Previous Page Table of Contents Next Page

Chapter 5 - Agro-forestry, a new fashion of old tradition?¹

¹ Adapted by F. Pétry, from a draft agro-forestry working paper, prepared by David French (1985), with the permission of the author and of IFAD, Africa Group, for which the paper was prepared.

I. A new interest
II. Agro-forestry; adding trees to farming systems?
III. Agro-forestry techniques
IV. Financial and economic implications
V. Preparation and implementation issues
VI. Conclusions


Agro-forestry has a long history, from man's earliest attempts at agriculture to the present time, even in the most industrialised countries. It is one of the most integrated systems for tackling energy and rural development needs together. It will therefore be described below relatively at length.

Typical home garden in south-east Asia - 1. bananas 2. mangos 3. coconuts 4. living fence 5. shade trees 6. vegetable garden 7. poultry

Agro-forestry ranges from traditional shifting cultivation (temporal integration) to elaborate systems of fruit trees, vines and cereals in spatial complementarity or poplar-crops systems, in Italy, for example. Recent developments include the pinus radiata, and sheep integration in New Zealand and Chile.

Agro-forestry may be defined as the intimate integration of trees and crop and/or animals in the farming system or, more widely, as the integrated management and use of forest and agricultural resources. It combines woody perennials with annual crops and/or livestock, either simultaneously or sequentially on the same unit of lands. Its purpose is usually to satisfy local consumption needs and generate additional income, while sustaining agricultural productivity and maintaining ecological stability.

The following categories of agro-forestry may be considered here.

i) Shelter. Systems in which trees provide shelter for crops, pastures or livestock. The main purpose is agriculture; trees fulfil a service role.

ii) Forest grazing. The grazing of livestock in forests or woodlands not originally planted with grazing in mind. The main purpose is forestry, with agriculture playing a supplementary role.

iii) Silvopastoralism. The deliberate integration of trees and livestock.

iv) Shifting cultivation - or periodic agriculture within a natural forest environment.

v) Agrosilviculture. Combinations of trees and crops.

vi) Agrosilvopastoralism. The integration of trees, crops and livestock - often in a temporal sequence as well as in a spatial pattern.

I. A new interest

Until relatively recently, the problems of agricultural production and deforestation were viewed separately. Thus, inadequate food production was seen to result from shorter fallows, inadequate use of improved technical packages, lack of incentives due to artificially depressed food prices, and so on. Deforestation, on the other hand, was thought to result from excessive cutting of wood for fuel. A growing consensus, however, now considers food and tree problems to be intimately related. For example, it is understood that most trees are not cut to provide fuel, but to clear land for planting crops. More important is the recognition that deforestation is an important reason for declining food output. It is hardly coincidence that many famine areas are those where trees and shrubs have largely disappeared.

In Africa, for example, production over the long run depends almost as much on trees as on seed, fertilizer, and other conventional inputs. Traditionally, most African farmers have cleared land, used it until fertility began to decline, and then moved on. Trees were important throughout this cycle. While the land was being used to produce food, for example, surrounding areas of tree cover served to prevent wind and soil erosion.

Often, some trees were left scattered across the farm itself. Acacia albida, for example, was known to provide a variety of benefits. During the hot season, animals attracted to the tree's shade would leave manure behind, increasing soil fertility. The tree's roots would fix nitrogen, and its fallen leaves would add organic matter to the soil. Competition with crops was limited since Acacia albida sheds its leaves during the growing season and draws its moisture from levels which the roots of food crops do not reach. Trees such as this were valued highly by farmers as "good neighbours" for crops and livestock. As the soil became exhausted and the farmers moved on, a variety of trees and shrubs would quickly reestablish itself. During the land's fallow period, the roots of this woody vegetation would return leached nutrients to the topsoil from layers of earth deep below the surface. Disintegrating leaves and twigs would restore the crumbly soil texture necessary for proper infiltration of rainwater. Eventually, the soil's productivity would be restored, and farmers would return Co grow crops there once again.

This cycle was interrupted when population pressures forced farmers to settle and to use land much more intensively. Fallows were shortened or eliminated, preventing the natural reafforestation that had been such an essential part of traditional agriculture. Trees that would formerly have been left in and around gardens were cut to provide firewood or to make way for more crops.

There is an added dimension to the problem in areas like the Sahel, where livestock occupies a central place in the farming system. With too many people come too many animals. These start by eating away the herbaceous ground cover, eliminating the best species first. The accessible parts of the trees are finished off next, after which herders must cut branches or pollard trees to provide feed. Eventually, both the ground cover and the trees are stripped away completely.

The results of both overcropping and overgrazing can be devastating. With too little humus, the soil loses its capacity to absorb water. Rain therefore washes across the surface of the land, carrying away precious topsoil. Erosion is made worse by the absence of trees that could break the flow of wind and water. The topsoil Chat remains is increasingly devoid of nutrients. At best, the land may be kept temporarily "alive" through the massive use of fertilizers. At worst, the land may die - and so may the people who work it.

Once ecosystems begin to collapse in this way, they are especially vulnerable to drought. Rather than being simply a hardship, bad rains can become a calamity. Many people now argue that this is what has happened in areas such as Ethiopia and the Sahel: people who were already using land in unsustainable ways were simply pushed over the edge by periods of inadequate rain ¹. There might have been less human suffering and less permanent damage to the land if traditional farming systems - including traditional trees and shrubs - had still been in place.

¹ See, for example, World Bank: "Desertification in the Sahelian and Sudanian Zones in West Africa" (Report No. 5210); October 9, 1984; p.5.

The above is an oversimplified picture of African farming systems. It remains true, however, that most traditional land use systems involve a strong interdependence among people, crops and/or livestock, and trees. This interdependence will long prevail. It may be in theory Chat African and Asian lands could be stabilized and made productive without trees through careful use of modern inputs and techniques. Nonetheless, most farmers and herders will indefinitely remain poised somewhere between the "traditional " and the "modern". To survive, they will need to retain or recreate farming systems in which trees occupy a central position.

II. Agro-forestry; adding trees to farming systems?

It is relatively new to think of farming systems as including trees, or to plant trees with farming systems in mind. As recently as 1982, for example, a major study of farming systems made no mention of trees except as an "off-farm" source of fuel ¹. At the same time, most tree planting projects were assuming that farmers would grow woodlots on marginal land far from their crops.

¹ W.W. Shaner, P.F. Philipp, & W.R. Schmehl, Farming Systems Research and Development; Guidelines for Developing Countries; Boulder, Colorado: Westview Press. 1982; see, e.g. p. 68.

Ideas are rapidly changing. A more recent study of African farming systems gives considerable attention to trees, especially in long sections devoted to soil conservation and the improvement of soil fertility 2. Simultaneously, social foresters have begun to give urgent attention to activities such as Niger's Maggia Valley Project, which planted windbreaks to protect farm La id from wind erosion. The convergence is not yet perfect: there are still foresters who seldom think of farms, and agronomists who seldom think of trees. Nonetheless, there is increasing agreement on the need to grow trees within farming systems, or even to conceive of a wider range of plants and animal mix in order to create a sustainable, productive farming system: for example, to a maize monoculture might be added Leucaena leucocephala, bananas, sisal, groundnuts, small livestock, pigeon peas, cowpeas, mangos, yams, Cassia siamea, or Acacia albida. Properly arranged, these would serve to add nitrogen and humus to the soil, prevent erosion, fence off the garden, and provide food, fruit, poles and firewood for consumption and for sale. Since the soil would be automatically replenished with nutrients and organic material, such a system could continue indefinitely.

2 William Jones and Roberto Egli, Farming Systems in Africa: The Great Lakes Highlands of Zaire, Rwanda and Burundi (Technical Paper No. 27); Washington, D.C.: The World Bank, 1984.

Conceived in this way, agro-forestry is a land use strategy that would: (a) increase the output of food and other products on a given unit of land; (b) ensure that increases in output were stable and sustainable; and (c) raise farmers' cash incomes. It is noteworthy that these are exactly the conditions necessary for greater food security in developing countries.

III. Agro-forestry techniques

Many of today's agro-forestry techniques will be adaptations of traditional practices. However, non-traditional problems are raised by the decline of shifting cultivation, growing pressure on land, and rapid deforestation. For example, there is new need for fast-growing, nitrogen-fixing tree species suitable for different ecological zones. In addition, ways must be found to compress larger numbers of trees, animals, and food crops on intensively managed farming systems. Intensive research is still required in this field. Some of the most promising agro-forestry techniques are outlined below.

Scattered Farm Trees

One agro-forestry intervention is simply to increase the number of trees or shrubs scattered among crops or pastures and along farm boundaries. These trees may have value greater than any crops they displace (as can be true of fruit trees). Alternatively, the trees may actually enhance the productivity of crops and pasturage by replenishing soils and helping to reduce erosion. For example, millet and sorghum may increase their yields by 50-100% when planted directly under Acacia albida. Often, trees can be planted that will serve multiple purposes; Leucaena leucocephala, for example, can be used for poles, firewood, fodder and soil enrichment.

Scattering trees across the farm will not by itself meet all domestic needs for tree products or ensure sustainable agriculture. However, this kind of agro-forestry action is most likely to make immediate sense to farmers. Benefits are diverse and in line with felt needs. New species can be introduced cautiously and cried alongside proven ones. Risks are minimized. In a number of countries, this approach is now the principal way in which agro-forestry ideas are being introduced.

Improved fallows

Another technique that can be introduced without making radical changes in the farming systems is the improvement of fallows. This is required by the fact that fallow periods in many areas have been substantially shortened. As a result, only shallow-rooted grasses of little value to the soil establish themselves naturally before crops must again be planted. The usefulness of short fallows can be greatly improved by planting green manure crops such as Desmodium and Tephrosia spp., Lupinus mutabilis, Dolichos, Indogofera, etc. In addition to improving the soil, such crops may yield livestock feed or food for human consumption.

It is worth noting that fallows can be improved without the use of "trees". Nonetheless, this kind of action falls directly within the scope of agro-forestry as defined here.

Alley cropping

A much more complex agro-forestry technique is alley cropping, as developed by the International Institute for Tropical Agriculture (IITA). This involves the growing of crops between closely-spaced lines of trees or shrubs, usually legumes. To prevent competition with crops for light and water, the trees are regularly cut back and are kept pruned during the growing season. Larger cuttings are available for firewood. Leaves and twigs are spread as mulch among the crops. This mulch replenishes the soil with humus and nutrients, as well as providing a barrier against erosion. Among the tree species found to work best in IITA's trial plots are Leucaena Leucocephala and Gliricidia sepium.

The details of alley cropping systems will vary. For example, herbicides may be applied (as IITA does), or weeding may be done by hand. Mulch may be spread before the crops are planted (as at IITA) or afterwards (as farmers have tried with mulching in northern Cameroon). Trees may be cut at ground level (as at IITA), or just the branches may be lopped off (as farmers do on Mount Kenya, leaving a bare pole that does not shade the crops). The appropriate combination of techniques in a given area will depend en such factors as the availability of labour for mulching and weeding, the presence of pests Chat might be spread by mulching, specific crops to be grown, and the farmers' willingness to make large changes in farming practice.


Mulch may also be used outside of alley cropping systems to protect or rehabilitate land. For example, the application of branch mulch helps restore nutrients to the soil, reduces erosion by breaking the impact of falling raindrops, and encourages water infiltration by slowing run-off and making the soil more porous.

Buffer strips

Strips of vegetation planted along the contour lines of slops can serve as "buffers" against water erosion. Perennial grasses have long been used for this purpose in many countries.

In more complicated systems, buffer strips may include grasses, fodder legumes, trees and shrubs. In such cases, the strip can yield mulch, forage, fruit and fuelwood, as well as controlling erosion.


On gentle slopes, water erosion may be prevented through buffer strips and mulching. Where slopes are more severe, however, construction of terraces may also be necessary. This job will be made simpler if buffer strips are included in the form of dense vegetation planted along the risers and edges of each terrace. If this is done, a much smaller earth structure will be required to control run-off than would the case with conventional, "naked" terraces.


Planting lines of trees to break the wind sweeping across open land can prevent erosion and slow the evaporation of moisture from the soil. Species can be chosen that will also yield fodder, poles or firewood.

Individual farmers may act to protect land by planting trees on the windward boundaries of their farms. More ambitious schemes seek to control wind erosion over much larger areas. For example, a CARE project in Niger's Maggia Valley established 250 km of windbreaks in seven years. These consisted of staggered, double rows of trees, primarily Azidirachta indica and Acacia scorpioides, planted across the valley at 100-metre intervals. Roughly 2,500 hectares of farmland were protected in this way. Initial indications are that millet production has been increased by 15% above yields on comparable, unprotected land.

In contrast to most agro-forestry techniques, large-scale windbreak schemes necessarily involve whole communities. This raises delicate issues in terms of sacrifices (whose land will be pre-empted for planting the trees?) and benefits (whose land will be best protected from the wind, and who will get such by-products as fodder and firewood?). Social analysis will obviously play an important role in this kind of project.

Live fencing

Many of the trees and crops in an agro-forestry system are vulnerable Co being eaten or trampled by livestock. Farmers will try to keep the animals away from the plants. However, adequate supervision of the animals may be Coo time-consuming, and wire or metal fencing too expensive. To solve this problem, farmers can plant "live fences" in the form of tightly spaced lines of species such as Prosopis juliflora. Acacia scorpioides, or Euphorbia spp.


Farm woodlots have often been promoted to provide firewood, either for domestic consumption or for sale. As far as domestic consumption is concerned, it will sometimes be appropriate for a family to plant such a woodlot on marginal land. However, recent estimates suggest that household firewood needs may be harvested from as few as 50-100 trees. This is most likely to be true if the trees are scattered around a farm rather than packed into a woodlot, where yields per tree are lower. Any serious agro-forestry scheme would include several times that number of trees as part of windbreaks, buffer strips, or scattered farm plantings. Under these conditions, there would be little need for a separate woodlot to produce fuel.

The growing of firewood for sale will usually be less attractive. Almost everywhere in developing countries, farm-gate prices of firewood are extremely low. Under these conditions, any attempt at firewood farming will lose money, either in absolute terms or relative to alternative uses of the same land. Where woodlots are established, these are most likely to be for production of high-value outputs such as building poles or timber, although markets for such commodities are limited.

Integrated techniques

In practice, several of the above techniques may be combined to improve existing farming systems. A number of such integrated techniques are proposed by D. French (1985, op. cit.).

IV. Financial and economic implications

A. Benefits and costs to the farmer
B. The government's perspective

The sections above consider, in technical terms, some of the interactions to be expected among trees, crops, land, and animals within agro-forestry systems. However, farmers and governments will wish to know more about financial and economic implications of these systems before moving to introduce them. Some preliminary information is therefore provided here on benefits and costs of agro-forestry.

These data should be treated with care, since little research has been done on the economics of tree planting or agro-forestry. The limited information available varies greatly in methodology and reliability. Nonetheless, the indications are consistent chat agro-forestry can have substantial returns at a relatively low cost. To confirm that this is true, new agro-forestry projects must include components to generate and evaluate economic data.

A. Benefits and costs to the farmer

An agro-forestry innovation may have reverberations throughout the farming system. Possible impacts of special interest to farmers will include changes in crop yields, curtailment of productivity losses now resulting from land degradation, and subsistence or cash outputs of the agro-forestry plants themselves. Against any such benefits must be set the value of land, labour and cash that would be required to establish and maintain an agro-forestry system.

Increasing crop yield

In the humid tropics, it has been estimated that Leucaena leucocephala can annually fix 500 kg of nitrogen in a hectare of land.¹ Senegal and Niger, tests have indicated that nitrogen and organic matter under Acacia albida trees are at least double their levels on open land. 2

¹ FAO Forestry Dept., "The Role of Forestry in Food Security"(Document for Committee on Food Security, 10-17 April 1935) p.11.

² Peter Felker, "State of the Art: Acacia Albida as a Complementary Permanent Intercrop with Annual Crops"; Riverside: University of California, April 1978; pp. 10, 16.

Even without improving soil quality, windbreaks can reduce wind damage to crops and thereby raise yields, as in the Maggia Valley Project mentioned above. Best results are achieved by simultaneously taking steps to slow wind or water and to improve the soil.

It is often assumed that a major "cost" of agro-forestry is the reduction in output when land is taken out of crops to plant trees. However, if 57. of a field is pre-empted for a windbreak, and another 6% for scattered farm trees, only a 12% increase in yields is required on the remaining land in order for the total output of crops to remain the same. This increase is well within the possibilities suggested by data above. At least in terms of crops foregone, there may well be cases where agro-forestry has no cost at all.

Avoiding losses of productivity

It is implied above that agro-forestry would replace present systems that are lower-yielding but stable in output. Of course, present systems are often not stable at all; they are deteriorating. Studies in a number of African countries indicate that topsoil is disappearing at annual rates of up to 8-15 mm. on grazed land, and at much higher rates on cultivated land.¹ The topsoil that remains is steadily degrading in quality and the ability to sustain food production.

¹ EcoSystems Ltd., "Southeast Shinyanga Land Use Study" (Final Report, Volume 2); Nairobi, December 1982; p. 53.

Under these circumstances, potential crop yields under agro-forestry, should not be compared with present levels' of output. The proper comparison is with likely future levels of output, which may drop rapidly toward zero in the absence of agro-forestry. In many cases, the benefits of an agro-forestry system would therefore be very great, even if yields were only maintained at current levels. As already indicated, we may expect in practice to achieve more than this.

Subsistence benefits of agro-forestry plants

Agro-forestry plants also produce direct benefits in the form of poles, firewood, fodder and other goods required for subsistence.

Other subsistence benefits of agro-forestry plants include food for people (fruit, nuts, leaves, roots) and for animals (e.g. Leucaena leaves and Prosopis pods). In addition, various species produce tools and utensils (e.g. the mortars made from Acacia albida), medicines, cooking oil, rope, etc. All of these are items to which families may attribute considerable (if unquantifiable) value.

Trees provide many different forms of food - a. fruit, b. nuts c. pods, d. leaves, e. bark/sap, f. fungi

Many products of the trees may also be sold for cash: building poles, fruit, charcoal. Acacia albida pods, gum arable, shea nut butter, etc. Given a proper harvesting regime, these benefits can be realized without damaging the trees' ability to protect and enrich the land.

Net benefits to the farmer; an example

In the absence of farm-specific data, any attempt to project the economics of a representative agro-forestry scheme will be largely imaginary. The following is an imagery one hectare scheme based on estimates and on field surveys of tree planting and related activities carried out in Malawi.

- The hectare is currently used to produce maize for sale. Annual revenues are $300, costs are $225, profits are $75.

- A 5-metre strip is to be pre-empted for a windbreak on the windward side of the field. One hundred Azadirachta indica trees, or twice as many as necessary to control wind erosion alone, will be planted over a four-year rotation. Starting in Year 5, one-fourth of the trees will be coppiced annually and sold as building poles for $.40 each.

- 200 farm trees (fruit, A. albida, L. leucocephala, etc.) will be scattered in the field and around its boundaries, pre-empting another 6% of cropland. Starting in Year 3, two cubic meters of firewood worth $1 at the farm gate will be trimmed from these trees annually. Starting in Year 7, $5 worth of fruit will be sold each " year.

- By Year 10, maize yields and revenues on the remaining land will be 22% higher than they would have been without the trees. (Costs of planting and harvesting the maize remain constant.) This assumes an absolute increase in yields of 11% over ten years, compared with a decrease of 11% if the trees had not been planted. The net rate of improvement is a steady 2% per year.

- Labour to plant and tend the trees will be worth $2.00 in Year 1, $1.40 in Years 2-4, and $1.20 in Years 5-10. Labour is valued at US$ .40 per day.

- Tree seedlings cost $.01 each.

Given these fairly plausible assumptions, costs and benefits of the project over its first ten years will be as shown on the following page. Note chat we are interested here only in the changes that result when trees are added to a hectare of land on which maize has traditionally been grown. Thus, we specify the total cost of planting and maintaining the new trees, along with the trees' direct output: poles, fruit, firewood. As far as maize is concerned, however, we include only the increase (or decrease) in maize profits that is attributable to having the new trees on the land.

These results illustrate several points that are likely to apply to any agro-forestry project. First, the direct cost of planting trees on this scale is quite small. Seedlings are cheap; and not much labour is required, especially if land is anyway being cleared and prepared for crops. The major cost is the (temporary) reduction in output of crops as land is diverted to trees.

Second, most of the financial benefits come from crop yields that increasingly outstrip what could be achieved without agro-forestry. Next most important are proceeds from the sale of building poles and fruit. Firewood is a relatively minor by-product of any such scheme.

Finally, the project has an impressive internal rate of return (61%). This should be comfortably above smallholders' discount rates, even though these may themselves be quite high (30-40% or more). If smallholders saw the project the way it is portrayed above, in other words, they would find agro-forestry to be very attractive.

The farmer's view

A hypothetical group of farmers would be likely to agree in part with our assessment of the costs and benefits of agro-forestry. Agreement would be most complete in terms of trees for poles, fruit and firewood. Tree planting for these purposes is a familiar activity in many developing countries, and farmers would be confident of the validity of our predictions.

However, farmers might not agree as readily with our estimate Chat crop yields will increase 1% annually with agro-forestry, and decrease 1% annually without it. Even if these results seemed possible in theory, there would be nothing in the farmers' own experience to confirm them. Farmers might therefore consider it too risky to accept these estimates as the basis for new investments. For their own calculations, they could assume a smaller benefit in which they had greater confidence. No matter what outsiders believed was "true", it would be these smallholder judgments that would determine whether agro-forestry was accepted.

One-Hectare Agroforestry Project

(Incremental Changes Due to Adding Trees to Maize Land)





















Net Increase in Maize Profits



































Net Decrease in Maize Profits






Tree Seedlings



































Rates of return are very sensitive to these judgments, especially since improved crop yields account for most of the scheme's presumed benefits. If crop yields were expected to improve at only one-half the rate stated above, the project's internal rate of return would fall to 35%, or no more than the discount rate of many farmers. Of course, farmers do not actually calculate these rates. However, they will measure an intuitive "rate of return" against an intuitive "discount race" to decide whether an investment seems attractive. Given the assumptions made here, they might well choose not to proceed.

Under these conditions, governments would have to decide whether some kind of encouragement was needed to bring agro-forestry into being. A government's enthusiasm for such action would depend on its own evalution of the economic costs and benefits of agro-forestry investments.

B. The government's perspective

Governments have discount rates (say, 12%) that are much lower than those of farmers due, in part, Co farmers' consideration of risk. Since the scheme above has an internal rate of return of 35% even with cautious assumptions about crop yields, any government would be likely to view the investment favourably.

Before coming to a final judgment, a government might wish to shadow-price the project's inputs and outputs. From an economic point of view, for example, governments often value farm labour at less than the market wage. On the other hand, the economic value of crops could be considered higher than the price actually paid to farmers by marketing boards. The value of firewood in terms of replacements such as kerosene could be far above its traded price. All these adjustments would increase the rate of return of an agro-forestry project.

Governments also need to consider the external benefits that do not accrue directly to farmers making agro-forestry investments. For example, siltation from soil erosion can negatively affect downstream irrigation systems, hydro-electric potential, and the cost of treating urban water supplies. Increased run-off can lead to flooding far from eroded areas. Any reduction in these kinds of damage should be considered an economic benefit of the project.

In practice, all these factors will tend to make agro-forestry even more attractive to governments than to individuals.

Although conclusive evidence is lacking, there seems a strong presumptive case for governments and donor agencies to begin supporting agro-forestry programmes. Such support need not be costly. Since agro-forestry is a way of modifying farming systems, techniques and inputs would logically be transmitted through the agencies already responsible for serving farmers. Even if subsidies were needed, this might involve little more than the distribution of low-cost tree seedlings, something that many governments are already doing in support of other objectives. Specific issues that governments and donors would face in promoting agro-forestry are considered below.

V. Preparation and implementation issues

A. Rural appraisals
B. Project components
C. National and local institutions

Issues of special concern for agro-forestry include the need for rural appraisals, the nature of project components, the choice of implementing institutions, and provision for monitoring and evaluation.

A. Rural appraisals

Agro-forestry projects must be consistent with local farming systems. Therefore, the first step in preparing such a project is to specify what these systems are like.

This is complicated by the fact that existing data on a country's farming systems may exclude information that is essential for thinking about agro-forestry. As noted earlier, for example, it is relatively new to consider trees as an integral part of farming systems. Available studies are therefore likely to leave unconsidered such points as current tree-planting practices, sales or purchases of tree products, and so on.

To remedy these deficiencies, an agro-forestry project may have to begin with an appraisal of key rural realities. Among the information to be collected in each project area would be the following.

- Physical characteristics (including altitude, rainfall, slopes, water supplies, soil condition, visible erosion). This is basic background for evaluating the need for agro-forestry and the local suitability of various techniques.

- Current uses of trees and shrubbery. This suggests the kind of subsistence products that an agro-forestry system would be expected to provide.

- Sales and purchases of agro-forestry products (including poles, fruit, firewood, fodder, etc.). This provides data for economic analysis, and indicates opportunities to replace purchased items or to expand sales by raising agro-forestry products.

- Current tree planting (including species, source of seedlings, intended use). This shows the present state of silvicultural knowledge.

- Farmers' perceptions of deforestation and erosion (including any perceived impact on crop yields). This gives a sense of how critical farmers think their problems are, and indicates current awareness of agro-forestry relationships.

- Land and tree tenure. This shows whether farmers have a right to their trees, and therefore whether they have an incentive to plant.

The job of collecting these data is less formidable than it may at first appear to be. Using a carefully-structured questionnaire, most information could be acquired from farmers through intensive interviews lasting not more than a half-hour per household. Allowing time to move from village to village, one person could average eight interviews per day. During project preparation, a hundred households might therefore be interviewed in the project area. Additional interviews could be carried out as the project was getting underway. These would seem minimum requirements for structuring any new agro-forestry project.

B. Project components

The exact nature of any agro-forestry project will depend on the results of the rural appraisal outlined above. However, it is safe to assume Chat most projects will involve some form of research, demonstration, on-farm trials, extension, and monitoring and evaluation.


Research must be done on the impact of including trees and other agro-forestry plants within farming systems. Both exotic and local agro-forestry species should be tried. Harvesting regimes should be designed to provide the products that people actually want, as determined by rural appraisals.

Since many agro-forestry species grow slowly, research on these issues is a long-term endeavour. However, it may also be a long-term endeavour to convince large numbers of people to adopt complete agro-forestry systems. If research is begun immediately, results may therefore be known at just the point when people are looking seriously for proven techniques.


Even while research is still underway, it will be possible to demonstrate basic agro-forestry practices whose results are relatively well understood: for example, windbreaks; scattered plantings of Acacia albida, Leucaena spp., and fruit trees; live fencing; etc. Demonstration areas should incorporate crops, trees, and harvesting regimes that are consistent with local practices and desires. Management of these areas should be flexible, with species and techniques added or subtracted as they prove more or less convincing.

An important part of this work would be to demonstrate agro-forestry plots alongside fields planted in traditional ways. This would allow farmers and other visitors to see quickly the differences made by agro-forestry techniques.

On-farm trials

The ultimate test of an agro-forestry system is its performance on farmers' fields. At the farm level, results (and perceptions of results) may be very different from findings on research or demonstration plots. Agro-forestry projects will therefore need to sponsor on-farm trials of the most promising techniques.

To be useful, these trials should as much as possible be conceived and evaluated by the farmers who are to carry them out. This implies the participation of farmers who are already fairly conscious of agro-forestry relationships. Results would be used to modify research and demonstration programmes, as well as to determine the agro-forestry message to be passed on by extension services.


The first extension job is to teach the extension agents themselves about interactions among trees, land, and food. Then, guidelines would be provided on how to evaluate farming systems, so that agents could guess what interventions might be appropriate under specific local conditions. Much information for this teaching programme would come from rural appraisals.

Once the extension agents understood agro-forestry concepts, they could begin to introduce promising species and ideas. This would be done gradually, starting with straightforward techniques such as boundary plantings of fruit and leguminous pole trees. More complex techniques would be channelled through the system after being proved successful during on-farm trials.

Monitoring and evaluation

Monitoring will be extremely complex, requiring measurement of yields (of both crops and trees) in very different places: research plots, demonstration areas, farmers' fields. These data are needed to assess the actual impact of various agro-forestry techniques.

Equally important is to determine farmers' responses to agro-forestry. This should be judged by the number of farmers who become involved, and by what they say about their experiences.

C. National and local institutions

Each agro-forestry project would require a multi-disciplinary implementing group. A social scientist would be needed to conduct rural appraisals, to help design on-farm trials and extension messages, and to monitor project activities in terms of farmers' reactions and socio-economic results. The research programme would require joint effort by an agronomist, a forester, and an animal husbandry expert. An agro-forestry extension officer would supervise the communications support system, in collaboration with the national extension service. Additional technical and extension staff would be based at out-stations to run agro-forestry demonstrations and on-farm trials.

If the process is functioning properly, participation of farmers is guaranteed throughout. The original choice of research priorities and extension messages builds on the situation and attitudes of farmers. Farmer's' groups are identified or created to carry out on-farm trials and participate in other activities. Monitoring gives voice to farmers as the project proceeds. At the end, farmers are asked to help evaluate what has happened. Of course, this participation will not happen automatically; vigilance must be maintained to ensure that the farmer's point of view remains at the project's heart.

Ideally, all of this work would be centred within the ministry already responsible for farmers and farming systems. It might prove desirable, therefore, to create an agro-forestry "cell" within the ministry of agriculture or rural development, made up of the kinds of professionals specified above.

In many countries, there exist non-governmental organizations (NGOs) that have close and continuing relationships with distinct groups of farmers. Some of these NGOs will already have experimented with tree-planting, generally as a means of meeting family needs for wood energy. Within an agro-forestry project, these NGOs could play an important role by carrying out demonstration activities and on-farm trials. However, close links would have to be maintained with the government, which presumably would retain primary responsibility for research and extension.

VI. Conclusions

In conclusion, it appears from recent research that agro-forestry can, in many cases, contribute to solving the fuelwood crisis by providing benefits which straightforward forestry or farmwood lots cannot provide, and thereby making it attractive to individual farmers, as well as communities, to plant trees.

Development workers must recognize first that farmers, in most countries, have nearly always practised agro-forestry in the fields, as well as around the homestead, until extension and development workers insisted on the advantages of single-cropping on each plot of land ...

Similar errors should not be repeated within agro-forestry projects by advocated new species and new associations of species, before the advantages of the local practices have been studied through discussion with farmers. Farmers' knowledge, and particularly women's, is usually very extended in this area, which provides both food and fuel of direct regular use in the household.

To propose new or additional agro-forestry pratices is, however, possible, as shown above, on two conditions:

1) that it is studied as an element of the overall farming, or rather farm-household, system; and

2) that its value to the farmer is checked both by discussion with farmers themselves (including women - whether heads of household or not) and by a financial and risk analysis at the farm level.

In general, agro-forestry practices will prove attractive indeed to farmers and governments alike, provided the mix of elements proposed include the production of at least one high-value product (quality timber, fruit, nuts, etc...)

Previous Page Top of Page Next Page