Version classiqueVersion mobile

Rural societies in the face of climatic and environmental changes in West Africa

Benjamin Sultan
Richard Lalou
Mouftaou Amadou Sanni
et al.

Part IV. Adaptation of farming systems, innovations

Chapter 19. Reintroducing livestock to increase the sustainability of village landscapes in West Africa

The case of groundnut basin in Senegal

Élise Audouin, Jonathan Vayssières, Mariana Odru, Dominique Masse, Séraphin Dorégo, Valérie Delaunay et Philippe Lecomte

Texte intégral


1In West Africa living conditions and food security improvement are still key issues (Lahmar et al., 2012). Farming systems sustainability is a priority in a context in which the rural population is dominant. Traditionally, integrated agro-sylvo-pastoral systems are in the majority in West Africa (Dugué et al., 2012; Jouve, 2001). They are based on the complementarity of ruminant herds and flocks of ruminants, cereal crops and trees (Jouve, 2001). The sustainability of these systems depends to a considerable degree on transfers of fertility from rangelands to cultivated fields via the night corralling of livestock (Freschet et al., 2008; Jouve, 2001). Among other things, these agrarian systems have had to adapt to changing rainfall patterns and strong demographic growth. Indeed, West Africa experienced a long drought period from 1970 to 1995 (Cormier et al., 2000) and the population doubled during the same period (Dongmo et al., 2010; FAO, 2003; Serpantié and Ouattara, 2001). In reaction to this and to respond to increasing food requirements, rural populations have cultivated marginal land (Courtin and Guengant, 2011; Faye and Landais, 1986; Schlecht et al., 2004). These areas were traditional used for grazing. The extension of cultivated land has therefore resulted in a reduction of herbaceous fodder resources and has increased the pressure on wooded areas, especially in dry years. There was particularly strong pressure on the wooded areas during the 1970-1995 drought period, resulting in regression caused by over-exploitation (Courtin and Guengant, 2011; FAO, 2003). As a result, the number of head of livestock decreased and feed was refocused on crop residues, increasing the tensions centred on this resource (Freschet et al., 2008; Jouve, 2001; Rufino et al., 2010). The current trend is that of harvesting residues (Faye and Landais, 1986). They are collected for sale or storage to cover the feed requirements of livestock during the dry season (Dugué, 1985; Fall-touré et al., 1997; Schlecht et al., 2004). It results from a switch from a collective management system to an individual one (Courtin and Guengant, 2011; Dugué, 1985; Jouve, 2001). The decrease in the presence of livestock combined with the harvesting of crop residues (in addition to the crop main product itself) affects soil fertility as exports of nutrients are no longer compensated by equivalent input in the form of manure (Buldgen et al., 1992; Dugué, 1985; FAO, 2003).

2In this study we examined the case of the Sereer populations in the Senegalese groundnut basin. Whereas the landscape already seemed to be saturated in 1960 (Lericollais, 1999), the population of this part of Senegal doubled from 1963 to 2009 (Delaunay and Lalou, 2012). Adaptation strategies differed according to the village (Dugué, 1985). Their common feature is the increase in the transhumance of cattle. In order to keep livestock in village landscapes, called terroirs (see below for definition), some village communities have kept the common fallow while others have developed fattening operations. One might then wonder whether, in a context of landscape saturation and strong climatic variability, an intensive system based on the reintroduction of cattle through fattening is more sustainable than the traditional extensive cattle farming based on a fallow system. To answer the question, we compared two village landscapes that use these different agricultural strategies (fallow versus fattening) in the same soil and climate zone.

Material and method

The conceptual model guiding analysis

3The conceptual analysis model is of the stock-flow types in which three systems are embedded: terroir, household unit and field (Fig. 1). These different systems are crossed by incoming and outgoing flows of biomass. The biomass consists of the main materials circulating in the farming systems studied (Fig. 1), i. e. the main production of the tree layer, crops (grain, residues, etc.) and livestock (manure, animals, milk, etc.) together with the main inputs (mineral fertiliser, concentrated feed, foodstuff, etc.).

The system studied: the village terroir

4Favouring a systemic approach, we chose the ‘Africanist’ definition of terroir as ‘an area cultivated and used by a village community’ (Rabot, 1990). A terroir can be divided into landscape units that make it possible to model the spatial organisation of biomass and nutrient flows (dwelling zone, compound fields, bush fields, rangelands reserved for livestock) (Lericollais, 1999; Manlay, 2001; Tittonell et al., 2006).

Figure 1. Conceptual representation of the system studied in three embedded systems (the terroir, the household unit, and the field).

5This study is focused on two terroirs close to each other (about 8 km apart) in the same soil and climate zone in the Senegal groundnut basin (average rainfall 566 mm. year-1, Fatick meteorological station, 2013). Diohine, the first terroir, features a farming system with a traditional trend and based on a common fallow. Barry Sine, the second terroir, displays a spectacular increase in livestock fattening (Fig. 2).

Household and field units

6Historically, the Sereer are gathered by family membership within a holding. Each holding is a dwelling unit and generally consists of several households. Whereas historically most decisions were taken at the scale of the holding, the household is now usually the main decision unit for the management of agricultural resources (livestock, land, equipment, etc.). The household can be divided into five sub-systems or farming ‘workshops’: family, crops, livestock, effluent and the tree area (see Fig. 1).

Figure 2. Frequency of holdings carrying out fattening in 2012 in the various village terroirs of the IRD HDSS observatory at Niakhar (Source: Delaunay and Lalou, 2012).

7The field is the crop management unit. This sub-system is conventionally considered in studies on the evaluation of the sustainability of farming systems as in agronomy it is the classic scale for the evaluation of crop yields and fertility balances (Schlecht and Hiernaux, 2004).

Data gathering and field units

Household surveys

8The aim of the surveys was to describe the structure and farming practices of each household. The exhaustive list of the households in each village and the corresponding demographic data are from the Niakhar observatory database (Delaunay et al., 2013). Agricultural practices have been converted into biomass flows between the various farming workshops to calculate sustainability (see below). The survey guide was therefore divided according to the 5 agricultural workshops that generate biomass flows (see above). Incoming and outcoming flows were quantified for each workshop (Fig. 1). The reference year was the 2012 farming season (June 2012 to May 2013).

9Inter-household flows were also recorded. These consist of commercial flows of biomass (information drawn directly from surveys) and the flows generated by herds and flocks. Calculation of the latter flows is detailed below (Schlecht et al., 2004; Thornton and Herrero, 2001).

Biomass flow database

10The database of biomass flows is compiled in standard units (kg FW and kg DM). Local units (carts, bundles, etc.) were converted into standard units by weighing during the study or using conversion factors available in the literature (Vandermeersch et al., 2013).

11Flows that could not be quantified directly by a survey, that is to say inter-household flows via flocks and herds, were calculated according to the different practical seasons (see below).

12In the dry season after the harvests, livestock ranges freely in the fields and eats crop residues left in the field. Part of the biomass is then returned as faeces and urine. The same applies in the rainy season when livestock grazes fallows (the case at Diohine).

13Biomass ingestion flows are calculated according to the fodder available for each household. This is production in household’s fields with deduction of the household’s harvest and consumption by the household’s livestock (according to flock and herd size and the feed practices). If the balance is positive, the surplus is used by the livestock of a household with a shortage. If it is negative, it is counter-balanced by using quantities from the fields of households with surpluses.

14Return in the form of animal faeces and urine was calculated using the duration of presence of batches of livestock in Tropical Livestock Units (TLU.h) in the various landscape units and in the different fields for each relevant season (Schlecht et al., 2006).

Spatialisation of flows and stocks using a geographic information system

15We chose the spatial approach to biomass flows since movements of nutrients in the landscape form an essential component of the maintenance of soil fertility in the Sudan-Sahel zone (Rabot, 1990). Each field was therefore georeferenced. The different flows and balances calculated for the area of each field were then spatialised to represent flow distribution in the terroir in order to evaluate the landscape units that form fertility sources or sinks (Manlay et al., 2004; Dugué, 1985; Tittonell et al., 2006).

Calculated sustainability indicators

16The research conducted in West Africa over the past 30 years has shown that nutrients, and especially nitrogen, form a major limiting factor for agricultural productivity (Rufino et al., 2009; Schlecht et al., 2006). The nourishment of grain crops in sandy tropical soils in West Africa is based mainly on the extraction of organic nitrogen from the soil—finite reserves with a limited quantity (Waneukem and Ganry, 1992). Because of the dominant role of nitrogen in the farming systems in the zone studied, we chose this element to construct our main sustainability indicators. Thus, biomass flows (in kg DM.year-1) were converted into nitrogen flow (in kg N.year-1). The four indicators below were calculated at three levels (field, household and terroir) for better understanding of the functioning of the two terroirs studied.

Nitrogen balance

17Agricultural intensification can endanger the sustainability of a farming system if it is not coupled with maintaining soil fertility. The nutrient balance is hence a useful indicator of the sustainability of the system (Roy et al., 2005; Thornton and Herrero, 2001). Only the apparent N balance has been calculated here. It is the difference between N inputs and outputs calculated for the agricultural area in use (AA). It does not include vertical flows (gas emissions, symbiotic fixation, atmospheric deposition, etc.).

Nitrogen use efficiency

18A second evaluation of system sustainability can be performed using N use efficiency. This indicator is calculated by dividing N outputs by N inputs. It indicates the ‘return on investment’ as it shows how many N units are produced for each unit used (Vayssières, 2012).


19The food security of households remains the main preoccupation of the Sereer. Examining the productivity of farming systems is therefore essential for defining sustainability. The indicator is in kg DM. ha-1.year-1, kg N. ha-1.year-1 and kg N. inhabitant-1.year-1. It can be calculated at the three levels. At field level it is crop yields.


20Imports in a terroir or household reveal the dependence of the system on other systems. They also show the independence of the system with regard to climatic events and the degree of pressure on local natural resources (e.g. wooded areas). The indicator can be expressed in kg N. ha-1.year-1 or kg N.inhabitant-1.year-1 (Rufino et al., 2009).


Village structure

Dwelling and land use

21Diohine is an old village in which dwellings are concentrated and organised in districts. The dwelling pattern in Barry Sine is the result of numerous relatively recent population movements and is fragmented and dispersed (Fig. 3). There is thus a ‘district hierarchy’ in Diohine while there is a ‘holding hierarchy’ in Barry Sine.

Figure 3. Agro-ecological zoning of Diohine (top) and Barry Sine (bottom) in 2013.

22Diohine displays a pattern of organisation that is more similar the traditional model organised around the lowlands that are essential for maintaining livestock in the terroir. In contrast, Barry Sine has no lowland, ponds or tree savannah within the traditional village limits.

Human and animal populations, land available and rotations

23Population pressure (180 in Diohine; 320 in Barry Sine) and the livestock stocking rate (0.96 TLU.year. ha-1 in Diohine; 2.31 TLU.year.ha-1 in Barry Sine) are smaller in Diohine. The latter village has conserved an extensive free-ranging livestock system. Livestock grazes on 20% of the UAA under fallow for this purpose during the rainy season. In contrast, Barry Sine has only a few isolated fields under fallow and has developed intensive livestock farming with substantial imports of concentrated feed.

24Millet is still the main crop in both terroirs using 53% and 55% of the UAA in Diohine and Barry Sine respectively. Groundnut decreased in Diohine where it now uses only 7% of the UAA in comparison with 30% in Barry Sine (Fig. 4).

Figure 4. Cropping land use maps of Diohine (top) and Barry Sine (bottom) in 2012.

Diversity of households

25The households in the two terroirs were classified in 5 types according to structural data and farming practices (Table. 1). The patterns are uneven.

Table 1. Characteristics of the different types of household.

Table 1. Characteristics of the different types of household.

26The best-represented type of household is in the ‘Subsistence farmer’ category, especially in Diohine. The second type is the ‘Newly installed’ category, especially in Barry Sine where this is more individualisation of households. Young heads of household have little UAA and few livestock and are particularly dependent on the resources of the terroir (grass and wood).

27The ‘Traditional mixed farmers’ that is found only in Diohine consists of heads of holding who have kept a large proportion of fallow and manage medium-sized traditional and transhumant herds and flocks.

28‘Cash crop farmers’ and ‘Fatteners’ are only found in Barry Sine. Both focus on commercial production (groundnuts and fattening). The ‘Fatteners’ type consists of densely populated households with little UAA.

29‘Fatteners’ have a high TLU ratio in comparison with total TLUs. In contrast, there is little or no fattening in ‘Traditional mixed farmers’. Both types have the feature of making little plant biomass (crop residues) available to the flocks and herds of other households as their stocking rate is high.

30Thus the flocks and herds in ‘ Traditional mixed farmers’ also graze in the fields of other households and in particular in those of ‘ Subsistence farmers’ who have low stocking rates. These farmers thus draw on nitrogen in the fields of other households and that they concentrate in their own fields by night corralling. However, they loan their livestock for corralling, especially to the ‘Newly installed’ category and can thus increase their UAA by borrowing manured fields in the following year (Odru, 2013).

31In contrast, ‘Fatteners’ tend to keep livestock hobbled in the holding that does not participate in free ranging. To feed these animals, the households gather a large proportion of residues on their own fields and also sometimes purchase residues from other households in the ‘Subsistence farmers’ and ‘Newly installed’ categories.

Farming practices and equipments

The hinging of livestocks and crop farming

32The types of dominant households in each of the two terroirs are different. The different practices that result can be described with a distinction made between the main seasons related to crop management and livestock.

33The rainy season and the beginning of the cold dry season (R1, R2 and CD1) form a period during which livestock and crop farming may compete with each other. As a result, part of the ‘free-ranging ruminants’ (cattle, goats and sheep) leave for transhumance to Saloum and then Ferlo (0.11 TLU.ha-1 in Diohine and 1.10 TLU.ha-1 in Barry Sine). Competition between activities is reduced in Diohine thanks to common fallow and access to the lowlands, making it possible to keep more head of livestock in the terroir (0.56 TLU.ha-1 in Diohine; 0.35 TLU.ha-1 in Barry Sine). This is not the case in Barry Sine, where more fattening is carried out (0.01 TLU.ha-1 in Diohine; 0.57 TLU.ha-1 in Barry Sine). This type of livestock farming reduces relations between livestock and crop farming as the animals are hobbled within the holding and feed is strongly based on concentrates purchased from merchants and feeds imported in the village terroir (e.g. complete concentrated feed, millet bran, rice bran, etc.).

34At the end of the cold dry season and during the hot dry season (CD2, HD1, HD2 and HD3), crop and livestock farming become potentially complementary as soon as the by-products are collected. Indeed, the free-ranging livestock graze crop residues in the field and are paddocked at night in certain fields. The hot dry season in Diohine is more split than in Barry Sine, with two distinct lean periods (HD2 and HD3). In the first phase, ‘free-ranging ruminants’ have access to common grazings, feeding in priority on crop residues left in fields and the resources of the lowland (Odru, 2013). In the second phase, fodder is based on stored crop residues and trimmings from fodder trees (e.g. Faidherbia albida).

35The sheep fattening period generally aims at Muslim festivals such as ‘Tabaski’ (Eid al-Adha) while beef fattening is spread over the dry season each year for the production of counter-season meat. The ‘finishing’ type of beef fattening is performed in Diohine; just a few head of cattle are taken from the free-ranging herd for fattening. In contrast, ‘purchase/resale’ fattening is carried out in Barry Sine.

Fertilisation of cultivated land

36The nutrient flows generated by herds and flocks (during free-ranging, rainy season grazing and night corralling), manure and mineral fertiliser are the principle field inputs.

Figure 5. Representation of the pattern of practical seasons in the two village terroirs.

Free-ranging, rainy season grazing and night corraling

37At field level, the contribution of free-ranging livestock forms 32 and 15% of total N input in Diohine and Barry Sine respectively. A large proportion of by-products is left for livestock in Diohine and the withdrawal of plant biomass exceeds the deposition of faeces and urine. It results in nitrogen export of-8 kgN.ha-1 in the dry season during common ranging and-26 kgN.ha-1 in the rainy season during grazing of common fallow. In Barry Sine, common ranging generally fertilises the land (+ 5 kgN.ha-1) as the stocking rate is high, most of the plant biomass is harvested and the livestock has substantial complements.

38As a result of the maintaining of livestock during the rainy season thanks to fallow, night corralling of this season is applied in an area 10 times as large in Diohine (2.7% of the UAA in Diohine, 0.2% in Barry Sine) and more intensively in Diohine (manure deposition 3.95 t DM.ha-1 in Diohine and 1.75 t DM.ha-1 in Barry Sine). The opposite is observed for night corralling during the dry season. It is used 10 times more in Barry Sine (3% of the UAA in Diohine and 39% in Barry Sine) and is applied to all types of fields, however remote they are from the dwelling, where it is only used in compound fields in Diohine (Fig. 6). The average quantities of manure deposited in the dry season are higher in Barry Sine (2.40 t DM.ha-1 in Diohine; 2.57 t DM.ha-1 in Barry Sine).

Manure application

39Manure forms 14% of nitrogen input in Diohine and 12% in Barry Sine. In both villages, manure spreading is performed in priority in compound fields in accordance with the traditional pattern. In contrast, the agricultural equipment in Barry Sine means that a large proportion of the UAA can be manured (Fig. 7).

40Average manure spreading is fairly similar in both villages (1.69 t DM.ha-1 in Diohine; 1.64 t DM.ha-1 in Barry Sine). However, the households in Barry Sine are better equipped with carts (48% of Diohine households have a cart in comparison with 86% in Barry Sine). Better equipment combined with greater availability of manure means that the latter can be better distributed in space (manuring covers 24% of fields in Diohine and 31% in Barry Sine).

Application of mineral fertiliser

41The spreading of mineral fertiliser is not directly related to livestock farming but plays a preponderant role by supplying the greater proportion of nitrogenous inputs in the fields at Barry Sine (8% in Diohine; 26% in Barry Sine). Mineral fertiliser is spread on a larger area in Barry Sine (2% in Diohine; 27% of the UAA in Barry Sine) and in higher doses. Thus the village counterbalances application of organic fertiliser (faeces and manure) favoured for compound fields by the application of mineral fertiliser in priority in the bush fields (Fig. 8).

Crop management and yields

42Following analysis of the main inputs in fields, we now describe their impact in terms of yield. Yields of cereals are greater in Barry Sine and similar for groundnut in both villages (Table 2). It is noted that overall straw yields are significantly greater in Barry Sine. But legume hay yields are smaller there (Table 2).

Figure 6.
Figure 6. Location of night corralling in the rainy season and the dry season in Diohine (top) and Barry Sine (bottom) in 2012.

43The fate of the by-products also varies from one terroir to the other because of their different livestock systems. The free-ranging livestock system is dominant in Diohine. This implies that sufficient quantities are left in the field to allow free-ranging. In contrast, fattening is carried out in Barry Sine and this means that the locals gather the greater part of the by-products (Table 2).

Figure 7. Manuring intensity in Diohine (top) and Barry Sine (bottom) in 2012.

Sustainability indicators

At field level

44Barry Sine uses significantly more inputs per hectare at the field level than Diohine (13.5 kgN.ha-1 in Diohine; 23.5 kgN.ha-1 in Barry Sine). In contrast, the gain in production (+ 15 kgN.ha-1) is not compensated by the gain in inputs (+ 10 kgN.ha-1). Indeed, fodder exports—especially for feed for fattening—are greater in Barry Sine.

Figure 8.
Figure 8. Intensity of the spreading of mineral fertilisers in Diohine (top) and Barry Sine (bottom) in 2012.

45Finally, farming practices have little effect on the nitrogen balances of the two villages, whose averages and medians do not have a significant difference (averages of-20 kgN.ha-1 for Diohine; -23 kgN.ha-1 for Barry Sine) (Fig. 9).

46Comparison of the spatial distribution of field nitrogen balances and after an ANOVA test, Diohine converges towards the traditional model with higher N balances in compound fields than in bush fields. In Barry Sine, this difference in N balance with remoteness from the dwelling is not significant.

Table 2. Comparison of the yields of the main crops and the proportion of by-products left in the fields in Diohine and Barry Sine in 2012.

Table 2. Comparison of the yields of the main crops and the proportion of by-products left in the fields in Diohine and Barry Sine in 2012.

(in red, the lowest figure for a variable compared with the other village; in green, the highest figure for the variable compared with the other village.

Figure 9. Graph of the distribution of nitrogen balances by frequency in Diohine and Barry Sine in 2012.

47Heterogeneity is also greater in Diohine, probably because of the markedly different management practices used in compound fields and bush fields.

At household level

48Farming practices in Barry Sine are resulting in increased use of inputs such as mineral fertiliser, concentrate feeds and foodstuffs (incoming N flows: 3 kg N. inhabitant-1 in Diohine; 20 kgN.inhabitant-1 in Barry Sine) while production in Diohine is based essentially on the use of local resources with the recycling of biomass from livestock farming, crops and trees. This surplus of inputs is devoted essentially to fattening as purchases of foodstuff are similar in the two villages (1.62 and 1.66 kg N.inhabitant-1 in Diohine and Barry Sine respectively).

49Analysis in greater depth of the nitrogen flows at household level shows that the trend in Diohine is more towards on-farm consumption while in Barry Sine the focus is on the sale of farm production. Indeed, sales of plant and animal products are much higher in Barry Sine (1 kg N.inhabitantt-1 in Diohine; 13 kg N.inhabitant-1 in Barry Sine).

50The flows resulting from the movement of free-ranging livestock are dominant in Diohine, whereas inflows and outflows related to fattening are dominant in Barry Sine. These figures show the importance of livestock farming in the functioning of the two terroirs, even though the livestock systems differ strongly.

51The current practices and functioning of households result in N balances at this level that are greater on average in Barry Sine (13 kgN.ha-1 in Diohine; 24 kg N.ha-1 in Barry Sine). These nitrogen balances increase when the focus of the household is more on livestock. Indeed, ‘Subsistence farmers’ and ‘Fatteners’ have respectively the lowest and highest nitrogen balances at household level.

At terroir level

52The results at terroir level are similar to those observed at household level. The nitrogen balances are positive in both cases and the balance is higher in Barry Sine (+ 8.5 kg N.ha-1 in Diohine; + 24.9 kg N.ha-1 in Barry). The intensity of nitrogen flows is much higher in Barry Sine. Indeed, incoming flows are 7 times greater in Barry Sine (9.3 kg N.ha-1 in Diohine; 67.9 kg N.ha-1 in Barry Sine). Likewise, outgoing N flows are 35 times as high in Barry Sine (1.2 kg N.ha-1 in Diohine; 43.9 kg N.ha-1 in Barry Sine). These differences functioning result in N utilisation efficiency that is 4 times greater in Barry Sine than in Diohine (0.15 in Diohine; 0.64 in Barry Sine).

General discussion: the effects of fattening on the sustainability of terroirs in West Africa

53Soil impoverishment is a problem common to numerous countries in West Africa (Smaling et al., 1997). It is seen in Figure 10 that West African nitrogen balances are generally negative at the field level. Even though methodology may differ from one study to another, the balances for Diohine and Barry Sine are comparable with those for the sub-region (Schlecht and Hiernaux, 2004).

Figure 10. Nitrogen balances at the field level in West Africa.

54This study of biomass and nitrogen flows describes the case of Diohine, confirming that the presence of livestock strongly calls into question the traditional system based on the close complementarity between livestock farming, crop farming and trees (Lericollais, 1999). Indeed the fertility transfers traditionally governed by free-ranging ruminants (Manlay et al, 2004) are becoming insufficient to make up for N outflow via crops, whence the negative N balances at field level and slightly positive N balances at the household and terroir level in the case of Diohine (Table 3).

55This study shows that the farming system based on fattening allows an increase in the number of head of ruminants at terroir level. The livestock stocking rate is higher in Barry Sine and the animals are better fed (see above). This results in the availability of more manure and a larger supply of organic matter at the field level in Barry Sine (+ 2.2 kg N.ha-1). As a result, the yields of millet—the main crop in both terroirs—have increased on average. This gain in yields corresponds to a surplus of + 101 kg DM.ha-1 of ears and + 595 kg DM.ha-1 of millet straw (see above). These gains in average yields at the level of the terroir make it possible to feed more head of livestock and contribute to increasing meat production by 189 kg LW.ha-1, the sale of which is a useful cash return for purchasing mineral fertiliser and foodstuff. Fattening thus intensifies crop production indirectly and supports a population density 78% higher (see above), thus confirming the hypothesis of Duncan et al. (2013).

56In order to maintain a high stocking rate (2.31 UBT.ha-1), Barry Sine farmers as a whole import an annual 411 t DW of concentrated feeds (see above). The use of imported concentrated feeds is a debatable point. It is true that it reduces the autonomy of households and exposes them to fluctuation of prices of cereals and processing by-products, but it also reduces the pressure of farming systems on local natural resources and makes them less sensitive to climatic variations. Indeed, in the traditional system (like that in Diohine), great demand is made on trees in dry years to make up for the shortage of crop residues needed to feed livestock. But the sustainability of such a system is called into question strongly during prolonged drought like that of the period 1970-1995. With a system centred on fattening (such as that at Barry Sine), the shortage of fodder in dry years is felt less because most of the ration consists of concentrated feeds.

Table 3. Multi-level comparison of the sustainability indicators of the villages.


Barry Sine

Nitrogen balance (kg N.ha-1)










Nitrogen efficiency (Dmnl)










57Table 3 shows the great difference between results at field level and those calculated for the household and the terroir. Indeed, the nitrogen balance is negative at field level and amply positive at the higher levels. Analysis of N flows in greater depth accounts for these differences in the N balance, showing that a proportion of available organic matter (manure, faeces and urine) is not used in cultivated fields. Thus part of the nitrogen accumulates at the dwelling level in both villages. The difference between the N balances is even greater for Barry Sine (see above) where manure is even less well used. Analysis of N flows by household shows that the use of manure depends considerably on the type of household (see above). As a general rule, the more manure available, the less will be spread per TLU. Indeed, certain households (‘Fatteners’) concentrate only on the sale of livestock, pay less attention to crops and import a large proportion of the feed using the proceeds of fattening. In contrast, households that focus on both the sale of fattened livestock and a cash crop (‘Cash crop farmers’) make better use of manure. This has also been noted in other contexts of the development of sales-orientated livestock farming in sub-Saharan Africa (Rufino et al., 2009; Schlecht et al., 2006).

Table 4. The environmental, technical and economic benefits of the installation of manure pits. Per household and per year (Audouin, 2014).

Table 4. The environmental, technical and economic benefits of the installation of manure pits. Per household and per year (Audouin, 2014).

58Audouin (2014) evaluated for the two villages the environmental, technical and economic advantages of making better use of manure by installing manure pits, assuming that haulage equipment and labour would not be limiting factors (Table 4). Manure pits mean an increase in the biomass collected (rough stems that would otherwise be burned) and a reduction of losses of gaseous N (Blanchard et al., 2011).

59The analysis shows that the gain would be particularly substantial in the terroirs in which fattening is preponderant, as in Barry Sine (Table 4). The advantage is much smaller for households with small livestock stocking rates and where the dominant livestock system is the free-ranging type as in Diohine. Indeed, manure production is limited there.


60This study provides a comparative view of two village terroirs that have followed different pathways in order to adapt to the changes in their environment. Diohine, the first, is close to the traditional Sereer agro-sylvo-pastoral system organised around fallow and free-ranging livestock and is strongly dependent on trees during the lean period. The second, Barry Sine, is managed more intensively and centred on stabled fattening. Farming practices and the resulting biomass flows were described by means of surveys and used to calculate indicators describing the productivity of the terroirs and the evolution of soil fertility.

61Diohine is an old terroir in which the inter-generation social hierarchy is still strong. The weight of tradition has contributed to maintaining the collective management of resources by organising a common fallow that enhances interaction between households via free-ranging livestock. The ownership of livestock is strongly linked to social status and determinant for agricultural productivity and the maintaining of soil fertility as most household nitrogen input is based on common grazing and night corralling practices.

62The Barry Sine terroir is more recent. It has developed into a more individual and more intensive resource management system as most household nitrogen input is based on the importing of concentrated feed for fattening. The village has also conserved commercial groundnut growing, facilitated by mineral fertiliser spreading 5.7 times as high as in Diohine (in terms of nitrogen). Its agricultural equipment allows better distribution of nitrogen application between compound fields and bush fields. The Barry Sine terroir is thus more productive because of an increase in crop yields (+ 39% N for main products, + 45% N for by-products) and livestock productivity (x 9 kg N.ha-1 produced as meat).

63Analysis of biomass flows shows that livestock still plays a major role in the organisation of the nitrogen cycle in both villages. In Barry Sine, fattening increases the presence of ruminants in the village terroir and field input of nitrogen in the form of manure thus increases by 1.5. However, the nitrogen balances at field level remain negative in both terroirs (-20 kg N.ha-1 in Diohine; -23 kg N.ha-1 in Barry Sine). They reveal insufficiently sustainable management of soil fertility explained by the amount of outgoing flows from the fields and the limited return flows. The present farming systems are thus based on non-sustainable drawing on soil resources. Margins for progress are nonetheless possible as the nitrogen balances of the terroirs (+ 9 kg N.ha-1 in Diohine; + 25 kg N.ha-1 in Barry Sine) and households (+ 13 kg N.ha-1 in Diohine; + 25 kg N.ha-1 in Barry Sine) are positive. The differences between the balances show that nitrogen accumulates at dwelling level and that nitrogen recycling is not optimum. This results in particular from the less than optimum use of manure.

64The populations of the two terroirs are aware of the decrease in soil fertility and show particular interest in the intensification of farming systems by the application of animal manure. They are thinking in particular of increasing manure use by installing manure pits. This would be beneficial from the environmental point of view (better N balances at the field level). The advantages of this innovation are clearer for Barry Sine from the technical and economic point of view as fattening operations generate substantial quantities of manure.

65In a context of climate change and strong demographic growth, this question of intensification options for village farming terroirs is common to many parts of West Africa. Intensification by the introduction of livestock fattening seems to be a promising pathway. The option not only has the advantage of increasing livestock and crop productivity but also improves household finances and reduces their sensitivity to local climatic variations as feeding livestock is less dependent on the fodder availability, thus reducing pressure on local resources, including tree layer during droughts.


66The authors thank the farmers for their availability and the interest that they showed in this work. They also thank Ibrahima Thiaw and Mamadou Lamine Ndiaye, students in cartography at UCAD, who performed part of the field survey and drafted all the maps in this chapter.



Audouin E., 2014
Comparaison de deux terroirs en termes de flux de biomasse et de bilans azotés en vue de proposer des voies d’intensification écologique – Cas de Diohine et Barry Sine dans le Bassin Arachidier du Sénégal. Thèse de master, EIP (École d’ingénieurs de Purpan), Toulouse, 118 p.

Audouin E., Vayssières J., Bourgoin J., Masse D., 2014
«Identification de voies d’amélioration de la fertilité des sols par atelier participatif». In: Actes symposium 50 ans de Niakhar, IRD Éditions, sous presse, 22 p.

Blanchard M., Koutou M., Vall E., Bognini S., 2011
Comment évaluer un processus innovant ? Cas de l’amélioration quantitative et qualitative de la fumure organique au champ. Revue d’élevage et de médecine vétérinaire en pays tropicaux, 64 (1-4) : 61-72.

Buldgen A., Detimmerman F., Priraux M., Compère R., 1992
Les techniques d’embouche de moutons en région soudano-sahélienne sénégalaise. Nutrition et Alimentation, 35 (3-4) : 321-328.

Cormier M. C., Gueye C., Lericollais A., Seck S. M., 2000
«Sécheresse». In: La construction de l’espace sénégalais depuis l’indépendance, 1960-2000, [ 2013]

Courtin F., Guengant J. P., 2011
Un siècle de peuplement en Afrique de l’Ouest. Natures Sciences Sociétés, 19 (3) : 256-265.

Delaunay V., Douillot L., Diallo A., Dione D., Trape J. F., Medianikov O., Raoult D., Sokhna C., 2013
Profile: The Niakhar Health and Demographic Surveillance System. International Journal of Epidemiology, 42 (4): 1002-1011.

Delaunay V., Lalou R., 2012
Culture de la pastèque, du sanio et pratique de l’embouche bovine dans la zone d’étude de Niakhar. Enquête légère juin 2012, rapport d’analyse. IRD, Dakar, Sénégal, 12 p.

Dongmo A. L., Dugué P., Vall E., Lossouarn J., 2010
«Optimiser l’usage de la biomasse végétale pour l’agriculture et l’élevage au Nord-Cameroun». In: Savanes africaines en développement: innover pour durer, du 20 au 23 avril 2009, Éditions Seiny-Boukar L. et Boumarda P., N’Djaména, Tchad/Montpellier, Cirad, 10 p.

Dugué P., 1985
L’utilisation des résidus de récolte dans un système agro-pastoral sahélo-soudanien au Yatenga (Burkina Faso). Cahiers de la Recherche-Développement, 7 : 28-37.

Dugué P., Vayssières J., Chia E., Ouedraogo S., Havard M., Coulibaly D., Nacro H. B., Sissoko F., Sangare M., Vall E., 2012
L’intensification écologique: réflexions pour la mise en pratique de ce concept dans les zones de savane d’Afrique de l’Ouest. Actes du séminaire ASAP, Partenariat, modélisation, expérimentation: quelles leçons pour la conception de l’innovation et l’intensification écologique, Vall E., Andrieu N., Chia E., Nacro H. B., éd., novembre 2011, Bobo-Dioulasso, Burkina Faso, 15 p.

Duncan A. J, Tarawali S. A., Thorne P. J., Valbuena D., Descheemaeker K., Homann-Kee Tui S., 2013
Integrated crop-livestock systems – a key to sustainable intensification in Africa. Tropical Grasslands Forrajes Tropicales, 1: 202-206.

Fall-Touré S., Traoré E., N’diaye K., N’diaye N. S., Sèye B. M., 1997
Utilisation des fruits de Faidherbia albida pour l’alimentation des bovins d’embouche paysanne dans le bassin arachidier au Sénégal. Livestock Research for Rural Development, 9 (5): 1-17.

Faye A., Landais E., 1986
L’embouche bovine paysanne dans le centre-nord du bassin arachidier au Sénégal. Cahiers de la recherche-développement en milieu rural, 9-10 : 113-120.

Food and Agriculture Organization of the United Nations, 2003
Gestion de la fertilité des sols pour la sécurité alimentaire en Afrique subsaharienne. Rome, FAO, Italie, 66 p.

Freschet G. T., Masse D., Hien E., Sall S., Chotte J. L., 2008
Long-term changes in organic matter and microbial properties resulting from manuring practices in an arid cultivated soil in Burkina Faso. Agriculture, Ecosystems and Environment, 123: 175-184.

Jouve P., 2001
«Jachères et systèmes agraires en Afrique subsaharienne». In Floret C., Pontanier R.: La jachère en Afrique tropicale. Rôles, aménagement, alternatives. De la jachère naturelle à la jachère améliorée, le point des connaissances. Paris, IRD Éditions: 1-20.

Lahmar R., Bationo B. A., Lamso N. D., Guéro Y., Tittonell P., 2012
Tailoring conservation agriculture technologies to West Africa semi-arid zones: Building on traditional local practices for soil restoration.
Field Crops Research, 132: 158-167.

Lericollais A., 1999
Paysans sereer. Dynamiques agraires et mobilités au Sénégal.
Paris, IRD Éditions, coll. À travers champs, 681 p.

Manlay R. J., 2001
Organic matter dynamics in mixed-farming systems of the West African savanna. A village case study from South Senegal. Thèse doctorale, École nationale du génie rural, des eaux et forêts, Montpellier, 192 p.

Manlay R. J., Ickowiczc A., Masse D., Feller C., Richard D., 2004
Spatial carbon, nitrogen and phosphorus budget in a village of the West African savanna—II. Element flows and functioning of a mixed-farming system. Agricultural Systems, 79: 83-107.

Odru M., 2013
Flux de biomasse et renouvellement de la fertilité des sols à l’échelle du terroir. Étude de cas d’un terroir villageois sereer au Sénégal.
Thèse de master, Istom, Cergy-Pontoise, 109 p.

Rabot C., 1990
Transfert de fertilité et gestion des terroirs, Quelques points de vue.
Les Cahiers de la Recherche-Développement, 25 : 19-32.

Roy R. N., Misra R. V., Lesschen J. P., Smaling E. M., 2005
Évaluation du bilan en éléments nutritifs du sol. Approches et méthodologies.
Bulletin FAO engrais et nutrition végétale, 14, FAO, Rome, Italie, 85 p.

Rufino M. C., Hengsdijk H., Verhagen A. 2009
Analysing integration and diversity in agro-ecosystems by using indicators of network analysis.
Nutrient Cycling in Agroecosystems, 84: 229-247.

Rufino M. C., Dury J., Tittonell P., Wijk M. T. V., Herrero M., Zingore S., Mapfumo P., Giller K. E., 2010
Competing use of organic resources, village-level interactions between farm types and climate variability in a communal area of NE Zimbabwe.
Agricultural Systems, 104 (2): 175-190.

Schlecht E., Hiernaux P., 2004
Beyond adding up inputs and outputs: process assessment and upscaling in modelling nutrient flows.
Nutrient Cycling in Agroecosystems, 70: 303-319.

Schlecht E., Hiernaux P., Achard F. O., Turner M. D., 2004
Livestock related nutrient budgets within village territories in western Niger. Nutrient Cycling in Agroecosystems, 68: 199-211.

Schlecht E., Buerkert A., Tielkes E., 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.

Serpantié G., Ouattara B., 2001
«Fertilité et jachères en Afrique de l’Ouest». In Floret C., Pontanier R.:
La jachère en Afrique tropicale. Rôles, aménagement, alternatives. De la jachère naturelle à la jachère améliorée, le point des connaissances. Paris, IRD Éditions: 21-83.

Smaling E. M. A., Nandwa S. M., Janssen B. H., 1997
«Soil fertility in Africa is at stake». In Buresh R. J., Sanchez P. A., Calhoun F., eds: Replenishing Soil Fertility in Africa, Wisconsin, ASSA, CSSA, SSSA: 47-61.

Thornton P. K., Herrero M., 2001
Integrated crop-livestock simulation models for scenario analysis and impact assessment. Agricultural Systems, 70 (2-3): 581-602.

Tittonell P., Leffelaar P. A., Vanlauwe B., Wijk M. T. V., Giller K. E, 2006
Exploring diversity of crop and soil management within smallholder African farms: A dynamic model for simulation of N balances and use efficiencies at field scale. Agricultural Systems, 91 (1-2): 71-101.

Vandermeersch C., Marra A., Ndiaye P., Ndiaye O., Faye S., Levi P., Naulin A., Ekoudvidjin E., 2013
Rapport sur les enquêtes « Culture élevage », « Ménage équipement » et le « Suivi scolaire » : document technique et axes de recherche. IRD, Dakar, Sénégal, 270 p.

Vayssières J., 2012
Modélisation participative et intégration des pratiques décisionnelles d’éleveurs dans un modèle global d’exploitation. Thèse doctorale, Centre international d’études supérieures en sciences agronomiques, Montpellier, 179 p.

Waneukem V., Ganry F., 1992
Relations entre les formes d’azote organique du sol et l’azote absorbé par la plante dans un sol ferrallitique du Sénégal.
Cahiers Orstom, série Pédologie, 27 (1): 97-107.

Table des illustrations

Légende Figure 1. Conceptual representation of the system studied in three embedded systems (the terroir, the household unit, and the field).
Fichier image/jpeg, 382k
Légende Figure 2. Frequency of holdings carrying out fattening in 2012 in the various village terroirs of the IRD HDSS observatory at Niakhar (Source: Delaunay and Lalou, 2012).
Fichier image/jpeg, 225k
Légende Figure 3. Agro-ecological zoning of Diohine (top) and Barry Sine (bottom) in 2013.
Fichier image/jpeg, 409k
Légende Figure 4. Cropping land use maps of Diohine (top) and Barry Sine (bottom) in 2012.
Fichier image/jpeg, 200k
Titre Table 1. Characteristics of the different types of household.
Fichier image/jpeg, 183k
Légende Figure 5. Representation of the pattern of practical seasons in the two village terroirs.
Fichier image/jpeg, 297k
Légende Figure 6.Figure 6. Location of night corralling in the rainy season and the dry season in Diohine (top) and Barry Sine (bottom) in 2012.
Fichier image/jpeg, 163k
Légende Figure 7. Manuring intensity in Diohine (top) and Barry Sine (bottom) in 2012.
Fichier image/jpeg, 171k
Légende Figure 8.Figure 8. Intensity of the spreading of mineral fertilisers in Diohine (top) and Barry Sine (bottom) in 2012.
Fichier image/jpeg, 162k
Titre Table 2. Comparison of the yields of the main crops and the proportion of by-products left in the fields in Diohine and Barry Sine in 2012.
Légende (in red, the lowest figure for a variable compared with the other village; in green, the highest figure for the variable compared with the other village.
Fichier image/jpeg, 118k
Légende Figure 9. Graph of the distribution of nitrogen balances by frequency in Diohine and Barry Sine in 2012.
Fichier image/jpeg, 91k
Légende Figure 10. Nitrogen balances at the field level in West Africa.
Fichier image/jpeg, 186k
Titre Table 4. The environmental, technical and economic benefits of the installation of manure pits. Per household and per year (Audouin, 2014).
Fichier image/jpeg, 162k


© IRD Éditions, 2017

Licence OpenEdition Books

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search