3. AQUATIC TECHNOLOGY AND EXTENSION MESSAGE

Characteristics of a technology determine whether farmers perceive the technology as being easy to adopt. What matters for a farmer are the changes that are required for the adoption of the new technology. The perception of the farmers regarding the technological characteristics of the aquaculture system are of course influenced by the extension message. Therefore, this message plays a crucial role in the adoption process. The technology the Programme used was introduced in such a way that its characteristics were not perceived as requiring a very drastic change in the normal farming activities.

The following characteristics were important:

• ability of aquaculture to achieve farmers' goals better or at a lower cost than could be achieved previously.

• adoptable in such a way that it fitted into the normal farming practices.

• technology did it not have to be supported by other changes, nor did it require a whole set of new technologies.

• possibility to adopt aquaculture on a small-scale and expand later.

• possibility to practically see aquaculture at other farms.

Certain innovations are rigid and can only be adopted in one way. Aquaculture technologies can be adopted and adapted in many ways. Fish can be grown under many different conditions, using different species, intensities of management, scales of operation, etc. To enable small-scale farmers to adopt aquaculture it was therefore possible to adapt the technology in such a way that its characteristics were not perceived as too alien, complicated or expensive to adopt.

To facilitate the adoption of aquaculture by small-scale farmers, the technology had to comply with:

• the objectives of the farmers;

• their available means of production;

• the sites available for pond construction;

• their existing farming system;

• the prospective distribution channel for fish;

• the availability of fish seed.

Based on data from secondary sources, surveys and open interviews with farmers, it became apparent that aquaculture would be an extensive or semi intensive activity, that would fit in the strategy of risk avoidance instead of profit maximisation, and that farmers saw fish farming as an additional farming activity and not a specialisation.

Using a participatory approach, discussions with farmers on how to implement fish farming were encouraged as much as possible. Farmers were assisted in making decisions on where and how to build a pond, and how to manage it afterwards. This interaction gave the extensionist an opportunity to learn about the farmers' situation and helped to create a mutual background of information. During farm visits an exchange of information took place which followed a standard procedure. Farmers gave information about the resources they were willing to divert to fish farming and their aims, upon which a fish farming system was worked out together with the extensionist. This led to the adoption of a technology that fitted the conditions of the farmers.

3.1 Adaptation

Farmers do not just have the option of adoption or rejection, adaptation of an innovation or partial adoption of some components is possible. Because of capital scarcity and risk considerations, farmers experiment with recommendations, often adopting them in stages. A farmer may be uncertain whether the information he receives about a recommended innovation accurately describes his own prospects. In order to learn more about the innovation, the farmer may choose to adopt a component rather than the package, even when the package is more profitable.

An extension message should therefore not be seen as a ready-made product without the possibility of adaptation. The value of a technique depends on local conditions and has to be adapted accordingly.

ALCOM tried to develop extension messages that left open the possibility for experimentation. The message informed farmers and assisted them in making decisions. Understanding the reasons behind messages was considered much more important than the details of the message itself. Farmers were then perfectly capable of adapting technologies to changing situations through experimentation.

3.2 Production system

Earthen ponds was the production system most suitable for small-scale farmers in southern Africa. Extension packages often dictated in detail how a pond should be constructed. The description included the size of the pond, the slope of the bottom, the slopes of the dike, the width of the crest etc. in exact measurements, assuming that it was easier to follow instructions than to understand principles. Extension staff also wanted to make sure that there was no chance of any criticism on the pond features by supervisors. Besides, it was easier for extension staff to deliver standard messages than explanations. This sometimes resulted in large ponds with oversized dikes with wide crests and very gentle slopes, based more on requirements for large reservoirs than for small-scale fish farming.

Taking into consideration that farmers only had limited access to the extension service, ALCOM presumed that understanding the principles was more important than constructing the perfect pond. The understanding why dikes had to be compacted, why dikes should had sloping sides was considered essential for farmers to be able to continue independently and also for them to be able to advise others. The size of a pond was explained in terms of possible output and inputs requirements, and not in absolute terms. Sites, timing of construction, available tools to be used and inputs, etc. were all selected in discussion with the farmers.

3.3 Management practices

Fish farming was a new activity for most farmers, and as long the advantages of it were not proven, few farmers were prepared to spend money on it. Inputs for management therefore had to be found locally. Inventories were made together with the farmers of what inputs were available locally, it was explained how these inputs could be used for fish farming and this use was compared their present use. In most areas the normal inputs for fish farming, like manure and brans, were in short supply and in high demand. Vegetal material on the other hand was readily available.

The colour of the water was used as an indicator to determine the quantity of fertilisers, the consumption of feed as an indicator to increase or decrease the feeding level.

3.4 Harvesting strategies

The traditional harvesting strategy of draining a pond at the end of the production cycle is not appropriate when farmers' main objective of fish farming is to have access to a secure supply of fish for home consumption at any time. These farmers opted for the possibility of catching a few fish as and when needed. Traditional aquaculture extension projects for small-scale farmers never recognised this harvesting strategy. On the contrary, it was often vigorously discouraged by extension staff. This strategy of intermittent harvesting has been given little attention in research and extension programmes. Research on this issue has focused entirely on the bio-technical feasibility under intensive fish farming systems.

Although often not recognised, intermittent harvesting is practised by most small-scale fish farmers in southern Africa, a strategy that has the following advantages:

• provides an accessible supply of protein, which allows for a more varied diet;

• provides the possibility of small amounts of cash when needed;

• no water is lost during harvesting;

• no need for restocking after harvest;

• easy and quick method of obtaining a few fish.

The main disadvantage is of bio-technical character. In a classical fish culture system in stagnant water, ponds are stocked with juvenile fish of the same size and age. The total biomass is harvested after a certain culture period. The stock is harvested when the rate of increase of the biomass starts to decline. This point depends on the biomass in the pond and its productivity. the productivity can be enhanced through feeding and fertilising. The production of the fish pond can be maximised by optimising the stocking rate, the productivity and the culture period.

Another strategy to maximise fish production is to maintain the biomass of fish close to its optimum (level that realises the greatest increase) during the whole culture period. This requires a constant and optimal cropping of the system. In theory this system can achieve much higher production than the first strategy.

The system becomes more complex when the cultured fish reproduce during the production cycle, as is the case with tilapia culture of mixed sexes. It is difficult to estimate the success of reproduction and recruitment in a pond, and thus it becomes impossible to know the total biomass. Under these circumstances it is impossible to maximise production.

This means that technically speaking, intermittent harvesting is a possibility for maximising production. Small-scale farmers practise fish farming in earthen ponds, with extensive management and use a prolific breeding species. Under these circumstances it is impossible to maximise fish production through intermittent harvesting. This has always been an important reason for extension services to discourage this harvesting strategy.

Other reasons to promote batch harvest have been the interest from extension services to get production data from fish ponds, something virtually impossible when farmers harvest small amounts at irregular intervals, and the idea that the farmers need good quality fingerlings for restocking their ponds. Extension services were also often of the opinion that since a batch harvest provides a relatively large amount of income at one time to the farmer, this would increase the farmer's motivation to adopt and continue fish farming more than small amounts of fish at regular intervals.

Maximisation of production was never the only objective for farmers to adopt fish farming. From various studies it became clear that many farmers saw a fish pond as an easy accessible source of fish. Other animals were seldom slaughtered or sold, and then only during special occasions. Most farmers slaughtered even a chicken only for special occasions, or for visits of honored guests, but not because the family had little to eat. One of the great advantages of fish over other farm animals was that fish could be harvested and eaten when the family needed it. This made many farmers interested in fish farming.

Extension message:
The message to the farmers was that fish could be harvested in small amounts at intervals. It was also explained that since the fish reproduced, the pond would become over-stocked with small fish, and that it was not feasible to expect all these small fish to grow to a large size. In order to achieve a good production and large sized fish, it would be necessary for the farmer to drain their pond at least once a year, and restock it with the right number of fingerlings.

3.5 Stocking

A main concern for farmers was the issue of obtaining fingerings for initial stocking and for restocking of ponds. In most areas the farmers wanted to be independent from outside sources. The idea that fish would reproduce in their own ponds and that they could use these for restocking was for many farmers an important aspect that made them decide to adopt fish farming.

One reason for extension services to insist on batch harvests was to ensure that the farmers would stock their ponds with fingerlings of good quality. Most extension workers were of the opinion that good fingerlings could only be produced at government fish farms.

Use of own fingerlings is often discouraged for reasons of risk of inbreeding or improper selection. Inbreeding depression occurs when too few brood fish are used for the production of future generations. A minimum of 25 males and 25 females as breeding stock for each generation is normally considered enough to minimise inbreeding depression. Since most small-scale fish farmers stock their ponds with at least 100 fish, the risk of inbreeding should not be exaggerated.

The problem of selective harvesting, especially when intermittent harvesting is practised is more serious. When intermittent harvesting is targeting the larger fish in the pond, the farmer selects his stock towards slower growth and earlier maturation. Although the heritability for weight at a certain age fluctuates greatly between different studies, and realised heritability appears to be low for both Oreochromis niloticus and Oreochromis mossambicus, the selection towards slower growing fish is a genuine risk in ponds that are never drained and only partially harvested.

On the other hand, the quality of fingerlings at government fish farms should not be overestimated. Few farms maintain a selection programme or even follow basic selection procedures. In Zambia, the government fish farms often required new stocks from the wild to replace and improve the stocks at the farms. For small-scale farmers the issue of being independent from outside sources, and the accessibility and timely availability of seed was far more important than the assumed improved quality of the fingerlings from government stations.

In most areas the government fish farms offered fingerlings to farmers at heavily subsidised prices, and often did not insists on payments. This policy discouraged private fingerling production. At the same time the production at the government farms was low or extension services were unable to provide these fingerlings to the farmers in a timely manner, due to the scattered location of fish farmers and the long distances involved. As a consequence, many ponds remained unstocked, while fish farmers in the area had fingerlings available.

3.6 Species selection

Fish species were selected according to a number of criteria. A species that was easy to reproduce by the farmers themselves greatly enhanced the farmers' independence. The species had to be located at the beginning of the food chain with a flexible diet and be able to use a wide variety of natural food and low quality feeds. It further had to be able to withstand strong fluctuating oxygen levels and periodic low oxygen levels.

Most fish species are widely accepted in southern Africa, only the African catfish and the eels are met by some consumer resistance. Fish species that are known to the farmers, and extension and research staff, and that are available, are much easier to use in extension than species that have to be introduced. Besides, introduced species can pose a threat to the local environment and biodiversity.

In aquaculture a certain species is often preferred because it performs better and achieves higher yields than other species. However, under the management strategies adopted by small-scale farmers the production was in first instance determined by the level of management, and less by the ultimate performance potential of the species.

In the areas where ALCOM worked with aquaculture extension activities, the local cichlids were used. These were:

• in Zambia: Oreochromis andersonii, Oreochromis macrochir, and Tilapia rendalli;

• in Tanzania: Oreochromis niloticus; and

• in Mozambique: O. niloticus and T. rendalli.

O. niloticus was not an indigenous species in the project areas in Tanzania and Mozambique, but was used by the fisheries departments for aquaculture at these sites. In Mozambique the government department used Common carp, Cyprinus carpio as an additional species for stocking ponds of smallholder farmers.

ALCOM, upon request of the government of Mozambique, assisted with the introduction of Grass carp, Ctenopharyngodon idella, Silver carp, Hypophtalmichthys molitrix, and Bighead carp, Hypophthalmichthys nobilis. Although originally meant for stocking of reservoirs and ponds, these species were never used for aquaculture extension (see Chapter 7.3).

In all areas where the Programme worked, the price per kilogram of fish reduced after individual fish weights reached 70 grams. Production of small fish could thus be considered more lucrative than the production of larger fish. The perception of many farmers however was that all fish, including all recruits, should reach 150 or 200 grams. When fish stayed small, farmers started to complain about the species and the quality of their fish, while stunting was the main reason for the smaller sizes.