Chapter 4. Results

Macrolevel

Ghana

Figure 14 shows the simulated land use map of Ghana. Cocoa and oil-palm are the dominant crops in the wetter southwest of the country, and maize, millet, sorghum and groundnuts in the drier north. Cassava is found between these zones and in the southeast. Based on this land use map, Table 26 presents the total and per-crop nutrient balances for Ghana. The differences between the crops are large, e.g. coconut and cassava are very depleting, whereas fallow and groundnuts have an almost neutral nutrient balance.

The left-hand map in Figure 15 shows the N balance at the original 1-km resolution. Fallow and groundnuts have almost neutral nutrient balances, whereas coconut and cassava have strongly negative balances (Table 26). The pattern of this map is very scattered and the uncertainty of the results is very high because of the uncertainties in the land use map. Therefore, the map was aggregated to a 20-km grid (using the median value) in order to obtain a picture of the total farming system, instead of individual crops (right-hand map in Figure 15). The highest depletion rates are in the southeast and the central-west parts of Ghana, which corresponds to the cassava area.

TABLE 26
Nutrient balance per crop and total balance, Ghana

Kenya

The simulated land use map of Kenya (Figure 16) shows an intricate pattern of land use. Agriculture is concentrated in the central and west highlands. Tea and coffee is found mainly in the west of the country and maize mainly in the central-east part. Table 27 presents the nutrient balances calculated on the basis of Figure 16. Figure 17 shows the N balance in a spatially explicit way. Positive values in the central-east part are mainly due to pulses and fallow. There is considerable depletion around Mount Kenya and in west Kenya.

TABLE 27
Nutrient balance per crop and total balance, Kenya

Mali

The simulated land use map for Mali (Figure 18) shows a scattered pattern with millet and maize mainly in the drier northern zone. Rice is found mainly in the irrigated areas along the Niger River. Table 28 presents the nutrient balances for Mali based on this land use map. Food crops, such as millet, sorghum and maize, have more negative nutrient balances. The cash crops and legumes are less depleting.

TABLE 28
Nutrient balance per crop and total balance, Mali

The N-balance map of Mali (Figure 19) is mainly positive in the central part of Mali (rice and fallow), while the northern border of the agricultural zone has, on average, greater depletion (millet and maize).

Figure 20 shows all N flows per country. In Kenya, the input of mineral and organic fertilizer is relatively important, whereas Ghana has a larger input by atmospheric deposition because of Harmattan dust. Outflows by leaching and gaseous losses are somewhat greater in Kenya because more mineral fertilizers are used. Most striking is the importance of erosion in Kenya, which is caused by the topography and the relatively fertile volcanic soils. Annex 8 details the N, P and K flows per crop.

Mesolevel

Cocoa-based farming system in Ghana

Figure 21 presents the nutrient balances for the two selected districts. The balance for Nkawie District is considerably more negative than that for Wassa Amenfi District. The main reason for this difference is the area under cocoa, which is 58 percent of the total area in Nkawie District, and 90 percent in Wassa Amenfi District. The nutrient balance for cocoa is only slightly negative as opposed to most other crops. These other crops cause the total nutrient balance for Nkawie District to be more negative. Cassava, yam and cocoyam have strongly negative nutrient balances (Tables 29 and 30). These root crops receive little mineral or organic fertilizer, while crop products remove many nutrients

TABLE 29
Nutrient balance for Nkawie District

TABLE 30
Nutrient balance for Wassa Amenfi District

Figure 22 presents the individual nutrient flows for the two districts. Nutrient losses by crop products (OUT1) cause the largest negative values. The other outflows are more or less of equal importance. The most striking aspect is the very low input of fertilizer. The farming system depends almost completely on natural resources, which are inputs by deposition (IN3) and N fixation by trees and non-symbiotic N fixation (IN4). This results in a total N input of 12 kg/ ha. Annex 10 presents the nutrient balances for each crop.

Tea - coffee - dairy - based farming system in Kenya

The nutrient balance for the tea - coffee - dairy zone of Embu District is very negative. All crops except tea and coffee have strongly negative nutrient balances (Table 31). Nutrients in crop products, leaching and, in particular, erosion are the major contributors to this negative balance (Figure 23). Although the quantity of eroded soil is not very large, nutrient losses are large because the soil is nutrient rich (being developed on young volcanic deposits). As a consequence, they are rather fertile with high organic-matter contents.

TABLE 31
Nutrient balance of the tea-coffee-dairy zone, Embu District

Figure 24 shows the N flows for three important crops in the coffee-tea-dairy zone of Embu District. All inflows are very small compared with the outflows. The differences between the coffee (a cash crop) and maize and beans (food crops) are evident. Larger amounts of mineral fertilizer (IN1) are used for coffee than for the food crops. In this connection, fertilizer use on cash crops may cease if world market prices are too low, as is now the case for coffee.

Nutrient losses due to leaching and erosion are small for the perennial coffee crop. Beans obtain part of their N requirements through symbiotic N fixation (IN4). However, this quantity is quite small compared with the losses. Erosion (OUT5) is the most important outflow, mainly because of the large nutrient content of the soil. Annex 10 presents the nutrient balances for each crop.

Cotton-based farming system in Mali

The overall nutrient balance for the cotton-based farming system in Koutiala Region is moderately negative (Table 32). For P, the nutrient balance is positive. The nutrient balance for cotton compensates in part for the other crops, because cotton receives a large amount of mineral fertilizer and manure. The most important losses are through crop products (OUT1) and leaching (OUT3) (Figure 25). Losses by erosion are relatively small because the soil is very poor in nutrients.

TABLE 32
Nutrient balance for Koutiala Region

Cotton as a cash crop and millet and sorghum as food crops are the main crops in Koutiala Region. The differences between these crops are evident. Sorghum and millet receive little fertilizer, while losses due to crop products and crop residue removal are similar to cotton (Figure 26). However, millet and sorghum do receive fertilizer indirectly as the fertilizer applied to cotton has an effect on millet and sorghum in the next crop cycle. Annex 10 presents the resulting nutrient balances for each crop.

Microlevel

Farms in Embu District, Kenya

The VARINUTS project monitored 15 farms, 3 farms per AEZ (Table 3), for two years using the NUTMON approach. The nutrient balances show clear gradients with the AEZs (Table 33).

TABLE 33
Nutrient balance per AEZ

Source: SC-DLO et al., 2000.

Although these results are per AEZ, which is mesolevel, the study is based on averages from three farms per AEZ. This means that no mesolevel data were involved, which makes it a microlevel study. The farms at the higher altitudes rely more on external inputs and have larger losses owing to leaching, gaseous losses and erosion (Figure 27). Although AEZ1, AEZ2 and AEZ3 appear to have a comparable total balance, the results are significantly different when nutrient losses are compared with the nutrient stocks.

Figure 28 shows that AEZ1 has the greatest relative nutrient losses.

The variations between farms in each AEZ and between AEZs were very large for each soil property (Table 34). The average coefficients of variance (CVs) and the number of soil units identified can serve as indicators of on-farm variation. The average number of soil units identified per farm decreased from five in the upper parts of the district to three in AEZ5 (Table 35). The variation between the units is significant and especially large in the zones where few units were identified.

CVs were also calculated for each individual soil unit with more than two samples. Again, especially in AEZ5, high CVs were found. A possible reason is that the identification of soil units is more difficult because of the flat topography.

TABLE 34
Soil properties for all farms and AEZs

Source: Stoorvogel et al. (2000).

TABLE 35
Average coefficients of variance for soil properties

Figure 29 shows that farm nutrient balances are also variable. In part, these differences reflect soil properties, e.g. leaching is less in soils with a high organic-matter content, and in part, differences in management, such as crop choice, fertilizer application, and soil conservation measures.

Villages in Mali-Sud, Mali

The villages of M’Peresso and Noyaradougou in Mali were examined for the microlevel analysis. Both villages are within the CMDT region of Mali-Sud. At first glance, they have comparable farming systems (Kanté, 2001). Cotton is the basic cash crop and cereals, such as maize, sorghum and millet, the major food crops. Livestock is important. A closer examination shows that land pressure is considerably higher in M’Peresso (higher population density, higher ratio of cultivated land to total land) and, as a consequence, the management of crop residues is more intensive in M’Peresso. Similarly, one might say that Noyaradougou has a labour shortage such as to prevent the village from recycling all crop residues. The burning of residues is 3 percent in M’Peresso compared with 16 percent in Noyaradougou. In addition, cotton yield increments as a result of manure application are larger in M’Peresso. Noyaradougou uses more mineral fertilizers to compensate.

Tables 36 - 38 show some general characteristics of the two villages. The total N stocks (0 - 40 cm) range from 600 kg/ha in M’Peresso to 900 kg/ha in Noyaradougou. The total P stocks (0 - 40 cm) range from 400 kg/ha in M’Peresso to 600 kg/ha in Noyaradougou. These differences are expressed terms of the soil fertility management (Table 39) and the nutrient balances (Table 40).

TABLE 36
Characteristics of M’Peresso and Noyaradougou

* TLU = tropical livestock unit (standard bovine 250 kg).
Source: Kanté (2001).

TABLE 37
Average cultivated areas per farm, two villages, Mali

Source: Kanté (2001).

TABLE 38
Average quantity of mineral fertilizer, two villages, Mali

* Cotton complex NPK = 14:9.6:10 and cereal complex NPK = 15:6.5:12.5.
** Number of fertilized fields.
Source: Kanté (2001).

Farm household members were used as the basis for assigning households by three nutrient management groups (Class 1 = good, Class 3 = poor). Interviews focused on three groups for this purpose: older men, younger men, and women. The class assigned largely reflected: the number of household members; possession of animals and, hence, manure; and carts. Class 3 compensates for the lack of manure by applying more mineral fertilizer per hectare (but has fewer hectares). A further partitioning of the above generates the outputs per village per class (Table 41). An annual evaluation of the classification can result in farmers being promoted or relegated to another class.

TABLE 39
Observed differences between two villages, Mali

Source: Kanté (2001)

TABLE 40
Partial nutrient balances for two villages, Mali

Source: Kanté (2001).

TABLE 41
Partial nitrogen balance for different classes and villages, Mali

Source: Kanté (2001).