H.N. Le Houérou
2. The acacia albida agropastoral system in Africa
3. Fodder shrub plantations in the Mediterranean basin
6. Methods of establishment and utilization
Agroforestry can be defined as the use of forest trees on farmland and their integration within farming or agropastoral systems. It is an ancient technique for maintaining soil fertility in marginal environments and has been used for centuries in rural civilizations in various parts of the world. Agroforestry has recently been rediscovered by modern scientists and agronomists as an excellent tool for maintaining the long-term biological balance of agricultural and livestock production systems, especially, but not only, in the arid and semi-arid zones. Agroforestry has also been developed in montane areas and in the wet tropics. Some examples of agroforestry systems are:
a) the cultivation of mulberry trees for silk production in the Far East and, later, in Europe;
b) the Prosopis cineraria — millet agropastoral system in India (Rajasthan) and Pakistan (see Mann and Shankaranayan in the Proceedings of this Symposium);
c) the Acacia albida — millet agropastoral system in West and eastern Africa (Senegal, Mali, Upper Volta, Niger, Nigeria, Ethiopia, Zambia and Malawi);
d) the "Dehesa" agropastoral system of western and southwestern Spain (Estramadura);
e) the olive grove agropastoral systems in the Mediterranean basin;
f) the Prosopis tamarugo pastoral system in Chile;
g) the Leucaena leucocephala agropastoral system in Central America, the Indo-pacific Islands and, later, in Africa (Malawi);
h) the Mulga pastoral system of Australia;
i) the Opuntia Ficus indica agropastoral systems in Mexico, then in the Mediterranean basin (Sicily, North Africa), Brazil, Madagascar and South Africa (Karoo);
j) the Shea-butter trees (Yitellaria paradoxa) - sorghum system in the Sudanian zone of West Africa;
k) the Neré (Parkia biglobosa)/sorghum system in the Sudanian zone of West Africa;
l) the Borassuspalm - sorghum - groundnut system in the Sudanian and Guinean zones of West Africa;
m) the various Baobab farming systems in numerous African countries;
n) the Gum-Arabic (Acacia Senegal - sorghum bush fallow system in the Sudan;
o) the Carob orchards of Cyprus and other Mediterranean countries.
Most of these systems developed independently in differing types of environment under highly varied socio-economic conditions. They are therefore likely to be a rational response to the need for long-term sustained productivity in agricultural and agropastoral systems. The objective of this paper is to outline the scientific justifications for such systems.
This is not the place, given the short time and space available, to review all these systems in detail; I shall therefore restrict myself to discussing those with which I am more familiar, and attempt to reach some general conclusions.
Acacia albida is a forest tree widespread in the semi-arid and subhumid zones of tropical Africa from sea level up to about 1800 m2/. It occurs over a wide range of ecological conditions, mainly on deep, sandy, well-drained soils having a deep water table (2 to 3 m and more). This tree has a special biology in that, unlike most other tropical deciduous trees, it sheds its leaves during the rainy season and keeps them during the dry season, i.e. from October to June in the northern tropics. Acacia albida is protected and kept in millet fields in a number of peasant civilizations in Africa (among the Serere in Senegal, for instance). The density of trees is 10 to 50 per ha, the adult trees are 10 to 20 m tall and 5 to 10 m in canopy diameter. Canopy cover is thus 2% to 40% of the ground area. Millet is cultivated under and between the trees; as light interception by tree canopy during the millet growing season is negligible, the shade factor has no influence on the crop grown under the trees.
2/ Exceptionally 2300 m (in Jebel Mara, Sudan).
The leaves of Acacia albida are shed at about the time when ploughing begins, so that to a large extent they are incorporated into the soil. It has been calculated that the leaves entering the soils are equivalent to fertilization of up to 50 tonnes of manure per ha/year in dense stands of 50 large trees/ha. However, the average figure is probably more like 10 tonnes/ha/year. The amount of organic matter in the upper layers of the soil is twice as high under Acacia as it is on open ground (Charreau and Vidal, 1965). The geobiogenous elements returned to the soil annually through the litter have been evaluated as shown in Table 1 (in kg per ha/year), with a population of 16 trees/ha covering 24% of the ground (Charreau and Vidal, 1965):
Table 1. Geobiogenous elements returned to the soil from Acacia albida (kg/ha/year)
Ca |
Mg |
K |
N |
P |
S |
120 |
25 |
13 |
75 |
12 |
20 |
Overall microbiological activity is 2 to 5 times higher under Acacia (Jung, 1970). Mineralized nitrogen content is also 2 to 5 times higher, while phosphorus and other elements are also sharply increased (Charreau et Vidal, 1965; Jung 1970, Giffard, 1971; Dancette and Poulain, 1969). As a result, millet production in the system is almost twice that obtained from purely open-land farming systems without fertilization, i.e. conditions such as those found in semi-arid West Africa (1000 to 1500 kg of grain/ha/year, as against 500-800 kg).
In addition, the annual production of pods per tree averages 20 to 30 kg of DM, i.e. 400–600 kg/ha/year, and has a feed value of 0.85–0.95 FU per kg, with 8–12% DP giving a theoretical stocking rate of 1.5–2.0 sheep or 0.15–0.20 head of cattle per ha from pods alone. Moreover, this is a concentrate feed enabling livestock to make optimal use of the mediocre, protein-deficient roughage available during the dry season, such as millet and sorghum stubble and straw, which otherwise would be consumed in much lesser quantities and without much benefit for the animals. In the same areas of West Africa current stocking rates on natural pastures are about half the above figures (1 TLU/10 ha).
Branches and foliage are lopped on a rotation system about once every 3 to 5 years and fed to livestock, which readily consume them. The wood from lopped branches is used for fences and fuel. The total economic output of the system for the farmer is probably about three times greater than in open-land millet cultivation. In addition, soil fertility is maintained and soil erosion is kept at a minimum, unlike in open-field conditions.
This system is so successful that about a century ago Dina Djenne, Sultan of Zinder (Niger), issued a decree whereby whoever cut down an Acacia albida tree would be beheaded. Large-scale plantations have been undertaken in several countries of the Sahel for the past decade. Similar results have been found in the Prosopis cineraria millet agropastoral system in Rajasthan (Mann and Sankhabarayan, 1980). Another well known case of agroforestry production systems in the dry tropics of Africa is the millet sorghum and Acacia Senegal long term rotation in the Sudan, also called gum-arabic bush fallow (Seif el Din, 1969).
In order to stop range deterioration, erosion and desertification, large-scale planting programmes for fodder shrubs and trees have been undertaken over the past decade in several countries of the Mediterranean basin such as Iran, Syria and Tunisia. These plantations are used as dry-season feed reserves (particularly from August to November), providing roughage which is usually rich in protein. At the same time they are used for sand dune stabilization and the rehabilitation of desert encroachment areas. The main species used in large-scale plantations are the following:
Salt bushes: Atriplex nummularia from Australia, A. halimus from the Mediterranean basin, A. canescens from northern America, A. lentiformis from northern America:,
Acacia: Acacia cyanophylla, A. salicina, A. aneura, A. victoriae from Australia, A. raddiana from the Mediterranean basin, A. liguata from Australia.
Mesquites: Prosopis dulcis from N. America, P. chilensis from S. America.
Cactus: Opuntia ficus indica var. inermis from N. and S. America.
Saxaouls: Haloxylon persicum from Middle East, Iran, Russia; H. aphyllum from Russia.
Tree Medic: Medicago arborea.
Tree coronilla: Coronilla glauca.
Calligonums: Caligonum polugonoides from Iran, C. comosum from Libya, Tunisia, C. azel from Tunisia, C. arich from Tunisia.
Honey locust: Gleditschia triacanthos from N. America.
Other species, both local and exotic, are under experimentation, and some are very promising. Those already available on a large scale make it possible to meet most of the ecological conditions in the arid and semi-arid zones of the Mediterranean basin.
There is too little hard scientific information on the influence of these plantations on soil fertility. However, from qualitative field observations and from the few measurements carried out, it is obvious that productivity of the land is greatly increased (200 to 500% for successful plantations). Soil analyses in Tunisia show that in cactus plantations in the arid zones (rainfall of 200 – 300 mm) the organic and nitrogen contents of soil under cactus are at least double those on open-land test plots (Monjauze and Le Houérou, 1965).
In the Mediterranean basin additional agroforestry systems have been in use for centuries: olive groves, for instance, are often grazed and the leaves and twigs from pruning are fed to livestock. Production of browse per tree is 1020 kg, i.e. 200–2000 kg/ha/year, depending on tree density and size. As there are some 8 million ha of olive trees in the world today, overall forage production would be about 8 million tonnes/DM/year, the theoretical equivalent of the annual dietary needs of 15 million sheep. Again, there is little if any hard data on the output of this system, in which olive and livestock production are combined. This seems to me an interesting subject for research.
An intensive research programme on the Dehesa system is under way in Spain. (Unfortunately preliminary reports from this programme were not available to me when preparing the present paper).
The Dehesa is a parkland of oak trees consisting of mainly holm oak (Q. ilex s.l.), cork oak (Q. suber), with a density of 10–25 trees/ha. The land is periodically farmed for cereals (once in 5–8 years) and used as grassland for the rest of the time. Due to this periodical cultivation there are few ligneous species. The acorns are harvested by pigs of the Iberian breed and the pasture is used by sheep as well. Over the last decade the system has been endangered by the spread of African swine -fever, which has not yet been curbed. The present aim is to convert the system to intensive sheep production, made possible by the introduction of sown subterranean clover pastures together with phosphate fertilization.
Grass production under tree canopy in the Dehesa is usually much higher than on open land, again due to the greater organic content of the soil, which is fertilized by leaves and livestock droppings, resulting in a higher plane of nutrition for the animals raised. This is also the case for dry tropical African savanna, where natural grass production under the tree canopy is about two to three times higher than in the open, with a photosynthetic efficiency rate of up to 1.4%, as against 0.3% in the arid zone of Senegal (Bille, 1978) and 4.5 -1.0% in the semi-arid zone of Mali (Le Houérou, 1979). Potential evapotranspiration is reduced by 50–70% under shade and the grass remains green 4 to 6 weeks later in the early dry season (Pratchett, pers. comm.).
However, this is not true everywhere. In Australia, for instance, it has been shown that the removal of trees (Eucalyptus spp., Malaleuca spp), considerably increases forage production in the grass layer. But this case may be an exception rather than the general rule, and may be linked to allelopathic or phytotoxicity phenomena.
The establishment and utilization methods for sylvoagropastoral systems vary, with a wide range of ecological, technical and socio-economic conditions between different countries and areas.
The following three combinations are broadly typical:
a) expensive labour combined with a high technical level of farmers or labourers;b) cheap labour combined with a low technical level of farmers, shepherds or labourers;
c) expensive labour combined with a low technical level of farmers, shepherds or labourers.
In the first case, most of the operations must be mechanized, i.e. direct sowing, with the harvest either mechanized or directly browsed by animals. However, not all the species mentioned can be established by mechanical means, for example those multiplied by vegetative propagation such as cacti. The collection of seeds, also usually requires hand labour. As a result, agroforestry techniques are rather expensive in developed countries and are for this reason not popular among farmers.
In the second case the costs of establishment can be reckoned in terms of mechanized reafforestation techniques, i.e. growing of seedlings in nurseries and their manual transplantation in situ at the beginning of the rainy season. On the other hand, when management is traditional, direct utilization by browsing is very hazardous, as there is always a danger that overbrowsing will occur, destroying the plantation. In this case, since labour is cheap, a zero-grazing technique (cut-and-carry) should be used.
The third case is the most complex and occurs mainly in oil-rich countries. The methods used may be a combination of cases (a) and (b). Plantations are established at very high cost, but the price of local animal products is so high that agroforestry techniques are still viable. A typical combination might be direct mechanical sowing, followed by utilization through zero grazing; or, alternatively, hand planting of seedlings and direct grazing by livestock.
I have attempted in this paper to draw the attention of the participants in this symposium to a few successful examples of agroforestry in various countries having a variety of ecological and socio-economic conditions. It is clear that in many areas of the world there exist well balanced, agropastoral systems geared to sustained production in the long term, and which have survived for centuries through all kinds of vicissitudes.
Let us think twice before rejecting or destroying them. Our experience of mechanized agriculture has been a very short-term affair by comparison. The cost in energy of producing 1 calorie of food is skyrocketing (Speddling and Walsingham, 1978). The point of breakdown may be nearer than we think. We should therefore develop alternative strategies, and agro-forestry is one of these. The National Research Council of the National Academy of Science of the USA is already making efforts along these lines. Several publications have been devoted to this subject in recent years, and more are in preparation. In addition, a new international research institute has recently been set up in Nairobi: the International Centre for Agro-forestry (ICRAF). Development of agropastoral production systems is, indeed, an old approach to new problems or, as Democrites put it 2500 years ago, "the fruit of casuality and need".
Instit. Nac. de Invest. Agr. (1979). 'Resumen de trabajos sobre mejoras de la Dehesa extremana – l'. Estudios besicos. Instit. Nac. de Invest. Agr. Cap de Ahorros y Monte de Piedad de Carceres.
Charreau, C. Nicou R. (1971). 'L'ámelioration du profil culturel dans les sols sableux et sable argileux de la zone tropicale séche ouest africaine et ses incidences agronomiques'. Agron. Trop. 26, 209–255, 531–565, 903-978, 1183–1347.
Charreau, C. Vidal P. (1965). 'Influence de l' Acacia albida sur le sol. Nutrition minérale et les rendements des mils Pennisetum au Sénégal'. Agron. Trop., 67.
Dancette, C., Poulain J.F., (1969). 'Influence d'Acacia albida sur les facteurs pédoclimatiques et les rendements des cultures'. Sols Africains, XIV, 182, 143–184.
Giffard, P.L. (1965). 'Les possibilités de reboisement en Acacia albida au Sénégal'. Rev. Bois For. Trop. 95, 21–33.
Giffard, P.L. (1971).'Recherches complémentaires sur Acacia albida Del'. Rev. Bois For. Trop., 135, 3–20.
Jung, G. (1970). 'Variations saisonniéres des caractéristiques microbilogiques d'un sol ferrugineux tropical peu lessivé (DIOR) soumis ou non à l'influénce d'Acacia albida Del'. Oecol. Plant, V, 113–136.
Lebrun, J. (1968). 'A propos du rythem vegetatif d'Acacia albida del'. Collect. Bot., VII, 11, 625–636.
Mann, H.S., Shankarnaryan K.A., (1980).'The role of Prosopis cineraria in agropastoral systems in Western Rajasthan'. Intern. Symp. on browse in Africa, 1980 International Livestock Centre for Africa, Addis Ababa.
Portères, R. (1957). 'Un arbre vivant à contre saison en Afrique soudano-zambézienne (Faidherbia albida)'. Sci. Nat. 19, 19–24.
Radwanaki, S.A., Wickens G.E. (1967). 'The ecology of Acacia on mantle soils, in Zalingel, Jebel Marra, Sudan'. J. Appl. Ecol. 4, 569–579.
Ross, J.H. (1966). 'Acacia albida Del. in Africa'. Bot. Soc. Brot. XL, 187–205.
Speedding, C.R.W. Walsingham J.M. (1978). 'Energy and the future of agriculture'. New Zealand Agric. Sci., 12, (3): 76–80.
Seif el Din, A.G. (1969). 'The natural regeneration of Acacia senegal'. MSc. thesis, Sudan Univ. of Khartoum.
Schoch P.G. (1966). 'Influence sur l' évaporation potentielle d'une strate arborée au sénégal et conséquences agronomiques'. Agron. Trop., 21, 11, 1283–1390.
Trochain J.L. (1969). 'Le rythme biologique aberrant de Faidherbia albida (Mimosaceae)'. Trav. Lab. Bot. et Biogerogr. Faculté des Sciences, Toulouse, 6, 3, 7–13.
Vassal J. (1967). 'La plantule d'Acacia albida Del (Faidherbia albida) Del. A. Chev.)'. Bull. Soc. Sci. Nat. Toulouse, 103, 3–4, 583–589.
Wickens G.E. (1969). 'A study of Acacia albida Del. (Mimosoideae)'.Kew Bull., 23, 181–202.
