Indigenous knowledge of soil fertility management in southwest Niger

Geoderma 111 (2003) 457 – 479 www.elsevier.com/locate/geoderma

Indigenous knowledge of soil fertility management in southwest Niger Henny Osbahr a,*, Christie Allan b a

Tyndall Centre for Climate Change Research, School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ, UK b Department of Social and Life Sciences, University of Surrey Roehampton, West Hill, London SW15 3SN, UK

Abstract Agriculture in Niger is a complex and challenging operation. Farmers are faced with low-fertility sandy soils, variable rainfall, changing social and political situations and an unfavourable economic environment. Concerns about sustaining soil fertility have been voiced by agricultural scientists, who view agriculture as the maintenance of soil nutrient capital. Many of the biological and economic attributes of these farming systems have been documented, and are becoming better understood through the growing body of agroecological literature. Although agroecology recognises social, political and cultural interactions, much less is known about the specific details of farmers’ physical and biological knowledge and how this knowledge is used to make management decisions. Research in Fandou Be´ri village identified a local ethnopedological framework that had fundamental differences from scientific systems. Farmers’ defined soils according to location, potential for production and interaction with the wider ecological framework. They drew on varied ecological knowledge and experiences to make complex and dynamic farm decisions, as at other Sahelian villages. Local soil fertility management depends on individuals’ capabilities, perceptions of constraints and opportunities, and their ability to mediate access to different types of resources. This research suggests that there is a need to maximise the benefits of indigenous knowledge by integrating social and natural science, for example to help ‘precision farming’, and to use it to understand diversity. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Soil fertility; Ethnopedology; Indigenous knowledge; Niger; West Africa

1. Introduction The Republic of Niger is a land-locked country in the West-African Sahel. It is one of the world’s poorest countries in the UNDP indices on several accounts. The agricultural * Corresponding author. Tel.: +44-1603-593904; fax: +44-1603-593901. E-mail address: [email protected] (H. Osbahr). 0016-7061/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 1 6 - 7 0 6 1 ( 0 2 ) 0 0 2 7 7 - X

458

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

sector is the mainstay of the economy and farmers still rely largely on traditional farming methods. Unreliable rainfall and poor soils create an unpredictable and harsh environment. Niger suffered major droughts in the 1970s and 1980s and low rainfall years persisted into the mid-1990s. Economic restructuring in 1994 pushed up prices for farming products and cut subsidies for fertilisers; turbulent political and social conditions have created a sense of insecurity. However, despite these and similar constraints elsewhere in the Sahel, more people make a livelihood and with greater livestock densities than before the droughts in the 1970s. This suggests that there is much to be learned from a better understanding of Sahelian production systems. This paper compares local and scientific understandings of soils in one village in Niger. Firstly, Sahelian agriculture and the role of indigenous knowledge are outlined, and secondly, local soil frameworks are analysed. Finally, local perceptions of soil fertility change and the application of farmers’ agroecological knowledge to the management of soil fertility are discussed. 1.1. Sahelian agriculture Agriculture in dry Africa is undoubtedly suffering some serious environmental, socioeconomic and political problems (Benneh et al., 1996; Commins et al., 1996; Eswaran et al., 1997; Raynaut, 1997; Barbier, 2000). Changes in soil fertility have been associated with population growth, low inherent biological productivity, long-term rainfall decline, inadequate infrastructure, inappropriate agricultural policies, unsustainable intensification, international debt and unfavourable global market trends (Vierich and Stoop, 1990; Sivakumar, 1992; Spath and Francis, 1994; Reenberg and Paarup-Laursen, 1996; Reardon et al., 1997; Adedeji, 1999; Breman and Sissiko, 2000). Much concern has been raised over low and declining nutrient capital (Stoorvogel and Smaling, 1990; Smaling and Braun, 1996; Drechsel et al., 2001). Nutrient capital can be defined as the stock of nitrogen, phosphorus and other essential elements that becomes available to plants. All soils suffer depletion of that capital when brought into cultivation as nutrients are removed in crops or lost through leaching, erosion, etc. Sandy low-fertility soils, common throughout the drylands, are inherently more prone to this problem (Buerkert and Hiernaux, 1998). Penning de Vries and Djiteye (1992) concluded that nutrients were limiting in most of the agricultural areas of West Africa that had average annual rainfalls of above 250 mm. However, the nutrient stocks of individual plots, within farms and village territories, can differ considerably from the average, depending on differences in soil texture, land use histories, rainfall variability and the individual capabilities of the farmers involved. The agronomic literature on soil fertility under such conditions in the Sahel is extensive. Fertility is related to productivity and nutrient capital, and must include reference to the chemical, biological and physical properties of the soil. It is also generally agreed that fertility must be defined as the ability of the land to produce and reproduce; its capacity to support the growth of plants over time, under given climate and other relevant conditions (Ingram, 1990). Agronomic indicators of soil fertility include organic matter content, microbial activity, soil air and water balance (which is related to soil depth, texture and structure), and the availability of the most important nutrients (Finck, 1995;

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

459

Lal, 2000). The main sources of plant-available phosphorus, which is generally the limiting nutrient in the Sahel (Bationo and Mokwunye, 1991; Mahamane et al., 1996), are from the weathering of soil minerals, the mineralisation of soil organic matter, fertiliser applications and organic inputs. However, this scientific approach to soil fertility management fails to include the complexity of agricultural practice. First, while technical appraisal does emphasise the role of rainfall and bioproductivity constraints, analysis tends to use ‘typical’ models of farmers’ responses, and, in the varied cultural and natural environment of the Sahel, it has been claimed that this is inadequate (Guyer and Peters, 1987; Vierich and Stoop, 1990). For example, negative nutrient balances do not always justify recommendations to farmers for the immediate use of mineral fertilisers. This is because farm households very often have short-term objectives that run counter to that advice, but which fit prevalent socioeconomic, ecological, public policy and international agricultural circumstances. Cases of these circumstances are well-documented (Sikana, 1993; McIntire and Powell, 1995; Scoones and Toulmin, 1999a). Second, scientific systems attempt to be generic in nature, whereas indigenous knowledge systems are specific, interactive and allow heterogeneity (Tabor, 1990; Davis, 1996). There needs to be greater integration between social and natural science discourse and farming practice. 1.2. Changes in agroecological thought and the role of farmer knowledge The notion of progressive decline in dryland agricultural production, which is prevalent in the agronomic community (Powell et al., 1999; Breman et al., 2001) is being challenged (Howorth and O’Keefe, 1999; Mazzucato and Niemeijer, 2000; Mortimore and Adams, 2001). Soil fertility is no longer merely seen as just a physical or environmental process. There is a greater appreciation of the multiplicity of situations and complex interactions between society and nature, making generalisations inappropriate (Raynaut, 1997). To deal with the emerging complexity, research approaches have evolved, shifting from farming systems research into production research and currently focusing the broad notion of sustainable production. While many hailed ‘Farmer First’ thinking as a step in the right direction, the ‘Populist’ strategies it promoted have now been said to fail to integrate fully with local systems, or to translate the socio-cultural and political economic dimensions of knowledge creation, transmission and use into a usable framework (Scoones and Thompson, 1993; Chambers, 1997; Neefjes, 2000). The challenge is to listen and learn from farmers’ own knowledge, and this is not easy. There is still a concern about the analytical dichotomy between scientific and indigenous knowledge systems in relation to West African farming (Richards, 1985; Leach and Fairhead, 2000). Simpson (1999) pointed out that while many of the biological and economic attributes of traditional farming systems have been documented (Brokensha et al., 1980; Warren et al., 1995), and are becoming better understood through the growing body of agroecological literature (Altieri, 1998), much less is known about the specific details of farmers’ physical and biological knowledge and how this knowledge is used in making key management decisions. Raynaut (1997) contended that a full understanding of the complexities demanded a closer engagement at the microscale than most studies had achieved.

460

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

Mortimore and Adams (1999) also believed that the major deficiency in the dominant international research discourse was its lack of concern for the local. Agroecological approaches to understanding local-level farming systems have had a large influence in the debate. Agroecology is a combination of agricultural science, ecology and elements of anthropology (Hecht, 1998). As well as the agroecology itself, it addresses the social system within which farmers work and legitimates farmers’ knowledge. The underlying analytic framework owes much to theoretical and practical attempts to integrate the numerous factors that affect agriculture, including indigenous ethnoecology (McCorkle, 1994; Gray, 1997; Rain, 1999; Mazzucato and Niemeijer, 2000). Using this approach, specific management decisions cannot be divorced from economy or culture and are embedded in this wholeness or integrity (Reenberg, 1996). Part of the agroecological approach has been greater attention to local soil knowledge (Tabor, 1990; Osunade, 1992; and as reflected in the papers in this collection). There have been several studies in the Sahel, which have examined ethnopedologies as part of wider investigations into local land management (Dialla, 1993; Reenberg, 1996; Hopkins et al., 1995; Baidu-Forson, 1999). Studies of the ethnopedologies of the Djerma (Taylor-Powell et al., 1991; Manu et al., 1991, 1996; de Groot, 1995; Allan, 1997) and Fulani (Lamers and Feil, 1995; Krogh and Paarup-Laursen, 1997) have found many common terms within the separate traditions (not common to both). Following a focus on indigenous knowledge, many researchers have shown the ability of smallholders to adapt (Harris, 1999; Ellis, 2000; Abdoulaye and LowenbergDeBoer, 2000; Mortimore and Adams, 2001), while others have demonstrated the importance of understanding the strategies of differentiated social actors in soil management (Osbahr, 2001; Scoones, 2001; Scoones and Wolmer, 2002). The themes from new agroecological thought, in relation to soil fertility management and local ethnopedological knowledge, are explored further using the following Sahelian village as a case study.

2. Fandou Be´ri The case study is of Fandou Be´ri (13j31.8039N, 2j33.4486E) (Fig. 1), already the focus of various scientific and development projects.1 It is a small (population of about 750– 800), but regionally important settlement in the hinterland of Niamey in 1 1996 – 1998 Project Social and Environmental Relations in Dryland Agriculture (SERIDA), a collaboration between University College London, the International Institute for Crop Research in Semi-Arid Tropics (ICRISAT-Niger), the Institute of Hydrology and Brunel University (Batterbury et al., 1999; Warren et al., 2001a,b). Additional research has been carried out in the village on ecology, indigenous knowledge, hydrology, erosion and gender (Chappell, 1995; Longbottom, 1996; Osbahr, 1997; Allan, 1997; Piper, 1998; Matthews, 1998). The study village is in the IGPB-SALT site and ‘East-Central’ site of HAPEX-Sahel (The NASA Hydrologic – Atmospheric Pilot Experiment) (Goutorbe et al., 1997), and soil maps are available (Legger, 1993; Manu et al., 1991). The nearby village of Banizoumbou has been intensively monitored as part of the HAPEX and ORSTOM programmes for several years and, along with another nearby site, is currently being used by the international centres of ICRISAT and ILRI.

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

461

Fig. 1. Location of study village.

southwest Niger. The village territory is approximately 37 km2. The landscape is an open savanna parkland mosaic of farmland, degraded land, and rangeland. Most of the arable land is in a wide, shallow valley, filled with now stabilised Late-Pleistocene dune sands, which thin out northward onto an uncultivable, ferricrete plateau, used only for grazing and wood fuel. The sandy soils of the valley are poor and acidic, and as such, are susceptible to rapid leaching, P-fixation and contain free Al. Some of these sandy acid soils have complex patterns of erosion, exposed hardpans and are dissected with dry abandoned channels, in which there are soils with higher clay content. The rains come in a well-defined season from April to September, giving a mean annual rainfall of about 550 mm. Early-season storms bring high intensity rains and high winds, which cause erosion and consequent dust storms. There is an appreciable input to soils of far-travelled dust in the Harmattan season between November and February, and this appears to replace the calcium and potassium removed by crops, but not phosphorus (Warren et al., this collection). Villagers come from Djerma and Fulani ethnic groups and are mostly farmers who practice rainfed subsistence agriculture and rear small ruminants. The Fulani families farm loaned fields, but some remain almost wholly pastoral. The agricultural system is focused on cereals such as millet and sorghum that are intercropped with cowpea, groundnut and hibiscus in low-density patterns. Farmers have only basic equipment (using hoes rather than ploughs), rarely use mineral fertiliser because of its expense and apply only limited amounts of organic fertilisers, relying on short-fallow rotation (averaging 3 years) for nutrient replacement. Many rely on off-farm income to supplement their household cash flow, made easier by the good access to roads and the capital. There

462

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

are four primary constraints for the farming community: the unreliable rainfall (above all), low soil bioproductivity, limited capital and limited labour. Most farms are dependent on family labour, but farming suffers a shortage of labour, even at a population density of about 20 people per square km. Nevertheless, the area of cultivation increased from 11% in 1950 to 23% in 1992 (Batterbury et al., 1999). This kind of extensification is shared with villages elsewhere in the region (Heasley and Delehanty, 1996; Amissah-Arthur et al., 2000) and can be partly attributed to changes in tenure law (Lund, 1998). There have been changes in the crops grown. Cotton has not been grown since the droughts of the 1980s and farmers have had to look to the sale of the remaining crops for additional income. The loss of vegetation on common land and increased harvest of crop residues for household use appears to have been one of the factors behind the reduction in the numbers of herds passing through the village, once important sources of manure.

3. Methods Studying farmers’ knowledge of soil is methodologically demanding because of the high variability in ecological, cultural and socio-economic factors. The study used a multidisciplinary approach, incorporating Grounded Theory (Strauss and Corbin, 1990), participatory methods (Guijt and van Veldhuizen, 1998) and soil science to build an understanding of soil knowledge and management on 20 farms (households were selected through stratified proportional sampling). The research was conducted during the 1997 and 1998 agricultural seasons (April –August). The rainy season was chosen because migrant workers have returned to their farms by April and make the most farm management decisions during this time. Trained Djerma and Fulfude translators were used to conduct focus groups, transect walks, farm visits, qualitative interviews and semi-structured questionnaires (including the household heads, male village elders and women). Through a process of participatory review and repeat visits, information was cross-checked and organised at three scales: field, farm and village territory. 3.1. Developing local narrative through visualisation Visualisation techniques were used to develop soil maps of the village territory. Each member of the group was asked to indicate and describe the location and soil type of their current fields. After discussing the soil types found in the territory, the group was asked to draw a map showing the distribution of the soil types. Individuals and groups were presented with photographs and samples of the different soil types and asked to identify the soil type. This gave insights into the role of intrinsic physical characteristics and spatial locations in the classification framework. Farmers were asked about perceived changes in soils, soil management, and the meanings of the soil and land use terms recorded by previous studies (Taylor-Powell et al., 1991; de Groot, 1995). Individuals were also asked to create network diagrams to convey everyday experiences, common knowledge and practice by representing their individual perceptions of the choices and priorities that influence resource allocation and investment.

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

463

3.2. Methods for eliciting understandings of soil fertility Farmers were asked to rank the soil types by texture, colour and fertility and to rank their fields by fertility for 1996 and 1997. Fertility was ranked for both ‘good rain’ years and ‘poor rain’ years (locally defined). Since not all groups used the same number of soil types, rankings were scaled by dividing by the total number of soil types ranked. A modified proportional piling technique was used to quantify the fertility differences between soil classes. Farmers were asked to use the local boko measure (bundle of millet) to indicate comparative yields for the different soil classes and fields represented by identical circles. For comparison, the field with the most fertile soil (as identified by each group) was said to be producing 100 boko. 3.3. Comparative soil sampling and laboratory analysis Available ammonium and nitrate in the four locally defined soil types of five fallow fields was measured from nine duplicate soil samples in each site (10 g of each soil was shaken in 25 ml 1 M KCl for 1 h, the sediment allowed to settle and the supernatant decanted). The nutrient content was determined using standard methods described for ammonium (Rowland, 1983; Gentry and Willis, 1988), nitrate (Dorich and Nelson, 1983; Tel and Heseltine, 1990), phosphate (Bray and Kurtz, 1945) and organic matter content (Tiessen et al., 1991). Soil pH (H2O) was measured with a pH meter in a 1:1 solution. Grain size was measured using a set of soil sieves of 1000, 500, 250 and 45 Am mesh size.

4. Ethnopedological framework Table 1 gives the local classification for soils in Fandou Be´ri. The seven groups of farmers consistently divided the soils into two major soil classes, hard soils (locally known as botogo and gangani) and sandy soils (tassi). The hard soils were further divided by perceived fertility, the botogo soils being considered more fertile than the gangani. A fourth type of soil (korabanda) was identified by a characteristic black surface. Farmers’ opinions differed as to whether korabanda occurred only on hard or sandy soils (it was observed on both during the survey) and approximately one-third of farmers interviewed Table 1 The main soil types identified by farmers in Fandou Be´ri and their distinguishing characteristics (from: Allan, 1997) Soil name

Texture

Colour

Fertility

Botogo Gangani

dark brown red

most fertile but dependent on rain fertile but dependent on rain

Tassi

hard; compacted and heavy to work hard; highly compacted, needs animal traction for cultivation loose and sandy

red, white or black

Korabanda

skin on hard or sandy soil

black skin over any colour soil

least fertile but still productive in low rainfall years fertile in low and high rain years

464

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

did not describe korabanda as a separate soil type, whereas all groups described the other three soil types as distinctly different. This summary gives a conceptual framework. The division is primarily texture-based and has two exclusive classes. There is a secondary division in one of these classes based on fertility, this being characterised by a colour difference. Here again there are two exclusive classes. Thus, there are a total of three distinct classes of soil (tassi, botogo and gangani) and a floating class defined by a soil crust (korabanda) that can occur on any of the other three soils. 4.1. Comparing soil identification characteristics Despite giving texture and colour as important identifying characteristics, farmers were unable to distinguish each soil type by these characters alone. When presented with photographs of each soil type, most farmers could identify the korabanda soils from the distinctive black surface markings, but were unable to identify, or frequently wrongly identified the others; this was despite relatively good photographic reproduction. Farmers invariably asked where they had come from, suggesting the need to use a spatial reference as the primary method of soil identification. The mapping exercises were used both to elicit a range of information about the local ethnopedology and for triangulation, whereby the descriptions of the soil types and location of the fields farmed by each informant were compared to the soil distribution map drawn up by each informant at the start of the session. The high degree of agreement created a local soil distribution map (Fig. 2), although two fields of korabanda soil were located outside the ‘korabanda area’. These exercises highlight an important difference between how soils are identified and differentiated in the local ethnopedological framework and a scientific framework.

Fig. 2. Farmers’ map of soil types (from: Allan, 1997).

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

465

Indigenous soil knowledge contained a wealth of information, and direct comparison with scientific soil classifications is not always straightforward. In common with other ethnopedologies (Osunade, 1992), colour and texture were used as the primary characteristics. However, they could not be used to identify soil samples removed from their physical location with the history attached to that piece of land. In contrast, a soil scientist often identifies soils from samples far removed from their site and uses characteristics, including colour and texture, intrinsic to the soil samples to differentiate them. The scientific study of the soils in the territory, conducted during the NASA HAPEX-Sahel experiment (Legger, 1993), roughly correspond to the three zones described by the farmers. However, in the scientific survey they are categorised into units, such as Arenosols or Podzols, that can be separated out from context and still retain distinguishing characteristics. This is not the case with the local soil framework. 4.2. Comparing soil fertility The relative fertility of the different soil types was perceived by the farmers in terms of crop production. In a year with adequate and evenly distributed rainfall (a ‘good’ rain year), the groups consistently ranked hard soils (botogo and gangani) as more fertile than the sandy soils, and all but one group ranked botogo as more fertile than gangani (Table 2). The modified proportional piling exercise was used to produce values that were comparable for both rainfall regimes. The number (N) is given for each average value in Fig. 3. Notwithstanding, Fig. 3 does suggest a picture of the relative productivity of the soil types as perceived by the farmer groups. Fertility was related both to intrinsic soil properties and critically the amount of rainfall in a given year (Table 2 and Fig. 3). Few groups ranked the fertility of korabanda, but of those that did, half ranked it higher in fertility than gangani and half ranked it lower. This resulted in a similar average fertility rank for the two classes. In years when rainfall was low or poorly distributed (temporally or spatially) (a ‘bad’ rain year), tassi soils were given the highest fertility rank (although still only attaining 70% of hard soil fertility levels in ‘good’ rain years, Fig. 3), while botogo and gangani were equally low; notably, korabanda retained its relatively high position in the rankings in these years (Table 2 and Fig. 3). Gangani soils were believed to be unproductive in a ‘bad rain year’. The pattern of ranks for the sample fields in the years of survey most closely resembled the ‘good rain year’ ranking pattern. It is clear from the fertility rankings that soil fertility is not perceived as an intrinsic property of

Table 2 Average scaled soil rank (number of groups who ranked each type given in square brackets) (from: Allan, 1997) Soil types

Good rain year

Bad rain year

Sample fields

Botogo Gangani Korabanda Tassi

0.98 0.66 0.62 0.28

0.46 0.43 0.79 0.98

1.00 0.80 0.53 0.33

[7] [7] [5] [7]

[7] [7] [5] [7]

[7] [7] [7] [7]

466

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

Fig. 3. Comparative expected yields (average number of boko or bundles of millet estimated by each group)—N for each average is given in square brackets under each soil type in the form [N ‘good’ rain year, N ‘bad’ rain year] (from: Allan, 1997).

the soil, but rather as an interaction between a combination of soil characteristics (encompassing nutrient content and texture) and rainfall. Discussion with the groups provided qualitative evidence to support the relationships represented in Fig. 3. Farmers described a complex situation, one in which the productivity of any given field in any given year is difficult to predict with high spatial variability in rainfall. In previous studies of both Djerma and Fulani, rainfall was also found to be central to local perceptions of soil productivity (Manu et al., 1991; Krogh and Paarup-Laursen, 1997). Manu et al. (1991) reported that in ‘good’ rain years, crops would be sufficient regardless of soil nutrients, while in years of lower or more erratic rainfall, low soil fertility became the more important limiting factor. This study found that it was both the temporal and spatial (on a field-by-field scale) distribution of rainfall during the growing season that determined the relative productivity of the soils. It is the soil characteristics that affect productivity in relation to rainfall that are most important in the local division of soil types. The ethnopedological framework is evidently based around perceptions of how soil qualities interact with rainfall to affect productivity, and this suggests that farmers seek to manage ecological uncertainties to maintain yield even in ‘bad’ rain years. 4.3. Results of laboratory tests The levels of nutrients measured in all the soils were low and variation between soil samples within and between grids was high, making patterns of inter-soil type nutrient availability difficult to distinguish (Table 3). However, tassi soils did have discernibly lower nitrate values, while botogo, korabanda and gangani had similar values for ammonium. The highest values of available phosphate and nitrate were found in soil from korabanda. Inter-site patterns of soil organic matter content and pH did not correspond to observed patterns in nutrient availability, but grain size did differ slightly between the soil types (Table 3). Organic matter levels were extremely low in all soil types, not one site having levels exceeding 0.8%, and inter-soil type differences showed no clear pattern. Average soil pH was between 5.2 and 5.5. The soils were sandy with generally very low silt

Code

F13 B F7 2A F7 2B F7 4A F7 4B F1 A2 F1 A3 F5 A F5 B F8 2B

Soil type

Botogo Gangani Gangani Gangani Gangani Korabanda – tassi Korabanda – tassi Tassi Tassi Tassi

Fallow age (years)

NH4 – N (Ag/g) average (S.E.) (N = 9)

NO3 – N (Ag/g) average (S.E.) (N = 9)

PO4 – P (Ag/g) average (S.E.) (N = 9)

Organic matter (% dry weight) average (S.E.) (N = 9)

pH (H2O) average S.E. (N = 9)

4 2 2 4 4 6

7.7 8.2 7.2 7.5 9.4 9.1

2.1 2.3 1.5 1.4 3.5 5.3

2.4 3.6 2.5 3.8 2.7 10.9

0.1 0.1 0.5 0.7 0.3 0.2

5.4 5.5 5.2 5.4 5.4 5.4

6

7.8 (3.4)

1.5 (0.5)

3.6 (1.3)

1 1 2

6.5 (1.6) 6.8 (2.0) 5.4 (0.5)

1.7 (0.3) 3.1 (0.9) 1.9 (1.2)

4.0 (1.7) 2.9 (1.6) 4.2 (1.9)

(1.0) (2.3) (1.6) (1.0) (2.0) (1.6)

(0.9) (1.6) (0.8) (0.7) (1.5) (0.4)

(1.1) (1.2) (0.8) (1.9) (0.5) (2.1)

Grain size (% dry weight in each size class) < 0.25 mm

0.5 – 0.25 mm

>0.5 mm

(0.1) (0.1) (0.1) (0.1) (0.1) (0.1)

40.9 50.0 42.3 56.4 57.0 45.8

47.0 37.3 48.4 36.1 34.6 41.7

12.1 12.7 9.3 7.5 8.3 11.8

0.7 (0.1)

5.3 (0.1)

41.6

48.9

9.5

0.3 (0.3) 0.3 (0.2) 0.7 (0.1)

5.5 (0.1) 5.5 (0.1) 5.3 (0.1)

34.0 35.7 29.5

58.7 56.9 65.4

7.3 7.4 5.2

(0.1) (0.1) (0.1) (0.1) (0.2) (0.2)

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

Table 3 Results of scientific soil analysis (from: Allan, 1997)

467

468

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

contents. The hard (botogo and gangani) soils had a higher percentage of the finest grains ( < 0.25 mm), which included clay. The tassi soils (including korabanda on tassi) had a correspondingly higher percentage of the medium sized grains (0.5 –0.25 mm). Of all the factors measured, grain size had the closest correlation to local divisions of soil types, indicating the importance of water-holding capacity on soil productivity. Available moisture increases with increasing proportion of fines, sandy soil having the lowest water holding capacity. Thus, botogo and gangani would be expected to have the highest available moisture when there is sufficient rain. However, these soils also have the highest clay content. Although clay was not measured in this study, Taylor-Powell et al. (1991) recorded clay contents of 2– 8%, with the highest values in the soils near the plateau (i.e. botogo and gangani) and the lowest values in the valley soils (i.e. tassi). Even small differences in clay content can make a great difference to the hard-setting tendency, which affects workability, especially under hoe cultivation. It is possible that in bad rain years or during intense rainfall events, the rain falling on the higher altitude areas of botogo and gangani would run off towards the tassi areas. This has two effects: erosion and the transport of water towards the higher tassi soils. The farmers’ perceptions of fertility were based around an interaction between soil characteristics and climatic factors, which produced a distinctive pattern of soil fertility that the scientific soil tests used in this study, were unable to discern. A study that measured available nutrients over several entire growing seasons may better detect the patterns perceived by the farmers. However, since intrinsic soil characteristics composed only part of farmers’ perceptions of fertility, there may be no direct relationship between measurable soil characteristics and local fertility ratings. The farmers’ ethnopedological framework relates to a complex management system that incorporates social and cultural as well as environmental considerations and it is these factors in combination which dictate the productivity of the land, not soil characteristics alone.

5. Perceptions of change in soil fertility The majority of the farmers (80%) reported declining soil fertility and crop yields on sandy soils. They had noticed a lighter colour and coarser texture for soils near the village, indicating a decrease in organic matter. Farmers attributed these changes to the increase in the area of permanent cultivation, the decrease in the area of long-term fallow (over 3 years), the limited inputs of manure from livestock, and to low rainfall. Elders believed that there were fewer areas of black sandy soils (tassi biri), that is those soils with high organic matter content, and more areas of red or white sandy soils (tassi kirey and tassi kware), which have lower organic matter contents. This is consistent with previous studies in the village (Batterbury et al., 1999). Only 10% of the farmers felt that there had been a decline in fertility of the hard soils and all of them believed that the restricted areas of tombo had lost little fertility (these soils gain their fertility from inputs near or on household compounds). In contrast to the beliefs of the Djerma, the Fulani groups reported no such declines in soil fertility on their fields, which they attributed to their constantly high inputs. The process of declining soil fertility during cultivation was reflected in the naming of stages of the rotation system. Farmers describe how one piece of land

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

469

progresses from fertile cleared sacara in the first year of cultivation to lalibanda the year after, to kwar kwari (white sand with less organic matter) in the third season and finally to boulouga (degraded land) after 5 years. After cropping for 5 years, farmers observe that some of the field will need to be rested as fallow (farezenon), or nutrients have to be added or, if particularly degraded, abandoned to long-term bush until the farmer has the ability to add enough inputs. Under farezenon farmers noticed the development of darker soil colours and korobanda.

6. Using local knowledge to manage soil fertility Farmers hold complex ecological knowledge or berey (local knowledge), as the ethnopedological framework presented in this paper has shown. There is great variation in the kinds and depth of such knowledge. The choices made by individual farmers depends on their capacity to enact successful agricultural performances and to exploit an evolving range of opportunities, and these in turn are based on the use of their local knowledge, as well as on their communication and creative capacities. The knowledge, here as elsewhere (Pawluk et al., 1992; Stoller, 1989), becomes intertwined with the belief and value systems that underlie culture (for example, in the use of metaphoric linkages). On an individual basis, knowledge is acquired through personal experiences and overlapping communication pathways, both of which are influenced by social factors, including age, gender and family ties. Uncertainties arise because any one farmer’s knowledge is incomplete and incapable of dealing with risks outside of their partial perspective and this creates a plurality of perspectives (as discussed by Mehta et al., 1999). For example, an overlapping set of influences affects knowledge distribution and communication. One of this set is familial status, based on lineage and traditional divisions. Family ties can allow access to specialised spheres of knowledge and sources of income. The traditional belief system and its associated practices is not equally spread, and appears to be eroding in importance in many locations, especially amongst the young who chose to enter new professions. Network mapping can capture some of the farmers’ everyday experiences and ‘common’ knowledge (assumed knowledge held within a culture). Such diagrams identify the links that exist in soil fertility maintenance practices and the conceptual arrangement of knowledge or differences in management between individuals. The varying levels of complexity in the diagrams represent the opportunities and constraints for individuals. These opportunities and constraints are defined by the household’s livelihood portfolio or predominance of particular types of ‘capital assets’ (see Carney, 1998; Scoones, 1998; Bebbington, 1999). Hence, the conceptual importance attached to soil fertility investment for farmers with poor soils and little access to livestock or labour is defined by constraints, whereas for farmers with exogenous knowledge or with higher social status, it is defined by opportunities. Thus, each ‘map’ becomes a statement about the way each individual views farm management, but also a reflection of their status, power, skills, identity and fears. The example given in Fig. 4 reflects the social and financial ‘asset’ opportunities open to a Djerma farmer with status in the village (a marabout and old family lineage).

470

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

Fig. 4. Conceptual network diagram of soil fertility management (from: Osbahr, 2001).

Furthermore, at the local level, the inherent temporal and spatial heterogeneity of natural resources is paralleled by equally variable household productive capacity (including knowledge, family labour and social networks, livestock, tenure, cash, and ability to Table 4 Soil fertility practices used in 1998 (from: Osbahr, 2001) Soil fertility practice

1998 % Farmers, n = 20

Fallow: (a) Short (1 – 3 years) (b) Long (4 years+) Inputs from livestock: (a) Corralling or grazing (b) Manure transported from homestead Mineral fertilisers Nitrogen-fixing crops Ash from burning Protecting trees Laying of millet stalks Recycling and composting organic materials a

field.

% Fields, n = 58

% Tassia, n = 41

% Gangania, n=8

% Botogoa, n=9

85 80 30 80 55 65

48 38 10 34 19 24

60 50 10 38 25 30

13 13 0 13 0 13

30 0 30 30 10 30

0 100 40 80 100 90

0 72 19 69 100 84

0 85 23 75 100 90

0 38 0 50 100 70

0 40 20 80 100 52

Although the local soil type is variable with all fields, values are given for the primary soil type found in the

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

471

enact resource entitlements) (Osbahr, 2001). There are many constraints to productive capacity for farmers in Fandou Be´ri. Table 4 shows that only 21% of the survey households used labour intensive methods (such as compost or hand-distributed applications of inputs), while most farmers were dependent on fallow rotation and a variety of low labour intensive methods. In other circumstances, farmers relied on their local knowledge of soil characteristics and plant indicators to focus effort in ‘precision farming’ methods, in which they target their planting strategies and additions of nutrients to specific places and soil types, with the aim of maximising nutrient recycling (Osbahr, 1997). For example, farmers value fertile niches around Faidherbia albida, termite mounds, old livestock pens

Box 1. One farmer’s agricultural season in 1998 (from Osbahr, 2001). Amadou Abdoullaye2 is a middle-aged Djerma farmer. He had inherited three fields of different soil types but considered the largest, nearest to the homestead, to have low soil fertility because it had been cultivated continuously for over 50 years. Below-average annual income meant he relied on help from his family to run the farm. His family needed to produce an annual minimum of 330 boko for subsistence. However, in 1997 grasshopper swarms caused much of the farm to be abandoned, producing a harvest of 146 boko, and presenting him with constraints in essential resources, such as seed, manure and labour for 1998. Mr. Abdoullaye worked as a migrant farm labourer in the Ivory Coast from November to February to earn cash to buy seed. With the lack of early 1998 rains, experience told him that the sandy soils would be more productive and he decided to clear a 3-year fallow on his large tassi field and mulch and burn the other fields. He put his household animals on the closest field but carried manure to the distant gangani field because he had not been able to pay local Fulani for grazing (7000 F CFA for 2 months). He and his family planted the millet seed on the tassi parts of the large field 3 days after the first rains at the end of April. Unfortunately, he misjudged his labour management as the result of a conflict with neighbours and missed seeding after the next rains, forcing him to plant a fast growing cultivar. Different crops were planted on the different soils reflecting the fertility and moisture-retention capability (Fig. 5). He partitioned his mid-season labour, preferentially weeding the tassi plots. Improved late-season rains allowed him to redirect work to botogo and gangani areas, which were additionally able to support sorghum, sorrel and hibiscus (planted at the same time as the millet) and he planted peanut on these soils 15 days later. However, much of the farm’s botogo soils had not been prepared and he was restricted to cultivating only those parts ready. High levels of weed growth, including Striga varieties, developed with the better rains and several parts were abandoned. Overall, 1998 brought more sustained late-season rains but his limited labour resources made prioritising action for soil fertility management and farm tasks especially difficult, making him vulnerable to unpredictable rainfall patterns (Fig. 6). 2

Name has been changed.

472 H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

Fig. 5. Seeding pattern used in 1998 (from: Osbahr, 2001).

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

473

and target applications of green manures, livestock grazing and burning. The importance of these practices has been widely qualified in numerous scientific studies (Juo and Manu, 1996; Brouwer and Bouma, 1997; Payne et al., 1998; Hiernaux et al., 1999; Wezel et al., 2000; Esse et al., 2001). To work with environmental microdiversity, farmers have developed an understanding that allows them to manage dynamic social, economic and environmental resources, and this knowledge may often appear to run counter to the understandings of conventional agricultural science. Their perspective essentially incorporates the ephemerality of soil status, the historical dynamics of soil fertility and the need for soil fertility maintenance to reflect natural variability. In the context of annual crop production, the convergence of these physical and social forces is possibly best conceived as a dynamic performance (Richards, 1989). It is more accurately a set of linked performances, through which individuals respond to the evolving set of physical and social conditions within each year, and from one year to the next (Box 1, Figs. 5 and 6). The general nature of these performances, as summarised by Moris (1991), applies very well to the situation at Fandou Be´ri. Risk and uncertainty dominate farmers’ lives; they must respond to an unfolding scenario of opportunities and constraints (for example, wet or dry years, early or late onset of the rains, pest outbreaks in a particular crop, unplanned labour shortages, and so forth). Farmers are able to bring to their farming an accustomed ‘script’, modified as the season develops. This does not preclude strategic natural resource management plans and indeed the process in adaptive strategies operating at different scales and over different timeframes has been well documented (Reij et al., 1996). Thus, local soil fertility management is the result of a combination of responses to unforeseen events, planned management, land use history and the farmer’s own particular knowledge. These form the basis of a household’s strategy at a particular time, in a specific place, whereby local knowledge is employed in an array of methods to manage soil fertility depending on labour and capital.

Fig. 6. Household farm labour and rainfall in 1998 (from: Osbahr, 2001).

474

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

7. Conclusions Our findings show that the local framework for understanding soil types was part of a wider conceptualisation of the local ecology. The soil classification was based on an understanding of the interactions of different environmental and social factors, as they affect the productivity of a particular field. The awareness of differing soil potentials under variable rainfall or anthropogenic inputs supported an essentially flexible and responsive management strategy, which may well be better at coping with the inherent and increasing variability in Sahelian systems than those suggested by orthodox agricultural science. It is clearly important for research into ethnopedologies to avoid easy comparisons of superficial similarities that may overlay a deeper divergence of perspectives. The comparison of agricultural science and local knowledge is not straightforward. The results of the soil chemical analysis, which showed low nutrients levels in all soil types, indicate that orthodox soil analysis may not distinguish the subtle differences in soil productivity represented in the local ethnopedology (although more thorough analysis might detect these). There were fundamental differences between local and scientific systems for defining and describing soil properties. Whereas scientific frameworks tend to be based on intrinsic soil characteristics that can be quantified independently of the context, the local framework defines the important distinguishing characteristics of soils in terms of factors such as location, potential and interaction in the wider ecology and history. Risk-aversion and dynamic soil use (responsive to rainfall, household labour, capital and ability to increase nutrient inputs) are more important in the local framework than are absolute soil fertility and crop production. Critically, the local knowledge is pluralistic, locally and historically embedded and socially constructed, factors that are not included in scientific soil frameworks. The local ethnopedological framework only represents one part of the farmers’ strategy for managing soil fertility. They draw upon varied ecological knowledge to make complex and dynamic management decisions. This research has shown that as well as biophysical, economic and social variability, local perceptions of the environment heavily influence soil fertility management. The local approach, incorporates measurements of crop yield, integrates complex soil – water relations and allows for inter-annual rainfall variability. Each household responds to these influences according to its capabilities, constraints and opportunities and ability to mediate access to resources. The strategies at Fandou Be´ri are similar to patterns in other Sahelian villages (Reenberg and Paarup-Laursen, 1996; Scoones and Toulmin, 1999b; Mortimore and Adams, 1999; Scoones, 2001). There is an urgent need for further studies to discover, develop and maximise the benefits of indigenous knowledge, such as is used in ‘precision farming’ techniques. They are particularly important where labour and capital are constraining. However, despite the local capabilities, it is acknowledged that many of the constraints on the use of technologies to improve soil fertility, such as the scarcity of inputs, labour, land, capital, and to some extent knowledge, can only be overcome by intervention at the policy level. In particular, there needs to be greater access to credit and economic rewards for farmers before investment in soil fertility matches local aspirations and knowledge.

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

475

Acknowledgements This paper draws on separate research carried out in Niger by the authors during the 1997 and 1998 agricultural seasons. It was funded by awards from the Economic and Social Research Council and Natural Environment Research Council, Royal Geographical Society, University College London Research and Travel Committee and the Eric Brown Geographical Trust. We would like to extend our thanks to the villagers of Fandou Be´ri, to our research assistants and local collaborators, and the SERIDA Project. We are grateful for comments from Katherine Homewood, Andrew Warren, John Pearson, Simon Batterbury and two referees. References Abdoulaye, T., Lowenberg-DeBoer, J., 2000. Intensification of Sahelian farming systems: evidence from Niger. Agricultural Systems 62 (3), 67 – 81. Adedeji, A., 1999. Structural adjustment policies in Africa. International Social Science Journal 51 (4), 521 – 525. Allan, C., 1997. Indigenous understanding of soil fertility and management issues in the Sahelian village of Fandou Be´ri, South-western Niger. MRes Thesis, University College London. Altieri, M.A. (Ed.), 1998. Agroecology: The Science of Sustainable Agriculture, 2nd ed. Westview Press, London. Amissah-Arthur, A., Mougenot, B., Loireau, M., 2000. Assessing farmland dynamics and land degradation on Sahelian landscapes using remotely sensed and socio-economic data. International Journal of Geographical Information Science 14 (6), 583 – 599. Baidu-Forson, J., 1999. Factors influencing adoption of land-enhancing technology in the Sahel: lessons from a case study in Niger. Agricultural Economics 20 (3), 231 – 239. Barbier, E.B., 2000. The economic linkage between rural poverty and land degradation: some evidence from Africa. Agriculture, Ecosystems and Environment 82 (1 – 2), 355 – 370. Bationo, A., Mokwunye, A.U., 1991. Alleviating soil fertility constraints to increased crop production in West Africa: the experience in the Sahel. Fertiliser Research 29, 95 – 115. Batterbury, S., Warren, A., Osbahr, H., Taylor, N., Skidmore, D., Seyni, S., Weigl, M., Waughray, D., Longbottom, J., Ousmane, H., Chappell, A., 1999. Land-use and land degradation in SW Niger: change and continuity. SERIDA GEC Report to ESRC, August. Bebbington, A.J., 1999. Capitals and capabilities: a framework for analysing peasant viability, rural livelihoods and poverty. World Development 27 (12), 2021 – 2044. Benneh, G., Morgan, W.B., Uitto, J.I. (Eds.), 1996. Sustaining the Future: Economic, Social and Environmental Change in Sub-Saharan Africa. UN Univ. Press, New York, 378 pp. Bray, R.H., Kurtz, L.T., 1945. Determination of total, organic and available forms of phosphorous in soils. Soil Science 59, 39 – 45. Breman, H., Sissiko, K. (Eds.), 2000. L’intensification agricole au Sahel. Kathala, Paris, 996 pp. Breman, H., Groot, J.J.R., van Keulen, H., 2001. Resource limitations in Sahelian agriculture. Global Environmental Change—Human and Policy Dimensions 11 (1), 59 – 68. Brokensha, D., Warren, D.M., Werner, O., 1980. Indigenous Knowledge Systems and Development. University Press of America, Washington, DC. Brouwer, J., Bouma, J., 1997. Soil and crop growth variability in the Sahel: highlights of research (1990 – 95) at ICRISAT Sahelian Center. Bulletin, vol. 49. ICRISAT, Niger. Buerkert, A., Hiernaux, P., 1998. Nutrients in the west African Sudano – Sahelian zone: losses, transfers and role of external inputs. Zeitschrift fur Pflandzenernahrung und Bodenkunde 161 (4), 365 – 383. Carney, D. (Ed.), 1998. Sustainable Rural Livelihoods: What Contribution Can We Make? Department for International Development, London. Chambers, R., 1997. Whose Reality Counts? Putting the First Last. IT Publications, London.

476

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

Chappell, A., 1995. Geostatistical mapping and ordination analysis of 137Cs-derived net soil erosion flux in south-west Niger. PhD thesis, University of London. Commins, S.K., Lofchie, M.F., Payne, R., 1996. Africa’s agrarian crisis: the roots of famine. Political Famine 34, 16 – 34. Davis, D., 1996. Gender, indigenous knowledge and pastoral resource use in Morocco. The Geographical Review 86 (2), 284 – 288. de Groot, A., 1995. La protection des vegetaux dans les cultures de subsistance: le cas du mil au Niger de L’ouest. Unpublished Report. Dialla, B.E., 1993. The Mossi indigenous soil classification in Burkino Faso. Indigenous Knowledge Monitor 1 (3), 1 – 3. Dorich, R.A., Nelson, D.W., 1983. Direct colorimetric measurement of ammonium in potassium chloride extracts of soils. Soil Science Society of America Journal 47, 833 – 836. Drechsel, P., Gyiele, L., Kunze, D., Cofie, O., 2001. Population density, soil nutrient depletion, and economic growth in sub-Saharan Africa. Ecological Economics 38 (2), 251 – 258. Ellis, F., 2000. The determinants of rural livelihood diversification in developing countries. Journal of Agricultural Economics 51 (2), 289 – 302. Esse, P.C., Buerkert, A., Hiernaux, P., Assa, A., 2001. Decomposition of and nutrient release from ruminant manure on acid sandy soils in the Sahelian zone of Niger, West Africa. Agriculture, Ecosystems and Environment 83 (1 – 2), 55 – 63 (Special Issue). Eswaran, H., Almaraz, R., Reich, P., Zdruli, P., 1997. Soil quality and soil productivity in Africa. Journal of Sustainable Agriculture 10 (4), 75 – 94. Finck, A., 1995. From the fertilisation of crops to the management of nutrients in crop rotations and farming systems: an overview. Fertiliser and Plant Nutrition Bulletin, vol. 12. FAO, Rome. Gentry, C.E., Willis, R.B., 1988. Improved method for automated determination of ammonium in soil extracts. Communications in Soil Science and Plant Analysis 19, 721 – 737. Goutourbe, J.P., Dolman, A.J., Gash, J.H.C., Kerr, Y.H., Lebel, T., Prince, S.D., Stricker, J.N.M. (Eds.), 1997. HAPEX-Sahel. Journal of Hydrology, Special Issue 188/189, 1 – 4 Feb., p. 1090. Gray, L.C., 1997. Land degradation in southwestern Burkina Faso: the environmental effects of demographic and agricultural change. PhD thesis, University of Illinois, Michigan. Guijt, I., van Veldhuizen, L., 1998. What tools? Which steps?: Comparing PRA and PTD. Issue Paper No. 79, Drylands Programme. International Institute for Environment and Development, London. Guyer, J., Peters, P., 1987. Introduction: conceptualising the household. Issues of theory and policy in Africa. Development and Change 18 (2), 197 – 214 (Special Issue). Harris, F., 1999. Nutrient management strategies of smallholder farmers in a short-fallow farming systems in north-east Nigeria. Geographical Journal 165 (3), 275 – 285. Heasley, L., Delehanty, J., 1996. The politics of manure: resource tenure and the agropastoral economy in southwestern Niger. Society and Natural Resources 9, 31 – 46. Hecht, S.B., 1998. The evolution of agroecological thought. In: Altieri, M.A. (Ed.), Agroecology: The Science of Sustainable Agriculture, 2nd ed. IT Publications, London, pp. 1 – 20. Hiernaux, P., Bielders, C.L., Valentin, C., Bationo, A., Fernandez-Rivera, S., 1999. Effects of livestock grazing on physical and chemical properties of sandy soils in Sahelian rangeland. Journal of Arid Environments 41 (3), 231 – 245. Hopkins, J.C., Berry, P., Gruhn, P., 1995. Soil Fertility Management Decisions: Evidence From Niger. International Food Policy Research Institute, Washington, DC. Howorth, C., O’Keefe, P., 1999. Farmers do it better: local management of change in southern Burkina Faso. Land Degradation and Development 10, 93 – 109. Ingram, J., 1990. The role of trees in maintaining and improving soil productivity: a review of the literature. Series CSC 90, AGR-15. Commonwealth Science Council, London. Juo, A.S.R., Manu, A., 1996. Chemical dynamics in slash-and-burn agriculture. Agriculture, Ecosystems and Environment 58 (1), 49 – 60. Krogh, L., Paarup-Laursen, B., 1997. Indigenous soil knowledge among the Fulani of northern Burkina Faso: linking soil science and anthropology in analysis of natural resource management. Geojournal 43, 189 – 197. Lal, R., 2000. Soil management in the developing countries. Soil Science 165 (1), 57 – 72.

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

477

Lamers, J.P.A., Feil, P.R., 1995. Farmer’s knowledge and management of spatial soil and crop growth variability in Niger, West Africa. Netherlands Journal of Agricultural Science 43, 375 – 389. Leach, M., Fairhead, J., 2000. Fashioned forest pasts, occluded histories? International environmental analysis in West Africa. Development and Change 31 (1), 35 – 59. Legger, D., 1993. Soils of the ‘West Central Site’, Niger. Department of Soil Science and Geology, Wageningen Agricultural University, The Netherlands. Longbottom, J., 1996. Productive bricolage: changing livelihoods and gendered strategies in response to food insecurity in southwest Niger. MA thesis, SOAS, London. Lund, C., 1998. Law, Power and Politics in Niger: Land Struggles and the Rural Code. Transaction Publishers, New Brunswick/LIT Verlag, Hamburg. Mahamane, I., Bationo, A., Seyni, F., Hamidou, Z., 1996. Recent achievement of research on indigenous phosphate rocks from Niger. In: Renard, G., Neef, A., Becker, K., von Oppen, M. (Eds.), Soil Fertility Management in West African Land Use Systems. Proceedings of the Regional Workshop, University of Hohenheim, ICRISAT Sahelian Centre, INRAN, 4 – 8 March 1997, Niamey, Niger. Margraf Verlag, Weikersheim, Germany. Manu, A., Geiger, S.C., Pfordresher, A., Taylor-Powell, E., Mahamane, S., Ouattara, M., Isaaka, M., Salou, M., Juo, A.S.R., Puentes, R., Wilding, L.P., 1991. Integrated management of agricultural watersheds: characterisation of a research site near Hamdallaye, Niger. TropSoils Bulletin 91-01. Manu, A., Salou, M., Hossner, L.R., Wilding, L.P., Juo, A.S.R., 1996. Soil related plant growth variability in the Sahel with special reference to Western Niger. TropSoils/TAMU Bulletin 96-01. Matthews, S., 1998. The development of a soil water balance model as an efficient tool for agroclimatic research in the Sudano – Sahelian Zone. MRes Thesis. University College London. Mazzucato, V., Niemeijer, D., 2000. Rethinking soil and water conservation in a changing society: a case study in Eastern Burkina Faso. Tropical Resource Management Papers No. 32. Wageningen Agricultural University, The Netherlands. McCorkle, C.M., 1994. Farmer innovations in Niger. Studies in Technology and Social Change, Series No. 21. Technology and Social Change Program in Collaboration with Center for Indigenous Knowledge for Agriculture and Rural Development. CIKARD, Iowa State University, Iowa. McIntire, J., Powell, J.M., 1995. African semi-arid tropical agriculture cannot grow without external inputs. In: Powell, J.M., Fernandez-Rivera, S., Williams, T.O., Renard, C. (Eds.), Livestock and Sustainable Nutrient Cycling in Mixed Farming Systems of Sub-Saharan Africa. Proceedings ILCA, Addis Ababa, pp. 539 – 551. Mehta, L., Leach, M., Newell, P., Scoones, I., Sivaramakrishnan, K., Way, S.-A., 1999. Exploring understandings of institutions and uncertainty: new directions in natural resource management. IDS Discussion Paper 372, November. Institute of Development Studies, University of Sussex, Brighton. Moris, J., 1991. Extension Alternatives in Tropical Africa. ODI, London. Mortimore, M.J., Adams, W.M., 1999. Working the Sahel: Environment and Society in Northern Nigeria. Routledge, London. Mortimore, M.J., Adams, W.M., 2001. Farmer adaptation, change and ‘crisis’ in the Sahel. Global Environmental Change—Human and Policy Dimensions 11 (1), 49 – 57. Neefjes, K., 2000. Environments and Livelihoods: Strategies for Sustainability. Oxfam, Oxford. Osbahr, H., 1997. Indigenous knowledge, fallow systems and indicator species: a case study from Fandou Be´ri, south-western Niger. MRes Thesis. University College London. Osbahr, H., 2001. Livelihood strategies and soil fertility at Fandou Be´ri, south-western Niger. PhD Thesis, University College London. Osunade, M.A.A., 1992. The significance of colour in indigenous soil studies. International Journal of Environmental Studies 40, 185 – 193. Pawluk, R.R., Sandor, A.J., Tabour, J.A., 1992. The role of indigenous soil knowledge in agricultural development. Journal of Soil and Water Conservation, July – August, 298 – 302. Payne, W.A., Williams, J.H., Moussa, K.A.M., Stern, R.D., 1998. Crop diversification in the Sahel through use of environmental changes near Faidherbia albida (Del) A Chev. Crop Science 38 (6), 1585 – 1591. Penning de Vries, F.W.T., Djiteye, M.A., 1992. La productivitie des paturages saheliens: une etude des sols, des vegetations et de l’exploitation de cette ressource naturalle. Pudoc, Wageningen, The Netherlands.

478

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

Piper, T., 1998. An investigation into the effect of algal crusts on soil fertility in the Sahelian village of Fandou Be´ri, southwestern Niger. MRes Thesis. University College London. Powell, J.M., Ikpe, F.N., Somda, Z.C., 1999. Crop yield and the fate of nitrogen and phosphorus following application of plant material and faeces to soil. Nutrient Cycling in Agroecosystems 54 (3), 215 – 226. Rain, D.R., 1999. Eaters of the Dry Season: Circular Migration in the West African Sahel. Westview Press, Boulder, CO. Raynaut, C., 1997. Societies and nature in the Sahel. Routledge/SEI Global Environment and Development Series, London. Reardon, T., Kelly, V., Crawford, E., Diagana, B., Dione´, J., Savadogo, K., Broughton, D., 1997. Promoting sustainable intensification and productivity growth in Sahel agriculture after macroeconomic policy reform. Food Policy 22 (4), 317 – 327. Reenberg, A., 1996. A hierarchical approach to land use and sustainable agricultural systems in the Sahel. Quarterly Journal of International Agriculture 35, 63 – 77. Reenberg, A., Paarup-Laursen, B., 1996. Determinants for land use strategies in a Sahelian agroecosystem: anthropological and ecological geographical aspects of natural resource management. Sahel – Sudan Environmental Research Initiative, working paper 10. Reij, C., Scoones, I., Toulmin, C. (Eds.), 1996. Sustaining the Soil: Indigenous Soil and Water Conservation in Africa. Earthscan, London, 272 pp. Richards, P., 1985. Indigenous Agricultural Revolution. Unwin Hyman, London. Richards, P., 1989. Agriculture as performance. In: Chambers, R., Longhurst, R., Pacry, A. (Eds.), Farmer First: Farmer Innovation and Agricultural Research. IT Publications, London, 218 pp. Rowland, A.P., 1983. An automated method for the determination of ammonium-N in ecological materials. Communications in Soil Science and Plant Analysis 14, 49 – 63. Scoones, I., 1998. Sustainable rural livelihoods: a framework for analysis. IDS Working Paper No. 72. Institute for Development Studies, University of Sussex, Brighton. Scoones, I. (Ed.), 2001. Dynamics and Diversity: Soil Fertility and Farming in Africa. Earthscan, London, 256 pp. Scoones, I., Thompson, J., 1993. Challenging the populist perspective: rural people’s knowledge, agricultural research and extension practice. Discussion Paper 332. Institute of Development Studies, University of Sussex, Brighton. Scoones, I., Toulmin, C., 1999a. Soil nutrient budget and balances: what use for policy? Managing Africa’s Soils, vol. 6. Institute for Development Studies, Brighton/International Institute for Environment and Development, London. Scoones, I., Toulmin, C., 1999b. Policies for soil fertility management in Africa. Report prepared for DFID, Institute for Development Studies, Brighton/International Institute for Environment and Development, London. Scoones, I., Wolmer, W. (Eds.), 2002. Pathways of change: crops, livestock and livelihoods in Africa. Lessons from Ethiopia, Mali and Zimbabwe. James Currey, Oxford. Sikana, P., 1993. How farmers and scientists see soils: mismatched models. ILEIA Newsletter 9 (1), 15 – 16. Simpson, B., 1999. The Roots of Change: Human Behaviour and Agricultural Evolution in Mali. IT Publications, London. Sivakumar, M.V.K., 1992. Climate change and implication for agriculture in Niger. Climate Change 20, 297 – 312. Smaling, E.M.A., Braun, A.R., 1996. Soil fertility research in sub-Saharan Africa: new dimensions, new challenges. Communications in Soil Science and Plant Analysis 27 (3 – 4), 365 – 386. Spath, H.-J., Francis, M.L., 1994. Deforestation and land surface change in the Hinterland of Niamey, Niger. Applied Geography and Development 43, 27 – 49. Stoller, P., 1989. The Taste of Ethnographic Things: The Senses in Anthropology. University of Pennsylvania Press, USA. Stoorvogel, J.J., Smaling, E.M.A., 1990. Assessment of soil nutrient depletion in sub-Saharan Africa, 1983 – 2000. Report 28, Winard Starting Centre for Integrated Land, Soil and Water Research (SC-DLO), Wageningen, The Netherlands. Strauss, A., Corbin, J., 1990. Basics of Qualitative Research: Grounded Theory Procedures and Techniques. Sage Publications, Newbury Park.

H. Osbahr, C. Allan / Geoderma 111 (2003) 457–479

479

Tabor, J., 1990. Ethnopedology: using indigenous knowledge to classify soils. Arid Lands Newsletter 30, 28 – 29. Taylor-Powell, E., Manu, A.A., Geiger, S.C., Ouattara, M., Juo, A.S.R., 1991. Integrated management of agricultural watersheds: land tenure and indigenous knowledge of soil and crop management. TropSoils Bulletin, vol. 91-04. Soil Management Collaborative Support Program, North Carolina State University, Raleigh, NC. Tel, D.A., Heseltine, C., 1990. The analyses of KCl soil extracts for nitrate, nitrite and ammonium using a TRAACS 800 Analyzer. Communication in Soil Science and Plant Analysis 21, 13 – 16. Tiessen, H., Hauffe, H.-K., Mermut, A.R., 1991. Deposition of Harmattan dust and its influence on base saturation of soils in northern Ghana. Geoderma 49, 285 – 299. Vierich, H.I.D., Stoop, W.A., 1990. Changes in West African Savanna agriculture in response to growing population and continuing low rainfall. Agriculture, Ecosystems and Environment 31, 115 – 132. Warren, D.M., Slikkerveer, L.J., Brokensha, D., 1995. The cultural dimension of development. Indigenous knowledge systems. Studies in Indigenous Knowledge and Development. IT Publications, London. Warren, A., Batterbury, S., Osbahr, H., 2001a. Soil erosion in the West African Sahel: a review and an application of a ‘local political ecology’ approach in South West Niger. Global Environmental Change—Human and Policy Dimensions 11 (1), 79 – 95. Warren, A., Batterbury, S., Osbahr, H., 2001b. Sustainability and Sahelian agriculture: evidence from Niger. The Geographical Journal 167 (4), 324 – 341. Wezel, A., Rajot, J.-L., Herbrig, C., 2000. Influence of shrubs on soil characteristics and their function in Sahelian agro-ecosystems in semi-arid Niger. Journal of Arid Environments 44, 383 – 398.