An e-publication by the World Agroforestry Centre
METEOROLOGY AND AGROFORESTRY
|
|
An e-publication by the World Agroforestry Centre |
|
METEOROLOGY AND AGROFORESTRY |
|
|
section 2 : basic principles Agroforestry systems: a production/protection land use strategy for steep land in the tropics
G. de las Salas
Consultant on Agroforestry Abstract This paper deals with the suitability of agroforestry systems as a protective/productive land-use alternative for some regions of the tropics.
Both economic and social role of forest trees and crops in combination (spatial and/or sequential) for the production of the rural poor is pointed out. Special attention is drawn to the environmental conditions for determining the potentials and limitations of agroforestry systems, notably: soil, climate, and microclimate, vegetation, pests and diseases. Perspectives of the use of agroforestry systems in soil conservation programs are analyzed.
Factors influencing the agricultural production systems like the agroclimatic conditions (distribution and amounts of rains, soil fertility, soil drainage, slope, microclimate among others), land tenancy, crops sequence and culture practices, must be sufficiently known. In the tropics the different systems of agricultural production and cattle raising predominant in each ecological region, are strictly related with the type of soils and the social pressure on the land. In tropical Latin America, on soils of high fertility, dominant production systems are perennial cultures like palm, coffee, cacao, sugar, cotton, banana and grasses for livestock. All these are stable systems which need renovation only after 15 or 20 years. On poor soils, the small farmer produces basic grains for his subsistence, utilizing shifting agriculture. This was efficient in the past when the population density was low and the requirements of the people were also low. The interval between seedlings was long enough to permit restoration of the soil. Under ever-higher demographic pressures and the growing need to produce crops at a commercial scale, it was forgotten that basic grain agriculture on poor soils and steep land is stable only when it is part of a system that provides soil restoration. Such a system is one that simulates the various storeys of a natural forest. Agroforestry systems appear to approach this condition and therefore have been favored by various research and donor institutions to demonstrate their efficiency in rural environments as a protective/productive land use strategy.
Fragility of Soils Many soils of the humid tropics are characterized by high acidity, aluminium toxicity, and low levels of available phosphorus, potassium and organic matter, all properties which indicates low fertility and limited potential for crop growth. Many of the nutrients utilized in traditional agriculture are naturally available from burning and from decomposition of organic matter rather than stored in the soil. Nutrient cycling mechanisms are located in the above ground living biomass and in the top roots. Under continuous vegetation cover or permanent cropping systems, very few nutrients are lost due to the closed system of bioelement recycling. However, as above ground biomass decomposes or is continuously disturbed the nutrient level is reduced due to leaching, soil erosion and nutrient export by crop harvest. Therefore, most soils of the humid tropics do not support continuous cropping under natural conditions. If crop yields are to be maintained, then fertilizer inputs, crop rotations and adaptive cropping systems will be required, such as: sustained annual crop production, pastures, plantation crops, forestry, alley cropping and the like.
There is widespread agreement concerning the scarcity of high quality agricultural land in developing tropical countries. Generally speaking, agricultural production will be able to keep up with population growth in the next decade provided it can grow by at least 2.5 - 3 percent per year without putting more — and not always suitable — land into crop and livestock production. The pressure to expand and intensify the crop land in the humid tropics has resulted in increasing demands on forest land. Sommer (1976) estimates the deforestation rate of the humid tropics at 110,000 km2 /year while Myers (1980) in a study for the National Academy of Sciences of the United States estimates high conversion rates from forest to agriculture in almost all the tropical countries studied, notably Brazil (400,000 km2 in few years); Colombia (2000 km2 / year); Peru (300,000 km2 by the end of the century); and Ivory Coast (35,800 km2 converted in 10 years). Forest areas will continue to be exploited and converted for agriculture and cattle raising as a result of foreign or internal pressures. This procedure has the following effects:
Forest Plantations The general picture in tropical America and perhaps in some parts of Africa is that afforestation is largely out of proportion with deforestation. Arguments for improving high-yield forest plantations intensively managed on suitable land, have never been more justified than now. Known examples are those from Venezuela (200,000 ha of Pinus caribaea on lowland) and Brazil (more than 2 million ha of eucalyptus managed for pulp production), both on savannah landscapes. In mountainous land, however, the situation has not always been satisfactory and has even been disappointing. Private owners have reafforested large areas mainly with conifers thinking in terms of income rather than in improving the environment. On volcanic soils the yields of these plantations have been satisfactory but the costs of exploitation (harvesting and transportation) are onerous so that there exists a 'bottle neck' to industrialize such plantations. The growing demand for wood in the tropics, along with intensifying competitive land use pressures, calls for a revision of land use planning in order to concentrate wood production on the most appropriate sites.
Trees and crops to do the job Selection of trees and/or crops to improve soil productivity while preventing soil erosion, requires careful consideration. Planting unsuitable species or tree/crop combinations can result in wasted effort and money and possible loss of time, if failures are slow to appear. There is no point in trying to put trees or crops in nearly soil-less sites because the growth and crop production will be poor. In such cases grasses would preferably be combined with an efficient legume like Cajanus cajan (pigeon pea) to ameliorate the physical condition of soil and to enrich it with nitrogen and organic matter. The selection of trees and crops as a land use strategy on steep land is largely a matter of assessing their adaptability to prevailing site conditions, their acceptance by the farmers, the farmers' needs for fire wood, poles, timber, food, etc. and the rentability of the system rather than its efficiency in preventing soil erosion. In fact, the first priority may be to obtain these by-products, if conservation is to have local acceptance. In Panama for example, Byrsonima crassifolia is tended by the farmer since its fruit is largely utilized in the preparation of juice locally named chicha de nance. In the watershed of Paute River, Ecuador, where erosion control has become a great problem, farmers accept technical assistance in increasing their crop production by means of measures that include planting fruit trees and other trees known in the region for centuries (e. g., Eucalyptus globulus, Prunus capuli) rather than to plant 'miracle' trees in their farms. E. deglupta is used in Costa Rica at the age of 1 year as training poles for banana trees. Okigbo (cited by Kunkle 1978) reports that in south-eastern Nigeria, cashew trees (Anacardium occidentale) were first planted as an anti-erosion measure, but now the fruits and seeds are also important by-products used for both food and commercial purposes.
Changes in farming methods and land use — brought about in great measure by modern agricultural policies in developing countries — have resulted in increases in soil productivity (e. g., coffee, rice, cotton). But these changes also have originated a series of environmental problems that include soil erosion and appearance of pests not previously known. As important factors contributing to the erosion problem, the following deserve mention:
This form of range management has led to various types of erosion, among which the so called 'cattle step' is the most widespread in mountainous landscapes. The damage done to forest and watersheds by this type of cattle management exceeds by far the poor returns from meat production if one takes into account the low carrying capacity of many tropical areas (mostly one animal per 5 or more hectares). Fortunately acid savannah landscapes are being successfully managed by grass varieties adapted to these edaphic characteristics (Hecht 1979). Some parts of the humid Amazon area are under experimentation with grass-legume combinations to increase cattle production. Various scientists (see Sanchez 1976, 1979; Hecht 1979; Toledo and Morales 1979; among others) are optimistic over the experimental results while others (Deurojeanni 1982; Goodland and Irwing 1975; Shubart 1977; Fearnside 1978) are pessimistic and foresee devastating erosion and they doubt the applicability of the experimentation to large areas. Although this subject is not the matter of this paper, its importance is unquestioned. Much research is needed to obtain an accurate appraisal of soil erosion problems.
Soils As pointed out previously, many of the soils of the humid tropics have low fertility, especially lowland soils. There are, however, areas occupied by fertile soils (Andosols, Nitosols, Vertisols, Inceptisols). Although notably smaller in extent, the majority of them are under coffee. On the other hand, foothills with relatively high fertility due to a large amount of organic matter and to a fertile parent material, are already occupied by vegetables, pastures and other land uses (e. g., cacao, cassava). Poor and partially steep soils remain under 'clean crops' like maize, onion and rice, among others, all of them subsistence crops. Unfertility of soils and steep landscapes are two strong limitations to bringing agroforestry techniques to the small farmer successfully. Furthermore, the small size of the farms makes it difficult for the farmer to plant part of the terrain with woody species, being the crop production essential for the farmer (at least in simultaneous tree/crop models). Some of the poorest soils tend to produce yields that decrease continuously for 8 to 10 years and then stabilize. The main problem here is that their management is still in its experimental phase and it is risky to extrapolate research results. However, there are a few successful examples. In fact, in the 'Sierra peruana', soil conservation measures (terraces, mulching, green manure) accounted for 60% of the improvement of the crop production of the small farmer using a crop/tree spatial model on treated slopes. A second example is that of the 'selva alta', also in Peru, where the German Agency for Development successfully experimented with woody species in private small farms. Ecologically, the problems in Andosols, Fluvents (alluvials) and Ultisols are minor, as organic matter could be increased, soil protective is feasible and the proper combination of chemical, organic and green fertilizers could be established, if the economic profit supports it. Most difficult is the management of Alfisols, Vertisols and even some Andosols that are abundant in mountainous landscapes in tropical America and perhaps also in Africa. In such soil types, the shallow depth (generally 30 to 40 cm), the presence of expansible clay and the erodibility of the top soil, respectively, make difficult the application of efficient cultural practices. Summarizing, crop/tree production — both spatial or sequential on small farmer's land of the tropics — is based more on economic than on ecological factors.
Moisture, temperature, and rain distribution are normally considered to be important factors in field crop production. Of these, moisture relationships (especially seasonal variation of evapotranspiration) determine the more intricate ecological relationships and therefore have been considered as the most important determining factor in crop ecology. In the humid tropics the chances of having abundant water supply in critical periods are higher than in the regions of less — and more variable — precipitation. The adoption of resistant varieties and crops more adaptable to different climates has been a long-term answer to the problem. Agroforestry systems are more difficult to evaluate since the interactions among plants of different size and root systems cause changes in at least the following factors:
Failures in identifying and monitoring these microclimatic factors can lead to losses in crop production. In general, however, farmers have, by trial and error, adopted the best practices to manage their crops.
This aspect constitutes an important matter of study to derive the limitations and potentials of agroforestry systems in tropical environments. It is well known that the more primitive (and consequently diversified) an ecosystem, the more resistant it is to pests and diseases. The incidence of diseases and pests is of utmost importance in the humid tropics. Attacks of insects, fungi, viruses, nematodes and other parasites affect the growing crop, and thus the final product. On the other hand, the indiscriminate use of pesticides has led to many failures. Ecological guidelines have to take into consideration besides the economic implications. A necessary prerequisite is a background of biological information. The control of diseases with chemicals has severe effects on the ecological complex, but these are not so serious as those caused by insecticides. The use of pesticides is, especially in tropical areas, spreading rapidly. At the same time, farmers and research workers feel an urgent need to develop effective alternative approaches, e.g., integrated pest management (IPM).
Agroforestry systems: where to implement them? Here we are concerned with the application of agrometeorological research to the changes in both macro- and microclimate. The results must be transferred to the rural poor in an intelligent and easy way, so that they can take advantage of the improvements in farm management methods. The combination of livestock with trees planted for fuel, helter and shade, fodder, or fence posts, can be very satisfactory in flat land. Successful examples exist in various countries of the American tropics. In hilly land, however, this kind of combination has to be practiced carefully since the lack of ground cover under tree plantations can lead to severe soil erosion. Pereira (1979) reports examples of such damage under eucalyptus in East Africa and Ethiopia and the same can be seen in the Andean countries of South America. Notwithstanding this fact, successful case studies deserve mention. Budowski (1981) comments that for at least 80 years, a local alder species, (Alnus acuminata), has been successfully planted at elevations between 1300 and 2500 m asl with high rainfall (2000 - 3000 mm) on good soils in the dairy region of Costa Rica. The presence of alder trees is said by farmers to increase fodder production. The trees are planted at wide spacings within pastures that are grazed (Pennisetum clandestinum) or cut (Pennisetum purpureum and Axonopus scoparius). Most alder trees found in pastures are regularly pruned. This species fixes nitrogen through large nodules. Some initial measurements indicate a mean annual increment in diameter of 2-3 cm with harvesting after 15-20 years. The wood is easily workable for multiple purposes. In Colombia, a 16-year-old stand of alder is being harvested by a match industry. Another example in the same country is the silvopastoral enterprise in more than 500 hectares in the highlands (1800 - 2200 m asl) near the city of Manizales in which Pinus patula, Cupressus lusitanica and Alnus acuminata stands are being grazed without diet supplements (except mineralized salt). The taungya system — as a part of a strategy to change land use or to protect watersheds — is well known. Pereira (1979) reports that the taungya ('shamba') system (pines planted between vegetables) in Kenya successfully avoided both significant soil erosion and loss of streamflow regulation. Today, with the pines approximately 30 years old and ready for harvesting, water use of the two species remains equal. The Forest Department is continuing with the plantation program in which agroforestry plays an integral part. In tropical America and the Caribbean, taungya has a successful history. In Costa Rica, tree species of a high commercial value like Cordia alliodora, Gmelina arborea, Eucalyptus deglupta and Terminalia ivorensis associated with maize, beans, cowpeas and string beans have demonstrated at an experimental level how to convert large areas of degraded lands into valuable tree plantations. Haiti is reafforesting thousands of hectares of steep eroded land. Thousands of farmers are expected to be involved in the programs which allow them to combine field crops with fuelwood trees (such as the excellent Azadirachta indica named locally as 'neem'). The rising prices of firewood posts and timber and the possibility of eventually obtaining cheap and large sources of fodder and biomass on marginal agricultural lands have generated great interest among funding agencies.
Soil protection under tree crops. Afforestation is normally an acceptable method to reduce erosion and help control flooding while at the same time providing timber and/or firewood. However, this statement must be regarded with caution, since on steep land with fragile parent material (e.g., saturated and decomposed ash under shale), afforestation could be the worst land use alternative. In such landscapes, big natural landslides occur frequently and the risk of a failure is very high. On stable soils afforestation can fulfil not only a protection objective but also a production one: getting products of direct value to rural people such as fruit, fodder, fence posts, timber, since these by-products can be as important as the protection itself. The most important function of tree or shrub vegetation for soil protection is to protect against the splash erosion caused by raindrops. Kunkle (1978) cites an experiment in Africa in which the 10-year average soil loss from bare soil was 126.6 ton/ha; but when the soil was protected from the raindrops with a fine wire gauze, the rate of loss was reduced to only 0.9 tons/ha. In upstream afforestation programmes in the Cauca Valley in Colombia, bamboo trees have been selected not only for soil conservation but also to yield benefits for farmers. For steep land, FAO (1976) has developed some useful conservation techniques. The following is a list of desirable characteristics of tree species for soil conservation programmes (Kunkle 1978):
Soil stability and crop/tree production In Honduras and Guatemala, burning the bush to prepare the soil for corn and beans is very widespread. This has led to a strong selection of fodder trees. In fact, there exists a traditional silvopastoral system, having an enormous potential and requiring only low inputs. Leguminous species like Gliricidia sepium and Leucaena spp have been associated with corn and beans since pre-Colombian time, a system attributable to the Maya culture. The cultivator traditionally leaves and tends the naturally-regenerated trees of laurel (Cordia alliodora) on the 'milpas' (abandoned bush fallow). This kind of land management deserves more attention and improvement, since it fulfils two goals at the same time: soil stabilization and crop or tree production. In humid environments, the tending of the natural regeneration in the native forest for fuel, forage and other products, permits a great opportunity to improve the land use of small watersheds. This opportunity is followed by the 'agroforestry chance' to introduce trees into the crop and livestock areas with the main objective of leaving seed-bearing trees to regenerate the pine (Pinus oocarpa) forest after the agriculture or cattle raising has been abandoned. It goes without saying that the involvement of the local community in the support of agroforestry and conservation programs is extremely important. In this respect, the FAO Conservation Guides 12 (FAO 1985) and 14 (FAO 1986) deserve careful reading.
Technical data and site-selection criteria as well as measurements on stream flow and discharges in agroforestry experiments in small watersheds vs. other land uses, should be worked out by hydrologists. Investigations of kinetic energy of raindrops in relation to soil erosion should be carried out. The Universal Soil Loss Equation (USLE) should be tested in agroforestry systems (and in monocultures as well) in order to evaluate the soil loss rate over a reasonable period of time. Small watersheds with various land uses including agroforestry systems (e.g., coffee with and without shade), should be accurately instrumented in order to record:
Foresters, agronomists and land - use planners should be realistic when stating the benefits derived from associated trees and crops on steep lands. The introduction of agroforestry systems alone does not necessarily result in reduced flood damage, reduced drought, increased rainfall or the control of erosion and sedimentation. There is a need for 'enlightened land-use techniques' for conserving soil while maintaining high productivity on critical steep mountain areas. Greater supervision and control of upland use, including grazing and logging, must be accomplished soon. A long-term strategy to control human and livestock populations in the uplands in addition to developing systems of producing trees and crops when appropriate, deserves high attention by land-use planners and decision-makers in tropical countries. In addition, the development of community organizations to restore wasted lands and non-destructive forest management practices should be encouraged.
Budowski, G. 1981. Agroforestry in Central America. In Agroforestry. Proceedings of a seminar held in CATIE, Turrialba, Costa Rica. 23 February - 3 March 1981. De Camino, R. 1985. Incentivos para la participacion de la communidad en programas de conservacion. FAO Guia de Conservacion Guide N. 2. Rome: UN Food and Agriculture Organization. Dourojeanni, M J. 1976. Una nueva estrategia para el desarrollo de la Amazonia per-ana. Invest. Agropecu. Pent 8: 41-58. FAO 1976. Hydrological techniques for upstream conservation. FAO Conservation Guide No. 2. Rome: UN Food and Agriculture Organization. FAO 1986. Strategies, approaches and systems in integrate watershed management. FAO Conservation Guide No. 14. Rome: UN Food and Agriculture Organization. Fearnside, P.M. 1978. Estimation of carrying capacity of human populations in apart of the Transamazonic highway colonization area of Brazil. Ph.D. Thesis Univ. of Michigan. Ann Arbor, Michigan. Goodland, R.J.A. and H.S. Irwin. 1975. Amazon jungle: green hell to to red desert! Amsterdam: Elsevier. Hecht, S.B. 1979. Spontaneous legumes on developed pastures in the Amazon and their forage potential. In P.A. Sanchez and L.E. Tergas (eds.), Pasture production in acid soils of the tropics. Cali, Colombia. Kunkle, S.H. 1978. Forestry support for agriculture through watershed management, windbreaks, and other conservation measures. In Proceedings of the Eighth World Forestry Congress. Jakarta, 16-28 October 1978. Myers, N. 1980. Conversion of tropical moist forest. Report to the National Academy of Sciences, Washington, D.C. Pereira, Ch. sir. 1979. Hydrological and soil conservation aspects of agroforestry. In H.O. Mongi and P.A. Huxley (eds.), Soils Research in Agroforestry: Proceedings of an Expert Consultation. ICRAF Nairobi, 26-30 March 1979. Sanchez, PA. 1976. Properties and management of soils in the tropics. New York: Wiley. Sanchez, PA. 1979. Manejo de suelos tropicales en la Amazoni suramericana. Suelos Ecuat. 8:1-11. Shubart, H.O.R. 1977. Criterios ecologicos para o desenvolvemento agricola das terras firmes da Amazonia. Acta Amazonica 7(4): 559-567. Sommes, A. 1976. Attempt tp an assessment of the world's tropical moist forests. UNASYLVA 28: 5-24. Toledo, J.M. and VA. Morales. 1979. Establishment and management of improved pastures in the Peruvian Amazon. In P.A. Sanchez and L.E. Tergas (eds.), Pasture production in acid soils of the tropics. Cali, Colombia: CIAT. |