Version classiqueVersion mobile
OpenEdition Books

Le projet majeur africain de la Grande Muraille Verte

Abdoulaye Dia
Robin Duponnois

Partie I. Espèces végétales de la Grande Muraille Verte

Keep Africa Covered Trees-based options to combat land degradation in the Sudano Sahel: A Contribution to the “Great Green Wall” of the Sahel Program

D. Pasternak, A. Nikiema, E. Haglund et J. Ndjeunga


The paper is a contribution from the crops and systems diversification team at the International Crops Research Institute for the Semi Arid Tropics (ICRISAT-Niger). It starts by describing in brief the major causes for land degradation in the Sudano Sahel. The causes are divided into natural and human induced causes. Natural causes are weathering of the soil, droughts and torrential rains. Human induced causes are the result of a crop-livestock production system that exposes the land to wind and water erosion, trees prospecting for firewood, and destruction of young trees sprouts by grazing animals and by bush fires. Wind and water erosion are the main reasons for land degradation.
The paper presents six potential trees systems suitable for the Sudano-Sahel: Traditional agroforestry (parklands), farmers managed natural regeneration (FMNR), Acacia senegal plantations, Sahelian Eco-Farm, Bio-reclamation of Degraded Lands and dryland fruit trees plantations. It gives the advantages and constraints of each system and conditions for success. It is recommended concentrating efforts on agricultural lands where most land degradation (desertification) is taking place. The FMNR system seems to be the one to yield best results with minimal investment.

Texte intégral


1The Sudano Sahel is the region south of the Sahara delineated by the 300-900 rain Isohyets. It stretches from Senegal by the Atlantic Ocean to Eritrea by the Red Sea. It covers an area of about 12 million km2. About 60% of the land of this region is classified as degraded land (Lal, 1988). ICRISAT Niger studied over a period of 20 years the drives for desertification in the Sudano Sahel. These studies served as the basis for the development and improvement of trees-based systems to both stop the land degradation process and to reclaim highly degraded lands.

Causes of land degradation in the Sudano Sahel

2Two major causes for land degradation (desertification) in the Sudano Sahel: Natural and human induced.

Natural causes

3With the exception of Ethiopia’s highlands, the Sudano Sahel is geologically old and its soil is highly weathered. Over the ages tropical and desert climate alternated. When the tropics prevailed acid rains resulted in leaching of soluble bases, phosphorus and nitrogen compounds. Silica was removed and iron and aluminumrich duricrasts were formed. These are the lateritic soils typical to the tropics. (Lal, 1988). When the tropics retreated the laterite was covered by sand. Both the sandy soils and the laterite have a low pH (4.0-5.5) and are aluminum saturated resulting in aluminum toxicity in many crop species (Woomer et al, 1994). Droughts are another ‘natural cause” for desertification. The severe droughts that periodically occur in the Sahel result in the death of many perennial plant species thus indirectly bringing about land degradation. The frequent very high rain intensity (up to 200 mm/hr) results in run off and flush floods that erode the soil. This is yet another natural cause for desertification.

Human induced causes

4The Global Assessment of Soil Degradation (GLASOD) clearly demonstrate that most of the land degradation in Africa occurs in agricultural lands (65%), less in the permanent pasture land (31%) and only 19% in the forest and woodland (Sivakumar and Wills 1995).

5In the Sudano Sahel agriculture is practiced in the 400-900 mm rainfall region. Efforts to combat desertification should therefore be concentrated in this belt.

6The agriculture system of the Sudano Sahel can be described in general as an agrosylvopastoral system (crops + trees + livestock). Rain-fed annual crops (millet, sorghum, cowpeas and groundnuts) are produced during the “rainy season”. Trees are kept in farmland to provide products such as fruits, vegetables wood and fodder while livestock graze crops residues exposing the predominantly sandy soil to erosion by wind.

7Livestock graze young trees and seedlings preventing the establishment of a tree cover. Woodcutting for firewood particularly in the vicinity of cities is also a big threat to the environment.

8Dry season bush fires that in most cases are initiated by farmers also destroy trees seedlings.

9Overexploitation of the land for crops production without replenishing nutrients through the use of fertilizer or manure (the so called “land mining”) is also a major driver for land degradation (Schlecht et al 2006)

Effects of wind and water erosion in agricultural lands

10In the dry season the “Harmatan” blows with speeds up to 70 km/hr. Rainy season sand storms are even more severe. Sterk et al. (1996) measured during a three months period soil and nutrient loss due to wind erosion. During this period a total of 45.9 tons/ha of topsoil was lost. The amount of nutrients that were removed equals the recommended rates of fertilizer application for a millet crop in the Sahel.

11The main causes for water-induced soil erosion are by order of importance lack of vegetation cover, cultural practices and slope management (Roose, 1989). Bationo et al. (1996) summarized measurements of water induced soil erosion in various regions of the Sudano Sahel. The rate of soil loss was a function of water runoff intensity, slope and ground cover. For example in Ouagadougou with an annual rainfall of 850 mm a slope of 0.5% and a runoff rate of 40.6% soil loss in a sorghum field was 10.2 tons/ha.

12Both wind and water erosion eventually removes the sand cover exposing the laterite layer. As long as the laterite is covered by sand or when it is protected by vegetation, it is a soft and permeable soil with better properties (water and cation holding capacities) than the sandy soil. When alternately exposed to sun and rain the laterite hardens through aluminum-iron cementation into an impermeable crust that reduces water infiltration and turns the land useless for conventional agriculture exploitation. (Lal, 1988). Crusted laterites comprise most of the degraded lands of the Sudano-Sahel.

Measures to curb land degradation

13Wind and water erosion can be controlled either through physical barriers such as stone lines, earth bunding, stone terraces etc or through biological means such as mulch, hedges, wind breaks (shelterbelts) agroforestry and trees plantations (Reij et al.,1996). Perennial plants are more effective than annuals in nutrient recycling and in improving subsoil conditions. Trees protect the land from both wind and water erosion. Recently Pender and Ndjeunga (2008) conducted a thorough study on the economic, socio-economic and environmental impacts of various approaches for Sustainable Land Management (SLM) in Niger. They demonstrated that trees planting gave the highest economic returns among a range of land management options (area closure, micro-catchments, planting pits etc). The internal rate of return from tree planting ranged from 27-45%.

14This article will concentrate on trees as a measure for sustainable land management (desertification control) and for rehabilitation of degraded lands

Trees systems for the Sudano Sahel


15Agroforestry is the predominant agricultural production system in the Sudano Sahel, which allow farmers to grow trees together with annual crops. These trees provide various products; fruits, vegetables, firewood and fodder, which are essential for the livelihood of rural farmers. In a traditional agroforestry system farmers do not plant the trees. They rather allow the regeneration in their croplands of trees that they consider useful. Agroforestry in each agro-ecological zone has its own trees species composition.

16Species diversity in agroforestry systems is influenced by ecological and sociocultural environment. In the West Africa Sudano region the major trees components of the agroforestry parklands are Vitellaria paradoxa, Parkia biglobosa, Adansonia digitata, Tamarindus indica, and Faidherbia albida. The main tree components of agroforestry systems in the Sahelian agroforestry systems are Acacia senegal, Faidherbia albida, Sclerocarya birrea, Prosopis africana, Piliogstima reticulate, Ziziphus spina christi and Balanites aegyptiaca (Nikiema, 2005). Even though these species were selected by farmers mostly for their economic benefits they have a significant effect on maintenance of soil fertility and prevention of soil erosion (Sturm, 1997; Ouedraogo, 1995). Agroforestry parkland is a traditional agricultural land use system that is sustainable as long as the overexploitation by increasing population does not prevent restoration of soil fertility and vegetation cover. Table 1 that was compiled by Nikiema (2005) gives the main trees species present in the Sudano Sahel parklands. It ranks them according to their importance to farmer, the frequencyof their presence and the degree of threat for their extinction. Recommendations are given for steps to be taken to ensure sustainability.


17FMNR is an exciting new approach for tree regeneration in the Sudano Sahelian region. This technology is based on encouraging the natural regeneration of trees as an alternative to reforestation efforts using nursery-raised trees or direct seeding. The practice was introduced in the 1980’s by an NGO called SIM-Maradi (Rinaudo 2007) and its spread in the Maradi-Zinder region has been largely spontaneous, with minimal external assistance. The area covered today by trees from FMNR is estimated to be more than 3 million hectares (Larwanou et al, 2006). FMNR is based on pruning and protection of regenerated trees from seeds or from trees sprouts. ICRISAT with the assistance of Bioversity International and World Vision is carrying a survey in the Maradi zone to identify the drivers behind this exceptional rate of adoption of a technology. We are presenting here preliminary results of this survey. These results still need final verification.

Table 1

Table 1

Species presence, uses and needs for management actions in the agroforestry parklands: H= high; M= medium; L= low; -= not available; ● = Farmer’s managed natural regeneration; ○ = Planting and/or direct seeding; *= Tree management. (Source: Nikiema 2005 modified)

18For the purposes of this study, FMNR was defined as the practice of pruning and protecting naturally occurring seedlings and/or sprouts with the goal of encouraging the establishment of mature trees. The survey data indicate that 93 percent of households practice pruning, 27 percent practice protection, and 26 percent practice both (those we define as FMNR practitioners or adopters).

A Note on the Survey Methodology

  • 1 “Program” and “non-program” villages refer to villages where FMNR has been formally promoted by a p (...)
  • 2 Researchers familiar with this part of the world have found it problematic to perform these kinds o (...)

19The survey data was gathered during the month of April 2008, in Niger’s Maradi region where a number of development programs have promoted FMNR. Secondary data were used to stratify all villages in the region according to market access (greater or less than 6km from village) and agro-climatic zone (growing period of greater or less than 75 days). The sample of 41 villages to be surveyed was drawn in proportion to the number of villages in each stratum. Within each stratum, an equal number of program and non-program villages were selected at random1. Ten households2 per village were selected by random draw for a detailed interview. Overall, 410 households participated.

Initial Results

Tree Preferences

20The level of importance that farmers place on different tree species is assessed using a methodology adapted from Tschakert’s (2006) study of Sahelian farmers’views on climatic and other stressors. Respondents were asked to list the most important trees for the well being of their household among the trees found in their fields. After generating a list of species, respondents were asked to rank their choices from most to least important. For each species mentioned, an incidence index (I) was calculated, ranging from zero to one, representing the proportion of participants who identified that particular species. Additionally, and importance index (P) was calculated based on the order in which the species was ranked. The importance index of species s to respondent i is defined as:

21Where rs is the rank of the species and ni is the total number of species listed by that participant. The mean of Ps was then calculated for the subset of respondents who listed the particular species. These two indexes were combined into a joint, “value index”, V, using the equation:

22The ten most valuable species, in rank order of value index, are presented in the table below. The final column gives, in order of frequency reported, the most commonly reported used of each species.

Economic aspects

  • 3 The matching method and other specifications can greatly impact the final results. The results prod (...)

23The table below summarizes some preliminary results on the impact of adoption of FMNR on several livelihood outcomes. Two estimation techniques were used. Propensity Score Matching (PSM) is a technique that “matches” similar farmers based on observable characteristics and then makes one-to-one comparisons between adopters and their matched non-adopters3. Econometric estimation involves constructing a model specifying the effect of marginal changes in explanatory variables on the outcome variable of interest (in this case, the livelihood outcome, using FMNR adoption as one of a set of explanatory variables). A precise theoretical model must underlie econometric estimates for them to be considered as valid, a step that has not been taken here. These econometric estimates are merely intended as a very rough way of checking the PSM results for consistency. The results of the two techniques are not identical, but the table below demonstrates that they are generally consistent in sign and magnitude.

Table 2

Table 2

Value index, Incidence index and Importance index for most common trees species in farmer’s fields

24All figures below refer to the 2007-2008 agricultural season.

25From these preliminary results it appears that the FMNR has a significant positive economic impact for the practicing households (higher yields, higher income), this probably being the major driver for the adoption of this practice

Table 3

Table 3

Impact of FMNR on a number of economic indicators at the household level


26Acacia senegal is a drought tolerant tree of the family Mimosaceae native to the Sahelian and Sudano-Sahelian regions of Africa. In many Sahelian countries A. Senegal plantations is initiated by programs and projects for reclamation of degraded land and for generation of income from gum arabic production. Gum arabic is defined as the dried exudates from the trunk and branches of Acacia senegal, A. leata or Acacia seyal. The gum of the latter species is of lower quality than the gum of A. senegal and is collected from wild plants only. Gum is produced only under the stressful conditions typical to the dry season of the Sahel. Thus the Sahel has a competitive advantage for the production of gum Arabic over other regions of Africa. The main producers of gum arabic are Sudan, Chad, Nigeria and Niger but the potential area of production covers the entire Sahelian belt.

27Currently A. senegal is the major plantation tree in the Sahel. It takes between 3-5 years from planting Acacia trees to gum production. Production peaks after 15 years and remains stable for 10 additional years. Coppicing the trees at this latter age will result in the rejuvenation of the plantation. The International market for gum Arabic is estimated at 80,000 metric ton per year. This means that significant expansion of the area planted with Acacia senegal must be accompanied by a significant growth in gum arabic demand. Gum arabic exports are growing at a pace of 5% a year.

28At an average value of $2,000/ton this gives a total international market of $160 million. Most of the income from gum Arabic goes into the hands of traders and processors. Producers can get as little as 50 FCA per kg (Macrae and Merlin, 2002). There is huge variability in gum production among various provenances. The accepted annual gum yield for a Sahelian tree is about 150 g. However yields up to 4 kg/tree have been reported in Chad (Macrae and Merlin, 2002).

29Nikiema and Pasternak (2008) compared gum yields from 10 years old local trees with that of 10 years old trees of a Sudanese provenance and found out that the local trees produced on the average 50 g/y of gum whereas average yield of the Sudanese provenance was 400 g/y.

30The current low gum yields in the Sahel and the low prices received by producers do not justify planting commercial plantations of A. senegal.

31A typical plantation with a density of 400 tress/ha gives after 7 years from planting an average yield of 50 kg. At a maximum value of 400 FCFA/kg, the total value from one-hectare plantation will be only 20,000 FCFA that is about half the value for an average yield of millet. A. senegal plantations have additional benefits such as firewood, improved soil fertility resulting in increased forage, and feed value from the Acacia seeds and pods.

32ICRISAT-Niger started on 2003 a long-term research project aiming at increasing gum production leading to the improvement of the profitability of A. senegal plantations.

Selection of superior germplasm

33About 200 ten-years old A. senegal trees from Sudan were randomly selected in a commercial plantation near Niamey. Gum yield was sampled from these trees during four consecutive years. Results are presented in Figure 1. Four years average yield of 9 elite trees ranged from 500 to 900 grams compared wit 350 g/tree-the average yield of the whole plantation.

34Scions taken from the plus trees were grafted on local A. senegal seedlings and grafted trees were planted in 2006 at ICRISAT station for yield evaluation. Gum arabic harvesting from the grafted plants started in 2009 three years after planting.

Figure 1

Figure 1

Gum Arabic yield from 9 elite trees in a commercial plantation compared with the average of the 300 sampled trees (from Nikiema and Pasternak, 2008)

Effect of Ethylene

35Bhatt and Ram (1990) demonstrated that application of Ethylene in the form of Ethephon markedly increased gum arabic yields of A. senegal trees.

36Current trials using local Acacia trees (Nikiema and Pasternak, 2008) show that Ethephon can more than double gum yields of local provenances (Fig.2).

Figure 2

Figure 2

Effect of three concentrations of Ethephon on Gum Arabic yields (grams) over a two years’ period

Figure 3

Figure 3

Relationship between number of wounds inflicted on A. senegal trees and gum arabic yield

37In spite of the fact that A. senegal is an old tree crop there is not enough information on best management practices for this tree in a plantation. One of the important questions is the effect of the number of wounds per trees on gum arabic yield.

38As it can be seen in Figure 3, there was a liner correlation between the number of wounds inflicted on 10 m years old trees and gum arabic yield The above work indicates that there is good scope for significantly increasing gum yield in A. senegal plantations. It appears that a yield of 400 kg/ha is attainable. This should give an annual revenue of 160,000 FCFA/ha. If and when and this goal will be reached A. senegal plantations will become a feasible undertaking.


39ICRISAT-Niger is developing new tree-based agriculture production systems with the objective of arresting land degradation while increasing household income. One of these systems is called the Sahelian Eco-Farm (SEF).

40The SEF is a rain-fed trees-crops production system. Two models were developed: Model 1 comprising the domesticated Ziziphus mauritiana called “Pomme du Sahel” planted inside micro-catchments (demi lunes) and intercropped with high value annuals. In the Sahel these annuals are dual-purpose cowpeas and watermelons.

41Model 2 comprising of hedges of Acacia colei and of Pomme du Sahel intercropped with traditional crops such as pearl millet and dual-purpose cowpeas. The objectives of the A. colei hedges are to reduce wind erosion, increase soil fertility and provide renewable firewood.

42After six years of research, an analysis was carried out to compare the economics of the two systems with that of a traditional pearl millet-cowpeas system. The results are given in Fig. 4.

43The Net Present Value of model 2 over a time horizon of 50 years is not different than that of a traditional millet-cowpeas system. However model 1, because of the incorporation of high value crops, is much more profitable than traditional systems and its adoption should be encouraged.

Figure 4

Figure 4

Net present value of the two SEF models compared with traditional pearl millet-cowpeas system


44Crusted laterites occupy much of the landscape of the Sudano-Sahel. Productivity of these degraded lands is very low because the crust prevents water infiltration. This land is used for grazing and for firewood harvesting. The degraded laterites are potentially more fertile than the prevailing sandy soils and also hold more water than the sand.

45One can distinguish two types of laterites: Hard laterites and soft laterites. The hard laterites are difficult to cultivate. These can be used for trees plantations. Soft laterites should be selected for the BDL.

46Indigenous technologies such as planting pits (called Tassa in Niger or Zaï holes in Burkina Faso), micro-catchments called demi-lunes and trenches that harvest run-off water have been developed over the years and are slowly expanding in Burkina Faso and in Niger (Reij et al 1996). The Zaï hole is the most popular of these technologies. Tender and Ndjeunga (2008) found out that the water harvesting technologies used in Niger have a low internal rate of return probably due to the high cost-benefit ratio. Digging of zaï holes and preparing demi-lunes are labor consuming and the monetary return received from the coarse grains is relatively low.

47In the BDL these structures are used to grow high value crops that give a high monetary return. The two main trees planted in the demi-lunes are Pomme du Sahel and Moringa stenopetala. This Moringa originates in Ethiopia and can grow and produce when planted in the demi-Lunes without the need of supplementary irrigation. The main vegetable species that are intercropped in zaï holes among the trees are okra and roselle. Additional species such as pigeon pea, sesame and cassava are being tested.

48The BDL gives a good opportunity for Sahelian women to own land. In this region women usually do not have ownership rights for croplands but can receive from the village authority patches of degraded lands.

49A 3-5 hectares plot of degraded land at the vicinity of the village is identified and fenced. Soil is scarified and the demi-lunes, trenches and zaï holes are constructed and planted. Each woman cultivates her individual plot (about 300 m2) inside the BDL complex.

50The BDL is still under development. Only data for annual crop yields are available. Okra yields from women fields were 1000 kg/ha. At 200 FCF per kg of okra the revenue from the okra can be as high as 200,000 FCFA /ha. It is estimated that when Moringa and PDS will start bearing fruit the revenue per hectare could amount to 400,000Fcfa.

51This system has high potential for dissemination because of the relatively low set up cost. It is easy to manage and it provides women a source of livelihood.


52In the savannah and the Sudano regions of Africa one can find large areas covered with two fruit trees crops; cashew (Anacardium occidentalis) and mango (Mangifera indica). The nuts of the cashew are being processed and exported in increasing quantities. In the region of Bobo Dioulassou in Burkina Faso a new fruit-juice company called Dafani is purchasing all the mango fruit produced in the area for juice production.

53Dryland fruit trees plantations can also be established in the Sudano Sahel region using hardy fruiting tree species. Their fruit could be processed into juices, nuts and oils. The three most promising species are: Tamarind (Tamarindus indica), marula (Sclerocarya birrea) and Pomme du Sahel (Ziziphus mauritiana).

54Tamarind is a large tree belonging to the Caesalpiniaceae family. Its pods are used for juices and sauces seeds are used for gum production. Tamarindus juice is very popular in many markets of the world. The wild tamarind has a sour fruit. However the sweet tamarind varieties selected in Thailand are superior in taste and apparently also in yield.

55Sclerocarya birrea is a wild tree of Africa semi arid tropics. It is very drought tolerant and grows in regions of 300 mm/y rainfall. The juice is very tasty and rich in Vitamin C and in antioxidants. The kernels have high oil content (about 50%). This oil is used in cosmetics. The kernels are eaten as nuts.

56The domesticated Ziziphus mauritiana called “Pomme du Sahel” can grow in regions of 500 mm rain/y and above without irrigation. A mature dryland tree can produce about 15 kg of fruit per year. The juice of PDS is tasty and high in vitamin C.

57The fruit of the above tree species can be processed into juices, dried fruit, oil and nuts.

58If a viable industry to utilize the products of these trees is put in place than fruit tree species could be planted over a large area of the Sudano-Sahel.

Discussion and conclusions

59Massive tree planting or tree regeneration should have a very significant impact in combating desertification in the Sudano Sahel.

60We will try here to lay down some principles to improve the success of tree planting in the Sudano-Sahel region.

61Increased tree population should be implemented mostly in the agriculture production area of the Sudano Sahel (rainfall from 400-900 mm/y) because agriculture practices are the major drive for desertification.

62Tree planting should be done in a diversified manner. Either as components of traditional agroforestry systems, as pure plantations of high value trees, and as components of new trees-crops system

63The practice of community plantations should be reconsidered. The failure or only part success of many past trees plantations was due to communal ownership, taking away the responsibility from the individual for tree maintenance and protection

64The Farmers Managed Natural Regeneration (FMNR) process that is spreading in eastern Niger is perhaps the easiest, cheapest and most sustainable option for augmentation of the tree population in the Sudano Sahel.

65The traditional agroforestry systems of the Sudano Sahel can be improved through grafting of superior germplasm into existing trees, by increasing the number of trees in the system and by adding additional high value trees.

66Sufficient research resources should be allotted to help the finalization of the current research with Acacia senegal aiming at the increase in gum arabic yield per tree and the identification of best practices for gum tapping. This research can be completed in a short period of time. If results will be as expected than a massive drive for the establishment of profitable A. senegal plantations should be undertaken. Additional effort should be given to solving the problem of the current high transaction costs in the gum arabic trade.

67Pilot programs for testing the new integrated trees-crops system (SEF and BDL) developed at ICRISAT should be undertaken. The BDL in particular is a very exciting approach for combating desertification because it integrates land reclamation with women empowerment, improved nutrition and increasing income.


Bationo, A., Buekert, M. P., Smaling, E. M. A. and Visker, C. 1996.

Technologies for restoring soil fertility. In A. Uzo Mokwunye, A. de Jager and E.M. A. Smaling (eds). Restoring and Maintaining the Productivity of West African Soils: Key to Sustainable Development>IFDC-Africa, LEI-DLO and SC-DLO.

Miscellaneous Fertilizer Studies No. 14. International Fertilizer Development Center.

Bhatt-J. R; H.Y.M. Ram 1990. Ethephon-induced gum production in Acacia senegal and its potential value in the semi-arid regions of India. Current-Science 59: 1247-1250

Caliendo, M. and S. Kopeinig. 2005. Some Practical Guidance for the Implementation of Propensity Score Matching. IZA Discussion Paper No. 1588.

FAO. 2003. Advancing assisted natural regeneration (ANR) in Asia and the Pacific. Accessed online at <> on December 1st, 2008.

Lal, R. 1988. Soil degradation and the future of agriculture in sub-Saharan Africa J. Soil and Water Conservation 43: 445-451

Larwanou, M., M. Abdoulaye, C. Reij. 2006. Etude de la Régénération Naturelle Assistée dans la Région de Zinder (Niger). USAID/EGAT and IRG, 48 p.

Macrae, J and Merlin, J. 2002. The Prospects and Constraints of Development of Gum Arabic in Sub-Saharan Africa. Report submitted to World Bank

Nikiema, A. 2005. Agroforestry parkland species diversity: uses and management in semi-arid West Africa (Burkina Faso). PhD thesis, Wageningen University. 102 pp.

Nikiema, A. and Pasternak, D. 2008. Intensification of Acacia senegal plantation in the Sahel. Annual Report submitted to the International Foundation. ICRISAT, Patancheru, A.P.

Ouedraogo, S.J. 1995. Les parcs agroforestiers au Burkina Faso. Rapport AFRENA/SALWA No 79, ICRAF, Nairobi, Kenya. 76 pp.

Reij, C. Scoones, I. and Toulmin, C. 1996. Sustaining the Soil. Indigenous Soil and Water Conservation in Africa. Earthscan Publications Ltd. UK. 260 pp.

Rinaudo, T. 2007. The Development of Farmer Managed Natural Regeneration. Lisa Magazine, 23(2), Jeune 2007, pp. 32-34.

Roose, E. Gestion conservatoire des eaux et la fertilité des sols dans les paysages Soudano-sahéliens de l’Afrique Occidentale. ICRISAT, Soil, Crop and Water Management systems for Rainfed Agriculture in the Sudano-Sahel zone> Proceedings of an International Workshop, 1-6 Jan.1987. ICRISAT Sahelian Center, Niamey, Niger, A. P. 502324, India

Schlecht, E., Buekert, A., Tielkes, E. and Bationo, A. 2006. A critical analysis of challenges and opportunities for soil fertility restoration in Sudano-Sahelian West Africa. Nutrient Cycling in Agroecosystems 76: 109-136

Sivakumar, M. V. K. and Wills, J. B. (eds.) 1995. Combating Land Degradation in sub-Saharan Africa: Summary proceedings of the International Planning Workshop for a Desert Margins Initiative.

Sterk, G., Hermann, L. and Bationo A. 1996. Wind-blown nutrient transport and soil productivity changes in southwest Niger. Land Degradation and Development 7: 325-335.

Sturn, H. J. 1997. The importance of agriculture parks for soil fertility management. In, G. Renard, A. Neef, K. Becker and M. von Oppen (eds.) Soil Fertility Management in West Africa
Land Use systems. Mrgraf Verlag, Weikersheim, Germany. pp. 293-298.

Tender, J. and Ndjeunga, J. 2008. Impact of Sustainable Land Management Programs on Land Management and Poverty in Niger (69 pp). Submitted to the World Bank TerrAfrica and the PAC programs.

Tschakert, P. 2006. Views from the Vulnerable: Understanding Climatic and other Stressors in the Sahel. Global Environmental Change, vol. 17 : 381-396.

Woomer, P. L., Martin, A., Albrecht, A., Resck, D. V. S. and Sharpenseel, H. W. 1994. The importance and management of soil organic matter in the tropics. In. P.L. Woomer and M.J. Swift (eds) The Biological Management of Tropical Soil Fertility. John Willey and Sons.


1 “Program” and “non-program” villages refer to villages where FMNR has been formally promoted by a program or project.

2 Researchers familiar with this part of the world have found it problematic to perform these kinds of analyses at the level of the “household”, given the complexities of polygamous and broadly defined families. The “production unit” is an often-used alternative, defined as family members who work together in the same fields. The distinction between this and other common definitions of the household (e.g., that used by the FAO, 2003) is that the production unit is defined by the sharing of productive resources rather than the sharing of consumption good. Throughout this discussion, household and production unit should be considered as synonymous.

3 The matching method and other specifications can greatly impact the final results. The results produced here used a radius matching method and a “minima-maxima” comparison to impose common support. See Caliendo and Kopeinig (2005) for more on PSM.

Table des illustrations

Titre Table 1
Légende Species presence, uses and needs for management actions in the agroforestry parklands: H= high; M= medium; L= low; -= not available; ● = Farmer’s managed natural regeneration; ○ = Planting and/or direct seeding; * = Tree management. (Source: Nikiema 2005 modified)
Fichier image/jpeg, 236k
Titre Table 2
Légende Value index, Incidence index and Importance index for most common trees species in farmer’s fields
Fichier image/jpeg, 264k
Titre Table 3
Légende Impact of FMNR on a number of economic indicators at the household level
Fichier image/jpeg, 180k
Titre Figure 1
Légende Gum Arabic yield from 9 elite trees in a commercial plantation compared with the average of the 300 sampled trees (from Nikiema and Pasternak, 2008)
Fichier image/jpeg, 512k
Titre Figure 2
Légende Effect of three concentrations of Ethephon on Gum Arabic yields (grams) over a two years’ period
Fichier image/jpeg, 40k
Titre Figure 3
Légende Relationship between number of wounds inflicted on A. senegal trees and gum arabic yield
Fichier image/jpeg, 36k
Titre Figure 4
Légende Net present value of the two SEF models compared with traditional pearl millet-cowpeas system
Fichier image/jpeg, 79k


International Crops Research Institute for the Semi Arid Tropics (ICRISAT-Niger)

International Crops Research Institute for the Semi Arid Tropics (ICRISAT-Niger)

International Crops Research Institute for the Semi Arid Tropics (ICRISAT-Niger)

International Crops Research Institute for the Semi Arid Tropics (ICRISAT-Niger)

© IRD Éditions, 2010

Conditions d’utilisation :