The mesolevel nutrient-balance calculation utilized tabular data. The approach involved establishing relations between land use and soils in order to compensate for the lack of spatial data. For example, in Ghana, farmers grow cocoa mainly on top and slope positions, where orthic Acrisols are located. As it is easier to adapt tabular data rather than spatial data, it is possible to include exceptions. In particular, compared with the macrolevel, it enables an improved description of crop management data, which is very important at the mesolevel. It was not possible to create a reconnaissance or semi-detailed land-use map for the mesolevel study areas. Mesolevel data for a map-based procedure were not available and macrolevel data would have provided no added value and failed to be sufficiently representative. At the mesolevel, a 1-km grid is too coarse for representing physiographic differences with sufficient accuracy. If higher resolution data were available, it would be necessary to develop a new procedure for the creation of a land-use map. This is because a suitability approach is not appropriate at the mesolevel, where socio-economic factors are far more important. Aerial photographs and satellite images with fast field checks can also provide the basis for creating a land-use map. However, these were not available for the study areas. The data availability for the three countries was very different, which prohibited a generic approach at the mesolevel.
IN1 Mineral fertilizer
The sources of data on mineral fertilizer were farm surveys, recommended fertilizer rates and macrolevel data, depending on the data availability in each study area. Recommended fertilizer rates are generally much higher than the actual application rates because not all farmers wish or can afford to apply these quantities. Therefore, in order to prevent overestimation, the fertilizer rates were multiplied by a factor representing the ratio between the harvested area at the mesolevel and the harvested area at national level.
IN2 Organic fertilizer
The amount of available manure was derived from the number of livestock within the study area, using excretion, nutrient-content and nutrient-loss factors. The application per crop was derived from farm surveys or estimates. Local nutrient-content values were used where available.
IN3 Atmospheric deposition
The atmospheric deposition was derived from the macrolevel. Rainfall data from local weather stations were used or, where these were not available, they were derived from the macrolevel.
IN4 N fixation
Nitrogen fixation was treated in a way similar as that for the macrolevel. Where available, specific data related to N fixation should be included, e.g. agroforestry systems with N-fixing trees.
IN5 Sedimentation
Irrigation was not relevant for the three case-study areas. Sedimentation was estimated for river-valley crops, e.g. rice. The LAPSUS model was not used at the mesolevel because of the lack of spatial data.
OUT1 Crop products
Crop production data were multiplied by the nutrient contents of the crops. Where available, local nutrient-content factors were used because these can differ significantly from the average continental values used at the macrolevel.
OUT2 Crop residues
Crop-residue removal factors were derived from farm surveys or estimated by local experts. These factors were multiplied by the production and the nutrient-content factors of the crop residues.
OUT3 Leaching
The calculations of the leaching of N and K utilized the macrolevel regression models.
OUT4 Gaseous losses
The estimation of the gaseous N losses utilized the macrolevel regression model.
OUT5 Erosion
Estimates of erosion were made for each crop. These estimates took regional differences in topography and soils into account and were based on literature and expert knowledge. The LAPSUS model was not used because of the lack of spatial data.
For Ghana, nutrient balances were calculated for two districts: Nkawie and Wassa Amenfi. The amount of data available in Ghana was small because there have been no studies on soil fertility and nutrient management in the cocoa area. Therefore, a significant amount of the data was derived from the macrolevel. A soil map (scale: 1:250 000) (Soil Research Institute, 1999) was available for both districts, but representative chemical characteristics of the soils were unknown. Therefore, the soil properties were derived from the WISE database (Batjes, 2002). Local weather stations were the source of the rainfall data, averaged for 1997 - 99. Nutrient balances were made for the following crops: cocoa, maize, cassava, plantain, cocoyam, yam, rice, vegetables and oil-palm. The area under annual crops and the macrolevel data provided the basis for estimating the area under fallow: 14 600 ha (13 percent) in Nkawie District, and 7 300 ha (2.7 percent) in Wassa Amenfi District. Annex 9 lists all the primary data.
TABLE 17
Harvested areas and manure for each crop
in two districts, Ghana
Crop |
Nkawie District |
Wassa Amenfi District |
||
Area (ha) |
% of manure |
Area (ha) |
% of manure |
|
Cassava |
11 838 |
15 |
7 700 |
15 |
Cocoa |
48 493 |
0 |
240 961 |
0 |
Cocoyam |
9 514 |
18 |
3 000 |
10 |
Fallow |
14 600 |
0 |
7 300 |
0 |
Maize |
11 455 |
50 |
5 650 |
40 |
Oil-palm |
- |
- |
900 |
0 |
Plantain |
11 725 |
15 |
5 000 |
10 |
Rice |
1 462 |
0 |
2 112 |
0 |
Vegetables |
- |
- |
250 |
20 |
Yam |
1 175 |
2 |
1 500 |
5 |
IN1 |
No fertilizer data available. Macrolevel data used. Only data for cocoa available. |
IN2 |
Livestock numbers known. Amount of available manure calculated. |
|
A loss factor of 50 percent was estimated because no active manure management takes place. Available manure allocated to crops as per Table 17. |
IN3 |
Macrolevel data used. |
IN4 IN5 |
No symbiotic and non-symbiotic N fixation according to the macrolevel equation. Irrigation not relevant. For rice, estimated sedimentation rate: 1 mm/ |
OUT1 |
year. Production data available. |
OUT2 OUT3 |
Macrolevel crop residue removal factors used. Calculated according to the regression models. |
OUT4 |
Calculated according to |
|
the regression model. |
OUT5 |
Estimated values per crop |
|
used (Table 18). |
TABLE 18
Estimated erosion per crop for
cocoa-based system
Crop |
Erosion (tonnes/ha) |
Explanation |
Cassava |
5 |
Stands two years, but during harvest high erosion risk |
Cocoa |
1 |
On slopes, but thick litter layer and stable root system |
Cocoyam |
7 |
In fresh fields and high erosion risk during harvest |
Fallow |
0.5 |
Permanent ground cover |
Maize |
5 |
Planted in fresh bare fields |
Oil-palm |
1 |
Relatively flat areas and stable root system |
Plantain |
3 |
Relatively stable root system |
Rice |
0 |
In valleys and therefore no erosion, but sedimentation |
Vegetables |
10 |
Near houses, more runoff, more bare soil |
Yam |
7 |
In fresh fields and high erosion risk during harvest |
TABLE 19
Soil properties for the upper AEZs in
Embu District
Soil properties |
AEZ1 |
AEZ2 |
AEZ3 |
pH |
4.4 |
4.5 |
4.5 |
N total (%) |
0.75 |
0.64 |
0.44 |
C total (%) |
3.3 |
2.5 |
2.3 |
P total (ppm) |
2 211 |
2 095 |
1 898 |
P Olsen (ppm) |
3.5 |
2.9 |
2.3 |
K exch. (cmol/kg) |
0.46 |
1.14 |
1.05 |
Bulk density (kg/dm3) |
0.80 |
1.20 |
1.55 |
Clay (%) |
35 |
60 |
70 |
CEC (cmol/kg) |
28.7 |
19.4 |
19.4 |
Source: Stoorvogel et al. (2000).
Many data were available for Kenya because it has been the subject of many soil fertility and nutrient management studies. However, not all of the data were useful because they were very scattered or based on only a few farms. The study area comprised the tea - coffee - dairy zone of Embu District, defined mainly by AEZ2 of the VARINUTS study (SC-DLO et al., 2000). The area consists mainly of Nitisols. A soil map of Embu District (Ministry of Agriculture, 1987) was available, but only one soil type was differentiated within the study area. This study utilized the soil properties as determined in the VARINUTS study (Table 19). Annex 9 lists all the primary data.
The nutrient balance was calculated for the following crops within the study area: tea, coffee, maize, beans, Napier grass, sorghum, cowpeas, cassava, potatoes, sweet potatoes and arrow roots. No data were available for Napier grass, so the study used a yield estimate of 35 tonnes/ha (Van den Bosch et al., 1998b), and for the harvested area a value of 3 percent of the total cultivated area was taken (Abdullahi et al., 1986). The fallow area was based on the non-cultivated area for the Manyatta and Nembure division, which is 24 km2. Of this area, 75 percent was estimated as fallow, the remaining area as buildings and forest. The resulting fallow area was 1 800 ha, which is 8 percent of the total cultivated area. The estimated production of fallow land was 10 tonnes/ha, of which 5 percent was removed for fodder and firewood.
IN1 |
Mineral fertilizer application based on Staverman (2003) for several crops (data were derived from 50 farm interviews in AEZ3 of Embu District). Recommended mineral fertilizer rates used for other crops (Table 20). |
IN2 |
Number of livestock known; amount of available manure calculated. Loss factor: 25 percent. Application factors based on Staverman (2003). |
IN3 |
Dry deposition not relevant; macrolevel data used for wet deposition. |
IN4 |
Symbiotic N fixation for beans, cowpeas and soybeans and non-symbiotic according to macrolevel equation. |
IN5 |
Irrigation not relevant; sedimentation occurs only in the flood lakes. |
OUT1 |
Production data available; local nutrient content factors of Van den Bosch et al. (1998) used. |
OUT2 |
Estimates for local crop-residue removal factors used. |
OUT3 |
Calculated according to the regression models. |
OUT4 |
Calculated according to the regression model. |
OUT5 |
Estimated values per crop, based on data from the LEINUTS study (De Jager et al., 2001) in Nyeri District, which is the same AEZ (Table 21). |
TABLE 20
Mineral fertilizer application rates,
Embu District
Crop |
N |
P |
K |
Reference |
(kg/ha) |
||||
Arrow roots |
0 |
0 |
0 |
No fertilizer |
Beans |
6 |
6 |
0 |
Staverman, 2003 |
Cassava |
0 |
0 |
0 |
No fertilizer |
Coffee |
59 |
13 |
0 |
Staverman, 2003 |
Cowpeas |
0 |
40 |
0 |
Recommended (200 kg/ha, TSP)* |
Fallow |
0 |
0 |
0 |
No fertilizer |
Maize |
18 |
8 |
2 |
Staverman, 2003 |
Napier |
33 |
6 |
0 |
Staverman, 2003 |
Potatoes |
45 |
50 |
0 |
Min. of Agr. and GTZ, 1998 (250 kg/ha, DAP) |
Sorghum |
10 |
4 |
0 |
Recommended (50 kg/ha, 20:20:0)* |
Sweet potatoes |
17 |
7 |
14 |
Recommended (100 kg/ha, 17:17:17)* |
Tea |
34 |
3 |
6 |
Recommended (135 kg/ha, 25:5:5)* |
TABLE 21
Estimated erosion per crop, Embu
District
Crops |
Erosion |
LEINUTS |
Explanation |
(tonnes/ha) |
|||
Arrow roots |
0 |
- |
Grown in valleys, equilibrium expected |
Beans |
7 |
7.2 |
According to LEINUTS study |
Cassava |
10 |
- |
Many erosion during harvest |
Coffee |
5 |
3.6 |
According to LEINUTS study |
Cowpeas |
7 |
- |
Similar to beans |
Fallow |
0.5 |
0.6 |
According to LEINUTS study |
Maize |
8 |
8.1 |
According to LEINUTS study |
Napier |
4 |
4.6 |
According to LEINUTS study |
Potatoes |
3 |
2.6 |
According to LEINUTS study |
Sorghum |
8 |
- |
Similar to maize |
Sweet potatoes |
5 |
4.3 |
According to LEINUTS study |
Tea |
1 |
0.8 |
According to LEINUTS study |
* Recommended fertilizer rates based on Ministry of Agriculture (1997 - 2001).
For Mali, the nutrient-balance calculation considered the CMDT Koutiala Region in Mali-Sud. The CMDT has undertaken many farm surveys in this area, which have also generated a lot of data relating to nutrient management. Therefore, a lot of high-quality data for important flows, such as mineral fertilizers, organic fertilizers and crop-residue removal, were available for this study area. However, spatial data were practically non-existent, only one handdrawn morphological map of the region was usable. The cultivated areas lie on the lower slopes with three different soil types, depending on the parent material (Sissoko, 1999). The study used an average of the soil properties of these soils (Table 22). The main crops in the study area are: cotton, maize, sorghum, millet, rice, groundnut and cowpea. These crops account for 97 percent of the total cultivated area. Research in two representative villages provided the basis for the estimate of the fallow area (Kanté, 2001). In these villages, 48 percent of the cultivated area was under fallow. The resulting estimate for the whole of Koutiala Region is a fallow area of 227 200 ha. Annex 9 lists all the primary data.
TABLE 22
Soil properties for the lower slopes in
Koutiala Region
Parent material |
Sandstone |
Schist |
Dolerite |
Average |
Clay (%) |
5.4 |
6.0 |
6.9 |
6.1 |
OM (%) |
0.6 |
0.5 |
0.7 |
0.6 |
CEC (cmol/kg) |
3.0 |
4.3 |
6.0 |
4.4 |
pH |
5.5 |
5.8 |
5.6 |
5.6 |
N total (%) |
0.03 |
0.04 |
0.03 |
0.03 |
P total (ppm) |
93 |
139 |
97 |
110 |
P ass. (ppm) |
5.7 |
3.0 |
3.5 |
4.1 |
K exch. (cmol/kg) |
0.03 |
0.03 |
0.03 |
0.03 |
Source: Bitchibaly et al. (1995).
IN1 |
Fertilizers applied: urea (46 percent N), cotton complex (14:9.6:10) and cereal complex (15:6.5:12.5). The amount of applied fertilizers per crop and percentage of fertilized fields was known (CMDT/SE, 1998, 1999, 2000). |
IN2 |
Data on organic fertilizer application available (SEP, 1997, 1998, 1999). |
IN3 |
Dry deposition and rainfall derived from the macrolevel. |
IN4 |
Symbiotic N fixation for groundnut (65 percent) and cowpea (55 percent) and non-symbiotic according to the macrolevel equation. |
IN5 |
Irrigation not relevant; sedimentation estimated at 1 mm/year for rice (grown in river valleys). |
OUT1 |
Production data available; local nutrient content factors of Kanté (2001) used. |
OUT2 |
Study conducted of crop residue (Table 23); local nutrient content factors of Kanté (2001) used. |
OUT3 |
Calculated according to the regression models. |
OUT4 |
Calculated according to the regression model. |
OUT5 |
Erosion estimated for each crop according to available literature. Average erosion on cultivated land (Table 24): 7 tonnes/ha (Bah, 1992; Vlot and Traoré, 1995). |
TABLE 23
Crop residue use for Koutiala
Region
Crops |
Fodder |
Litter |
Burying |
Grazing |
Burning |
Other |
Total removal* |
(%) |
|||||||
Cotton |
0 |
38 |
0 |
0 |
61 |
1 |
1 |
Cowpea |
83 |
0 |
0 |
15 |
0 |
2 |
100 |
Groundnut |
65 |
10 |
9 |
15 |
0 |
1 |
81 |
Maize |
44 |
11 |
9 |
36 |
0 |
0 |
80 |
Millet |
2 |
11 |
3 |
48 |
34 |
2 |
52 |
Sorghum |
2 |
20 |
1 |
48 |
28 |
1 |
51 |
* Total removal calculated as the sum of fodder, grazing and other.
Source: Camara (1996).
TABLE 24
Estimated erosion rates for Koutiala
Region
Crops |
Erosion (tonnes/ha) |
Explanation |
Cotton |
10 |
More than average erosion, because of burning after harvest |
Cowpea |
7 |
Average erosion |
Fallow |
0.5 |
Little erosion, because of permanent ground cover |
Groundnut |
10 |
More than average erosion, because of way of harvesting |
Maize |
5 |
Less than average erosion, as stubble is left on the fields |
Millet |
5 |
Less erosion, as stubble is left on the fields |
Rice |
0 |
No erosion, because rice is grown in the flat river valleys |
Sorghum |
5 |
Less than average erosion, as stubble is left on the fields |