Spatial crop growth variability in Western Niger: the knowledge of farmers and researchers

J.P.A. Lamers
P.R. Feil
A. Buerkert

For centuries, communities in the West African Sahel have practiced agriculture which is based on the transfer of cropping knowledge and experience from generation to generation. In the past decade, researchers have studied soil microvariability and subsequent crop growth differences in the major staple, pearl millet (Pennisetum glaucum [L.] R. Br.). A household survey was carried out in Western Niger together with an extensive literature search of baseline studies in the region, in order to determine whether indigenous knowledge related to crop growth variability can be combined with scientific knowledge with a view to sustaining soil fertility and millet production. This study illustrates the potential and limitations of local knowledge for the management of soil microvariability and millet growth in Niger.

Without the use of external inputs, millet yields in Western Niger, which range from 250 to 450 kg per ha, are strongly influenced by soil microvariability (Brouwer et al., 1993). This leads to patches of good and poor millet growth in close proximity to one another within a single field. Brouwer et al. (1993) define this soil microvariability as ... variability over relatively short distances (approximately 2-50 m), with differences in soil properties which affect hydrological behaviour and plant growth.... The causes of spatial variability within farmers' fields, and the current strategies employed by farmers to cope with this spatial variability are undocumented. Thus in presenting the viewpoint of farmers in Western Niger, the present study supplements on-station research into spatial crop growth and soil microvariability. The objective was to learn about farmers' knowledge and perceptions, and their strategies for managing crop growth and soil microdifferences. The study consisted of a combination of a field survey, conducted from June through October 1992, and a literature search of baseline studies carried out in the same region. The farmers' responses are discussed on the basis of the published literature.

The study was carried out in the villages of Sadore and Djiakindi, both in the vicinity of the ICRISAT research centre**1, 45 km southeast of Niamey. A total of 29 households were surveyed, all of which are engaged in a variety of activities, including agriculture, livestock, commerce, crafts and off-season labour. These activities take place at different times of the year and have long-, medium- or short-term time perspectives. The subsistence-oriented, rain-fed production systems are dominated by millet and, to a lesser extent, sorghum, both of which are usually intercropped with cowpea. Most households keep sheep, goats, and occasionally cattle, but the use of animal traction is rare. The mixed-farming systems, with an average farm size of 7-10 ha (McIntire et al., 1989:4-20), are characterized by limited family labour and low cash input. Most of the cropping budget is spent on hiring manpower for weeding. Additional cash income is earned through off-farm activities such as crafts, trade or temporary migration to neighbouring countries. During the dry season, hay from pulses and grasses is sometimes used to enrich the millet stover diet of the livestock.

Farmers' knowledge on spatial variability
In our study, farmers gave different explanations for spatial crop growth variability, which they believe is linked to differences in soil type. Colour and texture were the criteria which farmers said they used to differentiate between three types of soil: the reddish, sandy labu cheri, which is moderately fertile; the black labu bi, which is highly fertile, and the unfertile white labu kwarey (see table 1, columns 1 and 2). Of the 29 farmers interviewed, 69% attributed growth variations in millet to microtopographical features as well. These are small elevations and depressions in close proximity to one another, which farmers attribute to differences in soil fertility caused by wind erosion. The microhighs (or fandu, in the local Djerma language) are the fertile parts, which display a denser vegetation. The microlows (gorou) are less fertile and have a lower vegetation density, as was confirmed by a study carried out by Geiger and Manu (1993). According to farmers, crop growth variability is also related to trees and bushes. Guiera senegalensis and Annona senegalensis bushes were perceived as most effective in stimulating millet growth, whereas Balanites aegyptiaca and Sclerocarya birrea appeared to discourage millet growth. More than half (55%) of the farmers in our study made particular mention of the growth-enhancing features of Faidherbia albida trees, which are also well-known to researchers (e.g., Sall, 1992; Vandenbeldt and Williams, 1992). Differences in the dates of sowing, re- sowing, thinning and weeding also influence millet growth and development (see table 1, columns 1 and 2). This was confirmed by on-farm (McIntire et al., 1989) and on-station (Sivakumar, 1992) research.

The accuracy of farmers' comprehensive knowledge of agriculture, soil, animal husbandry and agroforestry has been confirmed by compartmentalized on-farm and on-station studies conducted in the same region (see table 1, column 3). And yet the explanations given by farmers and researchers were not always the same. For instance, farmers ascribed the 'F. albida effect' to manure deposited by animals which graze the palatable pods and litter underneath the trees. However, researchers found that although millet grain yields increased when F. albida windbreaks were planted, this increase could not have been due to manure, as this effect was excluded from the experiment (Michels, 1994). Sall (1992) found that underneath the F. albida the total soil carbon and nitrogen content increased by 40-100%, while biological activity was two to five times higher. He concludes that the higher yields are due to the increased biological activity caused by the litter, in spite of the fact that the nitrogen content of F. albida leaves is no higher than that of other species in Niger (Lamers et al., 1994). Vandenbeldt and Williams (1992), for their part, have shown that soil temperatures underneath F. albida are close to the optimum for millet growth and development, which allows the plants underneath to exploit the higher fertility attributed to this species. No change in the microclimate was mentioned by farmers.

¹Labu cheri (in the local Djerma language, labu means 'soil' or 'sand', and cheri means 'red') is a reddish sandy soil with moderate fertility. The labu bi (bi means balck) is a fertile soil with a higher clay content, while the labu kwarey (kwarey means white) is an infertile soil with a very low organic matter content.

¹ Labu cheri (in the local Djerma language, labu means 'soil' or 'sand', and cheri means 'red') is a reddish sandy soil with moderate fertility. The labu bi (bi means black) is a fertile soil with a higher clay content, while the labu kwarey (kwarey means white) is an infertile soil with a very low organic matter content. Farmers' strategies for dealing with spatial growth variability
Spatial growth variability is also the result of the management strategies employed by the farmers themselves to counteract both a decline in soil fertility and increases in soil degradation (see table 2, columns 1 and 2). Since farmers regard these as two interacting factors, they address them both by means of a single strategy; this consists in increasing soil fertility, primarily by the use of fallow and animal manure. However, the population growth has increased the pressure on land and it is therefore no longer feasible to allow entire fields to lie fallow. At present, farmers use manure for fertilization wherever possible, but this depends on their socioeconomic conditions. During our study, we were able to identify three groups: the rich cattle owners; the moderately wealthy farmers with small ruminants, who are able to procure manure; and the resource-poor farmers without livestock, who have to rely on crop residue management, within-field fallow, and trees and bushes. Farmers are forced to make use of the latter practices because of their limited resources, and the disadvantages of a broadcast soil mulch of crop residues, which can lead to increases in weed growth and extra labour needed for weeding. Farmers who have no other alternative than to use these traditional methods to restore soil fertility are faced with the continuing mining and degradation of their soils, in particular in the light of the predicted population growth. The increased demand for fallow and organic matter can no longer be satisfied by means of indigenous knowledge.

Although farmers are aware of the growth-stimulating effects of manure, they say that the quantities available are insufficient to manure all fields. The farmers lost most of their livestock in the Sahelian droughts of the early 1970's and 1980's, and stocks have not been rebuilt. Due to the shortage of manure, farmers have developed a sophisticated manuring system. In the first place, farmers manure specific fields, preferring household fields (cultivated by all household members) to individual fields (usually worked by individual members of the household). Not only the status of the field, but also the distance from the compound and the land tenure determine the choice of fields. This means that there are considerable differences in millet growth and development, not only between farmers but also between the various fields of the same farmer. Second, livestock keepers systematically shift the kraal where they are kept, particularly overnight, from one low-productivity spot to another. And third, some farmers distinguish between the quality of manure produced during the different seasons. The efficiency of these manure strategies has been confirmed by baseline studies focusing on isolated aspects (see table 2, column 3). Farmers know that although manure is an efficient way of increasing millet production, their desire to increase the number of livestock is detrimental to the range lands, which are already overgrazed (McIntire et al., 1992).

The strategy of selecting low-productivity spots for the application of high concentrations of organic material, such as millet crop residues, household waste and branches from trees and shrubs, is intended to stimulate the revegetation of eroded and crusty spots. Chase and Boudouresque (1987) have demonstrated that a mulch of branches improves such crusty spots, sometimes within the year. On the other hand, households in Niger have many alternative uses for crop residues, which limits the amount available for mulching (Lamers and Feil, 1993). Linear programming has shown that the concentrated spot application of mulch is more effective than broadcast crop residue applications, due to the higher labour productivity of the traditional techniques (Lamers et al., 1995). Even by moving compounds from one low- productivity spot to another every two or three years, farmers were able to increase millet production, as was recently demonstrated by an aerial survey (Buerkert et al., 1995a). It will be clear that such a strategy can only be employed on a limited scale, since compounds are small.

Farmers in Western Niger are aware of soil fertility management technologies such as within-field fallow; kraaling; spot applications of manure, crop residues and household waste; intentional relocation of their settlements; and making use of the microenvironmental differences between specific tree and shrub species. Although the efficiency of these strategies has been confirmed by research, their potential is limited, since farmers can only apply them on a very small scale. This is due to insufficient resources and both socioeconomic and institutional constraints; farmers may have neither cattle of their own nor land tenure, and may be unable to afford expensive inputs. Information is needed on how innovations that may complement the current methodologies fit into the complex household objectives and farming systems of subsistence farmers.

Researchers should be encouraged to consider not only isolated technical aspects but also farmers' resources and investment possibilities and the decision-making criteria involved in designing experiments. This case study shows that farmers' knowledge of soil types, soil fertility management and soil microvariability is based mainly on experience and observations. The analytical methods used by researchers could, therefore, supplement the farmers' indigenous knowledge and help them to make better use of their resources. But it is only in collaboration with politicians that it will be possible to develop sustainable farming systems for the Sahel.

Dr J.P.A. Lamers
University of Hohenheim,
Institute of Agricultural Economics and Social Sciences in the
Tropics (490)
70593 Stuttgart
Germany
Tel: +49-711-4593445
Fax: +49-711-4593762
E-mail: Lamers@rs1.rz.uni-hohenheim.de

Dr P.R. Feil
University of Hohenheim,
Institute for Rural Sociology, Agricultural Extension and Applied
Psychology (430)
70593 Stuttgart
Germany
Tel: +49-711-4593026
Fax: +49-711-4592652

Dr A. Buerkert
University of Hohenheim,
Institute of Plant Nutrition (330)
70593 Stuttgart
Germany
Tel: +49-711-4592344
Fax: +49-711-4593295
E-mail: buerkert@ruhaix1.rz.uni-hohenheim.de

References
Brouwer, J., L.K. Fussell and L. Herrmann (1993). 'Soil and crop growth micro-variability in the West African semi-arid tropics: a possible risk-reducing factor for subsistence farmers', Agriculture, Ecosystems and Environment 45:229-238.

Buerkert, A., F. Mahler and H. Marschner (1995a) 'Soil productivity management and plant growth in the Sahel: potential of an aerial monitoring technique', Plant and Soil (forthcoming).

Buerkert, A., R.D. Stern and H. Marschner (1995b). 'Soil variability and millet growth in the Sahel: From simple analysis of variance to ex-post blocking', Agronomy Journal (forthcoming).

Chase, R. and E. Boudouresque (1987) 'Methods to stimulate plant regrowth on bare Sahelian forest soils in the region of Niamey, Niger', Agriculture, Ecosystems and Environment 18:211-221.

Gavian, S. (1992) Manure and crop residue management strategies used by farmers in selected regions of the Tillaberi department, Niger. ILCA Niger Research Report 92/1.

Geiger, S.C. and A. Manu (1993) 'Soil surface characteristics and variability in the growth of millet in the plateau and valley region of Western Niger. Agriculture', Ecosystems and Environment 45:203-211.

Lamers, J.P.A. and P.R. Feil (1993) 'The many uses of millet crop residues', ILEIA Newsletter 9(2):15.

Lamers, J.P.A., K. Michels and R.J. Vandenbeldt (1994) 'Trees and windbreaks in the Sahel: Establishment, growth, nutritive and calorific values', Agroforestry Systems 26(3):171-184.

Lamers, J.P.A., M. Bruentrup and A. Buerkert (1995) 'The profitability of traditional and innovative mulching techniques of millet crop residues in West Africa', Agricultural Economics (forthcoming).

Michels, K. (1994) Wind erosion in the southern Sahelian zone: Extent, control and effects on millet production. Stuttgart: Ulrich Grauer Verlag.

McIntire, J., J. Hopkins, J.P.A. Lamers and L.K. Fussell (1989). The millet system of Western Niger I: Crop production. Addis Ababa: ILCA. (unpublished.)

McIntire J., D. Bourzat and P. Pingali (1992) Crop-livestock interaction in Sub-Saharan Africa. Washington D.C.: The World Bank, World Bank Regional and Sectoral Studies.

Powell, J.M. and F. Ikpe (1992) 'Fertilizer factories. Nutrient recycling through livestock'. ILEIA Newsletter 8(3):13-14.

Powell, J.M. and T.O. Williams (1993) Livestock, nutrient cycling and sustainable agriculture in the West African Sahel. IIED Gatekeepers Series No. SA37. London: IIED.

Sall, P.N. (1992) 'Site effects of Acacia albida Del.' pp. 103-105 in R.J. Vandenbeldt (ed) Faidherbia albida in West African semi-arid tropics. Proceedings of a workshop. Niamey/Nairobi: ICRISAT Sahelian Center/ICRAF.

Sivakumar, M.V.K. (1992) 'Climate change and implications for agriculture in Niger', Climatic Change 20:297-312.

Taylor-Powell, E. (1991) Integrated management of agricultural watersheds: land tenure and indigenous knowledge of soil and crop management. TropSoils bulletin 91-04. Texas: A&M University.

Vandenbeldt, R.J. and J.H. Williams (1992) 'The effect of soil surface temperature on the growth of millet in relation to the effect of Faidherbia albida trees', Agricultural and Forest Meteorology 60:93-100.

West, L., L.P. Wilding, J.K. Landeck and F.G. Calhoun (1984) Soil Survey of the ICRISAT Sahelian Center, Niger, West Africa. Texas: A&M University. Soil and Crop Sciences Department.

Endnotes
**1 ICRISAT stands for the International Crops Research Institute of the Semi-Arid Tropics, an international agricultural research centre operating under the umbrella of the Consultative Group of International Agricultural Research. The headquarters of ICRISAT is in India.