is not directly reflected in farmers' incomes. This situation may change as water resources come under greater pressure

from users. Zimbabwe has now officially adopted a system whereby charges will be levied on water use. Although details of how the system will be implemented are not available, it is a clear sign that water will increasingly be considered a commercial commodity in southern Africa. With this change, efficient integrated use of water will become a necessity, including the integration of aquaculture and irrigation.


IFAD. 1993. The Agricultural Sector in Zambia. IFAD Identification Report. Working Paper?, Rome, IFAD. 22 pp.

Chambers, R. 1988. Managing Canal Irrigation Institute of Development Studies, University of Sussex, UK. 278 pp.

Ubels J. and Horst, L. 1993. Irrigation Design in Africa: Towards an Interactive Method. Wageningen Agricultural University, Wageningen, Netherlands and Technical Centre for Rural and Agricultural Co-operation, Ede, Netherlands.

Rogers, E. M. 1983. Diffusion of Innovations. The Free Press. New York. 453 pp.

Wetengere, K., Osewe, K. & van Herwaarden, H. 1998. Development of semi-intensive fish farming in Morogoro Region, Tanzania. FAO ALCOM Working Paper 22. Harare. 54 pp.

15-1.jpg (96006 byte)