African Crop Science Journal African Crop Science Society ISSN: 1021-9730 EISSN: 2072-6589 Vol. 4, Num. 4, 1996, pp. 503-518

Recognising and overcoming soil constraints to crop production in tropical Africa

P.L. WOOMER and F .N. MUCHENA^1

Tropical Soil Biology and Fertility Programme, PO Box 30592, Nairobi, Kenya
^1 ETC-Kenya Consultants, P.O. Box 76378, Nairobi, Kenya

(Received 24 November, 1995; accepted 21 September, 1996)

Code Number: CS96093
Sizes of Files:
    Text: 57.5K
    Graphics: Line drawings (gif) - 36.8K

ABSTRACT

We have reviewed recent literature and research findings which address the types and coverage of soil constraints to crop productivity and the available management options with which these constrains are being overcome. The soil constraints fail into four broad categories, nutrient limiting, toxic, moisture restrictive and highly eroded. These constraints contribute to Africa's dilemma of population growth outpacing agricultural productivity. The African continent is approximately 30 million km^2 in size, 19% of which has been converted to agriculture. Of these agricultural lands, 34% may be considered as having no or slight soil limitations while severe nutrient deficiencies and toxicities occur within 32 % of them. These severe chemical stresses occur primarily on Oxisols and Ultisols. Other shallow and coarse textured soils present difficulties to water management and cover approximately 1.1 million km^2 of cultivated lands. The strategies necessary to overcome these soil constraints include better exploitation of the region's agromineral deposits, adjustments to fertilizer taxation, improved soil conservation including better maintenance of soil organic matter thereby improving nutrient use efficiency of organic and fertilizer inputs, further development of small-scale irrigation projects, dissemination of stress-tolerant crop varieties and finding new and better ways to work with farmers.

Key Words: Agricultural development, land use planning, nutrient availability, soil erosion, soil moisture

RESUME

Nous avons passe en revue les ecrits traitant de differents types de sol et de l'ensemble des informations sur les elements contenus dans le sol qui pourraient contraindre sa productivite, et sur les options d'encadrement disponibles grace auxquelles on pourrait venir a bout de ces limites a la productivite. En gros, on pent reunir ces elements qui reduisent la productivite du sol en quatre categories: ceux qui limitent la valeur nutritive des produits, ceux qui sont toxiques, ceux qui diminuent l'humidite et ceux qui sont fortement affaiblis par l'erosion. Ces elements qui limitent la productivite du sol contribuent au dilemme africain: la contradiction entre l'accroissement de la population et la productivite agricole qui le devance. le continent africain a approximativement une superficie de 30 millions de km^2 dont 19% ont ete rendus cultivables. On peut considerer 34% de ces terres cultivables comme etant entierement ou failblement productives tandis que 32% contiennent des elements toxiques et presentent des deficiences en elements nutritifs. Ces contraintes chimiques severes s'exercent essentiellement sur les oxisols et les Ultisols. D'autres types de sol a la texture peu profonde et grossiere presentent des difficultes quant a l'amenagement des eaux, et couvrent approximativement 1, 1 million de km^2 de terres cultivees. Les strategies necessaires pour venir a bout de ces elements qui refrenent la productivite des terres cultivables comprennent une meilleure exploitation des depots agromineraux de la region concernee, des ajustements aux charges riscales concernant les engrais, une amelioration de la conservation des sols qui comprend un meilleur entretien de la matiere organique du sol, et de ce fait ameliorant l'efficacite de l'utilisation des elements nutritifs des apports de matieres organiques et d'engrais, un developpement plus avance des projets d'irrigation sur une petite echelle, une dissemination des varietes de cultures qui tolerent toutes sortes de contraintes, et trouvant de nouveaux et meilleurs moyens de travailler ensemble avec les cutivateurs.

Mots Cles: Developpement agricole, disponibilite des elements nutritifs, humidie du sol, erosion du sol, amenagement des terres

INTRODUCTION

Large gaps exist between developed and developing nations in terms of agricultural productivity and the quality of life enjoyed by rural populations. Improvements in the quality of life in many developing nations may bc viewed as a race between agricultural production and population growth. In sub-Saharan Africa this race is being steadily lost because the modest gains in production are being more than offset by the rate of population growth (FAO, 1992). This relationship is illustrated in Figure 1 where decline in per capita agricultural productivity results from an imbalance of production and population between 1980 and 1990.

Smaling (1993) identifies four root causes which underlie this productivity gains gap in Africa; natural catastrophe, civil unrest, failure of the "Green Revolution" and continuous soil depletion. The frequency of natural disasters has been great during the past decade. These catastrophes include severe droughts in Sahelian countries, Ethiopia and Southern Africa, floods in Sudan and locust invasions. Perhaps more serious are the human induced disasters which accompany prolonged civil wars and tensions that weaken agricultural infrastructure. Such conditions have recently occurred, or continue, in Angola, Mozambique, Rwanda, Ethiopia, Uganda, Zaire, Somalia and Sudan. The "Green Revolution" which only a decade ago was viewed a promising vehicle to productivity gains has met with less than modest success in Africa. Few farmers are planting improved germplasm and fewer still are able to purchase the external inputs such as fertilizers and pesticides that are often necessary to realise the potential gains of improved varieties and hybrids (Okigbo, 1990; Woomer, 1993). The failure of the "Green Revolution" in itself has very complex causes, many of which are the same as those preventing alternative pathways to development (e.g. lack of agricultural infrastructure, poor market development, traditional land tenure and labour customs). Coupled with the by-pass of the "Green Revolution" is the decline in world prices of tropical agricultural commodities such as cacao, sugar and coffee.

Figure 1. Changes in population, agricultural productivity and per capita productivity in Africa between 1980 and 1990 (FAO 1992, 1995a).

The purpose of this concept paper is to provide some background to soil constraints to intensified agricultural production in Africa and to launch a discussion on improved mechanisms to overcome those constraints. In order to accomplish these objectives, we will first introduce the concept of the soil as a resource base for plant productivity, move to those constraints which limit the effectiveness of that resource base, provide a continent-wide evaluation of recent land use and soil conditions within Africa and then suggest several options that serve to more effectively utilise and maintain soils, relying primarily on examples from East and Southern Africa developed by the Kenya Agricultural Research Institute (KARl) and the Tropical Soil Biology and Fertility Programme (TSBF).

THE SOIL AS A RESOURCE BASE

Soils form over geological time based upon the weathering of parent material and are influenced by organisms, climate, topography and the movement of materials into and from the soil system. Plants rely on the soil system in several, interacting ways. Plant roots penetrate and explore the soil in order to acquire water, oxygen, nutrients and structural support for above-ground growth. It is when a plant is unable to meet a requirement provided by the soil resource base that the soil may be considered to be constraining. The nature of a particular soil as a resource base predetermines what sort and magnitude of soil constraints to improved crop productivity that soil may pose. The principal soil constraints may be grouped into four broad categories; nutrient availability and retention, soil toxicities, water availability and physical degradation due to erosion. The soil constraints to productivity are summarised below and their geographic distribution within the African continent described later in this paper.

Nutrient availability and retention.

Soils serve as both a reservoir and source of plant nutrients. The inherent fertility of soils is associated with mineralisation of soil organic matter (SOM). Decline in total soil organic matter may often be a root cause of nutrient exhaustion in farming systems provided with few external nutrient inputs (Sanchez, 1976). The most frequently limiting nutrients in soils of Kenya are nitrogen and phosphorus (Mochoge, 1993). The economic return to investment in recommended fertilisation rates and difficulty in using soil classification data alone to anticipate the response to N and P fertilisation is illustrated in Table 1 for several locations in Kenya. Similar results were obtained by the Fertilizer Use Recommendation Project (FURP, 1994) during several years of fertilizer response trials where very different recommendations are delivered for similar soils. While the inherent capacity of a soil to provide plant nutrients may be supplemented through the application of chemical fertilizers, less fertilizers are being applied per unit land area in Africa than in other regions of the world (see FAO, 1989) despite Africa having become a net exporter of fertilizers (Fig. 2).

TABLE 1. Maize yields under low external input and fertilised conditions and the return on investment to fertilizers at several on-station and on-farm experiments in Kenya (J.R. Okaiebo, personal communication)

---------------------------------------------------------------------------
Site                                   Farmer  Potential  Necessary  Return 
                      Soil           low input high input  inputs   ratio^1

           -----------------------      yield     yield                     
          FAO/UNESCO   USDA (1975)     (per ha) (per ha)  (per ha)
---------------------------------------------------------------------------
Katumani     Ferral    Oxic              2.4       4.1     60 kgN+     4.3 
Station      chromic   Paieustalf 
             Luvisol                                       20 kgP 
Kampi ya     Chromic   Udic Rhodustalf   2.6       3.2     60 kgN +    1.8 
Mawe Station.Luvisol                                        0 kgP 
Mutua Farm   Haplic    Affisol           2.8       4.5     90 kgN +    3.1 
             Lixisol                                       15 kgP 
Kathonzweni  Lixisol   Affisol           0.5       1.9     60 kgN +    3.0 
 Farm                                                      40 kgP 
Kyengo Farm, Haplie    Ultisol           0.6       4.9     90 kgN +    7.6 
 Wamunyu     Alisol                                        20 kgP 
Kombo Farm   Haplic    Ultisol           1.8        3.0    90 kgN +    2.2 
             Alisol                                        10 kgP 
Mweu Farm    Haplic    Affisol           2.4        3.5    90 kgN +    2.0 
             Lixisol                                       15 kgP 
Kitale       Orthic    Typic             5.0       11.2    60 kgN +   13.5 
 Station     Ferralsol Haplustox                           40 kgP 
Joy Farm    Acri-orthic  Oxic            2.8        3.0    60 kgN +    0.4 
             Ferralsol Rhodustalf                          60 kgP 
Kisasi,     Acri-orthic  Oxic            2.7        4.0    60 kgN +    2.8 
 Kitui       Ferralsol Rhodustalf                          40 kgP 
Likhorero    Acrisol   Ultisol           3.3        6.5    100 kgN +   4.3
 School
Katheka-     Acrisol   Ultisol           2.2        2.9    60 kgN +    2.1 
  Kabati     (Orthic)                                       0 kgP 
Kakamega     Eutric    Typic             5.7        7.0   100 kgN +    1.9 
  Station    Nitjsol   Paleustalf                          40 kgP 
Malava Farm  Eutric 
              Nitisol  Typic             2.2        5.5   120 kgN +    3.2 
           (Ferralsol) Paleustalf                         120 kgP

---------------------------------------------------------------------------

Another important mechanism to improved nutrient cycling is through the use of applied organic inputs and the retention of crop residues (Sanchez et al. 1989). Yet in many tropical cropping systems, little or no agricultural residues are returned to the soil leading to decline in SOM (Lal, 1986; Bouwman, 1990b; Post and Mann, 1990 Woomer and Ingram, 1990) which frequently accompanies lower crop yields (Lal, 1986) or plant biomass productivity (Woomer and Ingram, 1990). Application of organic residues and the avoidance of burning are important management practices that play an important role in reversing these trends along with the conversion to lower tillage systems (Haines and Uren, 1990) and the establishment of live mulches (Lal, 1986). In addition to agronomic issues, carbon storage in soils has achieved global significance due to increases in atmospheric, carbon-based greenhouse gases (Bouwman, 1990a, b).

Nutrient retention results from ionic attraction between plant nutrients in solution and the charged surfaces within the soil. These charged surfaces occur in both the mineral and organic fractions although in highly weathered tropical soils, there is a greater dependence on the role of organic because of a decreased charge density of mineral oxides. This is particularly the case in the Oxisol-Ultisol-Inceptisol sequences found on the oldest parent materials (Sanchez, 1976). As illustration of the importance of soil organic matter, the relationships between total soil nitrogen and cation exchange capacity with total soil organic carbon in a number of East and Southern African soils are presented in Figure 3. The importance of the relationship between CEC and SOM in low activity clays (Oxisols and Ultisols) is the opportunity for management whereas the lack of higher charged clays cannot be rectified. An important factor in continuous productivity of tropical soils is the maintenance and improvement of soil physical characteristics. Once this is achieved, production capacity of these soils can be further improved by the use of organic and inorganic fertilizers.

Figure 2. Fertilizer consumption, production, exportation and importation by Africa between 1977 and 1993 (FAO. 1989; b).

Soil toxicities.

Toxic levels of aluminum species and other cations (particularly Fe and Mn) occur in highly weathered, low pH soils that are dominated by oxide mineralogy (Russell, 1973) where high concentrations of these cations interfere with nutrient uptake and, when entering the plant, interfere with sugar phosphorylation and DNA synthesis. One common symptom of cation toxicity is the development of swollen, stubby roots (Sanchez, 1976) resulting from an inhibition of root elongation. A direct amelioration of aluminum toxicity may be accomplished through the application of lime. Sanchez (1976) recommends the application of 1.65 T ha^-1 per milliequivelent of exchangeable aluminum based on the solubility oral cations at different soil pH levels. This management is not always feasible, however, due to the unavailability of lime to many subsistence and local market farmers in the humid tropics, and to the frequency of high Al saturation in Oxisols and Ultisols where exchangeable Al may reach 3.1-4.2 cmol kg^-1 of soil within the tillage layer (see Sanchez, 1976). Computer software programmes are available to assist in the management of acid soils. The Acid Decision Support System (Anon., 1989) is an expert system based on the replacement of toxic cations by liming and which takes into account the differential sensitivity of crop species to the effects of aluminum toxicity prior to generating liming recommendations. An earlier version of this programme, ACID4, was used to assess liming requirements in the highland region of Rwanda (Yamoah et al., 1990).

Soil organic matter interacts with toxic cations in two different manners, both to the benefit of plants. Humic substances absorb toxic cations, resulting in their immobilisation and detoxification (Woomer et al., 1994). Organic acids from decomposing residues in soils also complex with AI in solution resulting in less toxic forms oral in the soil solution, again without changing the soil pH (Hargrove and Thomas, 1981; Hue et al, 1986, Bell and Besho, 1993). While management may be technically feasible at the small farm level, large application rates of organic residues are required to detoxified acid soils. Furthermore, the relative effectiveness of most farmer-available organic resources in Africa remains unknown. Relatively simple plant bioassays are available to measure the toxicity of test soils by comparing the root elongation of recently germinated mung bean seedlings (Hue et al. 1986; Bell and Besho, 1993) which could enable rapid testing of the efficacy of ameliorative practices.

Figure 3. The relationships between total soil carbon and nitrogen (above) and cation exchange capacity (below).

Water flow within the soil/plant system may be viewed as a single directional flow in which roots absorb moisture from the soil, this water is passed from the roots to the stem and leaves through a series of resistances and then finally evaporated from leaf stomata. The availability of soil moisture to plants is a function of water inputs, moisture retention and rooting depth. In general, roots are readily able to absorb soil moisture at field capacity, 0.1-0.5 bar, depending on mineralogy and soil structure, and become less able to do so until 15 bar is reached, referred to as the wilting point (Russell, 1973), although the universality of this relationship is challengeable (Sanchez, 1976). At a farm level, however, it may be said that unless irrigation is possible, water inputs result from precipitation and are beyond the control of farmers.

Moisture retention is both an intrinsic property of soils and subject to management. The pores of sandy soils are emptied of gravitational water at 0.1 bar, while silicate layered clayey soils retain this moisture until 0.5 bar (Sanchez, 1976). Other soils, including most of those in the tropics, fall somewhere between. Farmers manage moisture retention in many ways, the most obvious and important one being the reduction of water run off along the soil surface by terracing, contour ridges and other more elaborate water capture strategies. Added benefits to runoff reduction are control of nutrient losses and less soil erodibility. A key to reduced runoff and its consequent benefits is protection of the soil surface with mulch. Nill and Nill (1993) demonstrated that 60% surface cover of Guinea Grass leaves reduced runoff by 60% and controlled soil erosion in Southern Cameroon.

The depth of rooting is often an overlooked component to moisture availability and one that need not be universally associated with shallow soils. The ability of plant roots to extract moisture reserves from deeper soil layers may be inhibited by the inability of plant roots to penetrate to that depth by physical and chemical barriers (Sanchez, 1976). Little can be done to improve rooting depth in Leptosols or extremely rocky soils.

Hardpans develop in clayey soil immediately beneath the shallow tillage layer due to compaction from mechanical tillage. The formation of hardpans may be disrupted by occasional deep tillage. Acidic subsoils may limit a roofing system's ability to recover moisture reserves. This is a common limitation in Oxisols and Ultisols where moisture in the well structured surface horizon is depleted but abundant moisture in the acidic subsoil remains unexploited and moisture stress ensues. This problem has various solutions ranging from deep tillage and liming to the use of more acid tolerant, tap-rooted crops and cultivars. In many cases agricultural researchers and farmers are unaware of roofing limitations. Therefore, profile walls of root distribution as an experimental measurement must become more frequent before we can place the magnitude of the root penetration component as a problem into better perspective.

Soil erosion.

Erosion results in degradation of soil physical characteristics such as infiltration rate, soil structure and crusting. The physical removal of surface soil is serious for shallow gravelly soils, but more so are the resulting effects of poor soil physical conditions on crop production. Soil erosion also decreases the fertilizer use efficiency by increasing the nutrient losses. Soil erosion is seldom a direct constraint to an individual cropping cycle but rather a chronic depletion of the soil as a resource base. Soil erosion implies the transport of surface soils from one location to another. The principal agents of erosion are water and wind. The principle causes of soil erosion are deforestation, overgrazing, shortened cycles of shifting cultivation (Sanchez, 1976) and cultivation of slopes in absence of conservation. The annual losses from some soils in West Africa under different land uses are presented in Table 2. Note that small soil losses are observed under natural vegetation but as lands are cultivated and left in bare fallow the rate of soil loss increases. The protection afforded by natural vegetation from erosion includes protection from direct impact of rainfall by the canopy, presence of a surface litter layer to further buffer droplet impact and to impede runoff and abundant rooting within the soil surface horizon which stabilises soils. The importance of carbon cycling within an ecosystem must not be overlooked. As surface and root litter decays, organic materials assist in the formation and stabilisation of soil aggregates (Oades, 1984) which resist erosive forces (Sanchez, 1976). Soil structure and the distribution of macropores are further facilitated by increased rooting (Lal, 1986) and macrofaunal activity (Lavelie, 1988). The resilience of natural ecosystems vs improperly managed agroecosystems lend credence to the assertion of Swift and Anderson (1992) that the sustainability of a cropping system is promoted by maintaining plant species of different architecture and chemical composition.

TABLE 2. Annual soil erosion losses in four localities of West Africa (tons ha^-1) under different land uses (from Charreau, 1972 and Sanchez, 1976)

---------------------------------------------------------------------------
Locality                Forest land    Cultivated land    Bare soil
---------------------------------------------------------------------------
Ouagadougou, Upper Volta    0.1            0.6-8.0         10-20 
Sefa. Senegal               0.2              7.3             21 
Bouake, Senegal             0.1            0.1-26          18-30 
Abldjan, Ivory Coast        0.03           0.1-90         108-170
--------------------------------------------------------------------------

DISTRIBUTION OF SOIL CONSTRAINTS WITHIN AFRICA

The distribution and coverage of land use systems within Africa are presented in Table 3. The geographical distribution of cropland/fallow systems and the two major natural vegetation types from which croplands are derived is presented in Figure 4. Both Table 3 and Figure 4 were obtained from the Resource Information System (RIS), a geo-referenced data base under development by the International Institute of Tropical Agriculture's Agroclimatology Laboratory (Jagtap, 1991). The soil principle constraints to plant productivity and their coverage and land areas are presented in Table 4. Soils of modest fertility and which occur in favourable climates for agriculture are assigned a none to slight rating (Table 3) although these areas are subject to periodic conditions of unfavourable climate or to land degradation, so this category may be over estimated. Highly weathered soils are considered to pose severe constraints. Soil texture, depth, physical properties and climatic conditions are used to assign the other constraint types. Table 4 relates the predominant soil stress conditions to their respective FAO and USDA classification systems and provides the coverage of those soils. The soil properties which contribute to the various constraints are also included in Table 4. Figure 5 maps the distribution of agricultural soils; including fallows, in Sub-Saharan Africa, dividing them into areas with no or slight soil constraints (Table 3), chemical constraints (nutrient and toxicity constraints in Table 3) and a composite of physical limitations such as coarse, shallow and cracking soils (Table 4). This map was generated from the RIS (Jagtap, 1991). The major soil constraints to cropland are further subdivided by soil order, their respective land areas and the generalised plant stress conditions are presented in Table 5. Note that RIS soil classification is based on the FAO/UNESCO system (FAO, 1974, 1990) and the assignment of USDA Classification is only approximate and based on Sanchez (1976). Also note that the scale of the maps presented in this section is quite broad and must not be used to identify a precise limiting condition.

TABLE 3. Land area and coverage of land use systems within Africa (after Jagtap, 1991)^1

--------------------------------------------------
Land use                    Area (km^2)   Area (%)
--------------------------------------------------
Savanna and Dry Woodland    11,167,000    37.5 
Desert and Salt/Soda Lands   9,298,000    31.2 
Cropland and Fallow          5,649,000    19.0 
Forest                       2,644,000    9.5 
Marsh and Swamp                625,000    2.0 
Other                          420,000    0.8

Total                       29,803,000    100

^1 Data excludes lakes and includes neighbouring islands.
-------------------------------------------------------------

Figure 4. Coverage of forests (light shade), cropland/fallows (intermediate shade) and savannas (darkest shads) in Africa.

---------------------------------------------------------------------------
Soil constraint             Area (km^2)  Area (%)    Plant stress type
---------------------------------------------------------------------------
None/slight                  1,904,000    33.7      varies w/management 
Severe nutrient/toxicity     1,783,000    31.6      nutrient availability
Coarse texture/gravelly        789,000    14.0      physical (nutrient) 
Desert (irrigation dependent)  326,000     5.8      moisture 
Poor drainage                  310,000     5.4      oxygen availability 
Shallow                        288,000     5.1      physical 
Heavy cracking                 243,000     4.3      physical 
Salinity                         6,000     0.1      osmotic interference
---------------------------------------------------------------------------
Total                        5,649,000     100
---------------------------------------------------------------------------

Figure 5. Coverage of soils with no or slight constraints (lightest) and severe physical (darkest) or chemical limitations (intermediate shade) to plant growth in croplands of Africa.

---------------------------------------------------------------------------
Stress    FAO Order    USDA Soil   Area (km^2)  Soil properties 
type                   Taxonomy^a 
---------------------------------------------------------------------------
Slight    Luvisols^b   Alfisols    2,601,000    Partly weathered,           
                                                moderately fertile 
          Nitisols     Alfisols      990,000    Partly weathered,           
                                                moderately fertile 
          Cambisols    Inceptisols   941,000    Slightly weathered. fertlie

Chemical  Ferralsols   Oxisois     4,461,000    Highly weathered, infertile

          Acrisols^c   Ultisols      965,000    Highly weathered,           
                                                infertile, often toxic 
Physical  Arenosols    Psamments   2,402,000    Coarse texture, often       
                                                infertile 
          Lithisois    Lithic        855,000    Shallow rooting depth 
                        subgroups
          Vertisols    Vertisols     740,000    Cracking when dry, poor     
                                                drainage 
Moisture  Xerosols^d   Aridisols^e   171,000    Aridic, often gypsic or     
                                                calcic 
          Yermosols    Aridisols^e   155,000    Aridic, often gypsic or     
                                                calcic    

SOIL SURVEY, RESOURCE MANAGEMENT AND LAND USE PLANNING

Development planning attempts to devise programmes and projects to improve the utilisation of resources, including human, land, economic and technical resources, to achieve increased production, greater efficiencies, higher- quality products, better environmental conditions or some combination of desirable goals (Knox, 1981 ). Much development planning is concerned with land use and management for agricultural, silvicultural or other vegetation-based land use. Sound decisions for this planning require the information and predictions supplied by soil surveys. An appraisal of the soil resources of Africa has been provided by FAO/UNESCO Soil Map of the World at a scale 1:5 million (FAO 1974, 1990). The increasing need for food and agricultural products has put ever increasing demand on soil resources as the production base. Although expansion of the cultivated area is still possible in parts of Africa for increased food production, good soils are unevenly distributed and most of the soils in Africa are fragile and subject to degradation and erosion. Some of the soils present serious constraints to cultivation and require special management techniques. An inventory of soil conditions in sub-Saharan Africa illustrates that the major constrains vary across sub-regions with moisture availability being most common in East, Southern and West Africa, and nutrient availability most important in Central Africa (Sant-Anna, 1985).

Susceptibility to deterioration.

Susceptibility to deterioration is a very important soil quality and needs to be considered early in development planning. Erosion is the most obvious and serious hazard resulting from agricultural or other vegetation-based land uses. Susceptibility to erosion of soil types occurring in an area can be derived from soil survey information. Soil surveys also provide information on hazard restricted to specific soils, for example presence of limiting layers for root development such as hardpans etc., salt and sodium accumulation in soils irrigated without adequate drainage and subsidence of organic soils (Histosols).

Suitability for management operations.

From soil surveys, interpretations can be made with regard to the suitability of soils for management operations such as tillage either by hand labour, animal traction or motorised machinery. Information used for such evaluations and which are collected during soil surveys include slope, soil depth, rock outcrops, surface stoniness, drainage condition, soil consistence, etc. The development of irrigated agriculture very often demands large-scale investment. In addition to the cost of locating, accumulating and distributing water, heavy expenditure may be involved in land preparation and in the provision of drainage to reclaim or guard against soil affected by salinity or waterlogging. In order to attract the investment capital required, areas earmarked for irrigation projects must be studied to provide potential investors with a range of interpreted data sufficient to assess the risk and potential profit of investment. Soil surveys make an essential contribution to the development of the interpretative classification of land suitability for irrigation.

Soils information is important in planning efficient irrigation systems. Knowledge of the characteristics of individual soils is essential for planning the economic use of water. It provides a basis for ensuring that development is concentrated on the most suitable soils available and for determining the method and quantity of water application that will achieve optimum efficiency. It is important also that water application should be planned in relation to data on soils, particularly subsoil characteristics, if the hazards of salinisation and waterlogging are to be avoided.

Insufficient study of soils have in the past led to unsuccessful irrigation schemes. A good example is the 400 ha (Block C) of the Kimorigo/Kamleza Irrigation Schemes (Taita-Taveta District) which failed due to mounting problems of increasing salinity and sodicity, silting of the drains through continual flooding and the variations in ground water levels (Kanake, 1982). Had the soil investigations been carried out, these potential risks would have been identified.

Response to management.

Notwithstanding the importance of climatic parameters (rainfall, temperature, radiation, evaporation) in determining the potential of an area for agricultural production, the prevailing soil conditions determine what the land can be used for. It is therefore, important to know the variations of soils within the farm and how these differences relate to the envisaged use and management. Soil survey information would facilitate extrapolation of results and experiences gained from one area to similar areas where experiments have not been conducted. This would ensure that the available knowledge is used without unnecessary duplication.

Constraints on land use.

Soil surveys also give information on major limitations of land use. Some of the limitations may be of permanent nature, that is, they cannot be changed by management, for example soil depths, texture and slope. Others may be of temporary nature and can be corrected by management, for example soil fertility. This information is important for making decisions on the correct mode of development which takes into consideration the resource constraints. To facilitate storage and retrieval of soil survey information for planning and resource management, the Kenya Soil Survey have established a Geographical Information System (GIS). This GIS is also being used for digitising soil maps and preparation of interpretation maps for various land use alternatives.

Land use planning.

The function of land use planning is to guide decisions on land use in such a way that the resources of the environment are put to the most beneficial use for man whilst at the same time conserving those resources for future use. This planning must be based on an understanding both of the natural environment and the kinds of land use envisaged.

Agriculturalists in Africa are faced with some problems among which is the restricted access to good agricultural land where per capital size is continually getting smaller with the increasing population growth while the demand for agricultural products is continually rising. Consequently, there is a need to intensify land use. The strategy for this should aim at achieving high yields per unit of land. This may not be sustained through production based on the natural fertility of the soil types in Africa. Increased output of agricultural products can only be achieved if nutrients removed from the fields with the harvests are continuously replaced. Studies that can provide adequate, reliable and up-to-date data which can be used to quantify nutrient removal and fertilizer requirements need to be carried out. An example of such studies is the work carried out by the Fertilizer Recommendation Project (FURP) of the Kenya Agricultural Research Institute.

Sustenance of productivity of land calls for careful planning for utilisation and management of the soil and water resources. To achieve the most rational plans, land evaluation is called into play. Land evaluation is concerned with the assessment of land performance when used for specified purposes. It involves the execution and interpretation of basic surveys of soils, climate, vegetation, and other aspects of land and its productivity in terms of requirements of alternative forms of land use. The major objective of the land evaluation is to put at the disposal of the user, whether a planner, a farmer, a politician or extension worker or any other government official, relevant information about land resources that is necessary for planning and management decisions. The end result of land evaluation is thus a number of appropriate types of land use alternatives . together with their consequences on implementation. Such an undertaking is essential to facilitate identification and selection of the most appropriate options for sustainable soil productivity under intensive African agriculture.

OVERCOMING SOIL CONSTRAINTS TO INTENSIFIED AGRICULTURE

Fertiliser policy. As was stated earlier, the nations of Africa are net fertilizer exporters but the farmers of Africa lag far behind other regions in fertilizer use. Higher yield may only be sustained through the judicious use of the proper fertilizers or by the application of nutrient-rich organic resources. Intensified agricultural production is best achieved through the use of fertilizers and calls for the production and importation of the correct forms and quantities for timely distribution. These fertilizers must be supplied at reasonable cost and packaged in amounts suitable for use by small farmers. Fertilizer recommendations should be based on tested information which takes into account individual crop requirements and soil conditions. An example of such recommendations is the Fertilizer Use Recommendation Project which has improved the blanket recommendations for crops in Kenya to those for representative soil types occurring within individual districts in Kenya (FURP, 1994). Those countries within the region that impose high tariffs on fertilizer imports and which control fertiliser prices should consider liberalising those policies. Those countries that are exporting raw fertilizers should consider refining these fertilizers within the country. The need and effectiveness of subsidised fertiliser production and use should be reevaluated. At the same time, agriculturalists investigating crop responses to fertilizers must take prevalent economic conditions into account when designing and interpreting their experiments. An example of the return ratio obtained from several maize fertilisation experiments in Kenya is presented in Table 1. Note that the return to what might otherwise be recommended fertilizer rates varies greatly, including those obtained from on-ram research. This is due in part, to the over- generalisation of fertilizer recommendations. Another important policy dimension is the better mobilisation of under-exploited agromineral deposits, particularly rock phosphates ( McClellan, 1991; Van Kauwenberg, 1991).

Improved nutrient use efficiency.

The economic return to the use of fertilizers must be improved. This may be accomplished by identifying the form, rate of application, timing and placement that offer the highest rate of economic return to farmers. Too often, the results of fertilizer use trials are viewed in terms of their immediate crop return rather than as an agroecosystem-wide resource input that interacts with, and may be optimised through nutrient recycling processes. The use of under-utilised, farmer-available organic resources as a means of providing nutrients to the crop and improving fertilizer use efficiency warrants greater attention of agricultural scientists. There is a pressing need to better understand the soil as a resource base and to relate these understandings to soil inventory and land evaluation methods.

Continued and more effective soil conservation. Soil and water erosion pose a serious threat to the productive capacity of agriculture to keep pace with and, hopefully, exceed population growth. Cultivation of steep, non-terraced hill sides are the~most susceptible to erosion but, given the irregularity and intensity of precipitation in semiarid and sub-humid areas, even moderate slopes are often in jeopardy. Too often bench terraces, grass strips, hedgerows, windbreaks, mulches and contour ditches are not only under-utilised, but unknown within the cropping systems. One difficulty with farmer appreciation of the importance of soil conservation is that the return on labour and investments are not immediately evident and too often beyond the planning horizon. It is only when land has become severely degraded that those farmers who did not practice soil conservation become aware of the consequences by comparison with those farmers who did. Agricultural resource scientists and planners must better familiarise themselves with the labour requirement and capital expense of candidate soil conservation practices, and prepare farmer information packages which recommend socially acceptable and technically feasible options to the farmer.

Maintenance of soil organic matter. Soil organic matter is a key resource owing to its ameliorative effect on nutrient supply, detoxification of harmful soil constituents, moisture and nutrient retention and its role in soil structure formation. As a result of these properties, decline in SOM is often associated with decreased yields, even when large amounts of external inputs and energy are applied to the system. The level of organic matter within a soil is determined by its rates of formation and loss, yet the effect of applying many farmer available organic resources and agro-industrial wastes remains poorly understood from the perspective of organic matter dynamics. Recently, carbon sequestration in agricultural lands have assumed a global importance as adjacent natural ecosystems are destroyed and atmospheric CO2 increases. Studies of organic matter dynamics during the conversion of natural to managed, and within those managed systems over time must be initiated as a means to develop soil conservation measures that reduce carbon loss from soils. These studies must be further consolidated into improved management practices that are acceptable to farmers.

Development of environmentally sound irrigation projects.

Fresh water resources are under-utilised in many of the areas where the length of the cropping system is limited by the precipitation pattern or are susceptible to periodic drought but are of otherwise excellent potential for agriculture in terms of soil fertility and radiation. Additional benefits to the development of irrigation schemes are the opportunities for energy generation and rural electrification. Two hazards to water development by large-scale water conservation and irrigation projects are siltation of reservoirs and salinisation of irrigated lands over time, often with horrific environmental effects. While national scientists and the private sector are seeking solutions to these problems, a possible alternative is the support of smaller scale, farm-community managed projects.

Development and dissemination of stress resistant crop varieties.

The advancements by plant breeders in international and national centres often exceed the ability of extension and the private sector to multiply and disseminate improved plant genotypes. At the same time, plant geneticists often focus their breeding objectives on crop yields under optimal growth conditions rather than selecting and breeding plants for resistance to environmental stress. We recommend that extension specialists and Non Governmental Development Organisations seek to better familiarise themselves with the opportunity for the introduction of available improved plant germplasm and that agricultural scientists take greater advantage of the potential for overcoming soil constraints through the importation and testing of exotic plant lines.

Better understanding of and working with farmers.

Too often, agricultural planners and scientists forget that the farmers best understand their own lands and objectives. While national policy and top-down agricultural development strategies have their place, these may only be implemented through the active participation of farmers. It is the farmers who mobilise their available resources and take risks in order to assist their crops to overcome soil constraints to productivity in anticipation of a fair return to this effort and risk. Many farmers are acutely aware of land degradation but their priorities are, by necessity, food production and income generation during the current or next cropping cycle rather than in the more distant future. This dilemma between short term household security and longer-term conservation issues highlights the need for sound policy support of the soil as a resource base. Agricultural scientists are encouraged to visit farms and talk with farmers at the earliest stage of project formulation rather than only when a "new technology" is released for "on-farm testing".

ACKNOWLEDGEMENTS

The authors sincerely thank Dr. S. Jagtap of the International Institute of Tropical Agriculture for providing the Resource Information System to the authors, Dr. J .R. Okalebo previously with the Kenya Agricultural Institute and currently on the Faculty of Agriculture, Moi University for sharing the results of numerous fertiliser trials in Kenya, Mr. Gichuki Karanja for assistance in developing the East and Southern Africa soils data base and Mr. A. Ngaruiya and Alice Ndung'u for assistance in manuscript preparation.

REFERENCES

Anonomous. 1989. The Acid Decision Support System (ADSS, Computer Software). Centre for Soils Research, Bogor Indonesia and USAID Tropical Soils Project, University of Hawaii Department of Agronomy and Soil Science and North Carolina State University. Honolulu, Hawaii, USA.

Bell, L.C and Besho,T. 1993. Assessment of aluminum detoxification by organic materials in an Ultisol using soil solution characterisation and plant response. In: Dynamics of Organic Matter in Relation to Sustainability of Agricultural Systems. John Wiley & Sons, pp. 317-330.

Bouwman, A.F. 1990a. Land use related sources of greenhouse gases. Land Use Policy. Butterworth and Company, London. pp. 154164.

Bouwman, A.F. 1990b. Exchange of greenhouse gases between terrestrial ecosystems and the atmosphere. In: Soils and the Greenhouse Effect.

Bouwman, A.F. (Ed.), pp. 59-127. John Wiley and Sons Ltd., West Sussex, England.

Charreau, C. 1972. Problemes poses par l'utilisation agricole des sols tropicaux par des cultures annuelles. Agronomique Tropicale (France) 27:905-929.

Food and Agriculture Organisation of the United Nations (FAO). 1974. Soil Map of the World. Legend, FAO, Rome.

Food and Agriculture Organisation of the United Nations (FAO). 1989. FAO Yearboak 1989: Fertiliser. Food and Agriculture Organisation of the United Nations. Rome. Volume 39. 127 pp.

Food and Agriculture Organisation of the United Nations (FAO). 1990. Soil Map of the World. Revised Legend. FAO, Rome.

Food and Agriculture Organisation of the United Nations (FAO). 1992. FAO Yearbook: Production: 1991. Food and Agriculture Organisation of the United Nations. Rome. Volume 45. 265 pp.

Food and Agriculture Organisation of the United Nations (FAO). 1995a. FAO Yearbook: Production: 1994. FAO, Rome. 243 pp.

Food and Agriculture Organization of the United Nations (FAO). b. FAO Yearbook: Fertiliser 1994. Food and Agriculture Organisation of the United Nations. Rome. Volume 44. 121 pp.

Fertilizer Use Recommendation Project (FURP). 1994. Fertilizer Use Recommendations. Volumes 1-22. Kenya Agricultural Research Institute, Nairobi, Kenya.

Haines, P.J. and Uren, N.C. 1990. Effects of conservation tillage farming on microbial biomass, organic matter and earthworm populations in north-east Victoria. Australian Journal of Experimental Agriculture 30:365371.

Hargrove, W.L. and Thomas, G.W. 1981. Effect of organic matter on exchangeable aluminium and plant growth in acid soils. In: Chemistry in the Soil Environment. Stelly, M. (Ed.), pp. 151-186. American Society of Agronomy Special Publication 40. American Society of Agronomy. Madison, Wisconsin.

Hue, N.V., Craddock, G. and Adams, F. 1986. Effects of organic acids on aluminum toxicities in subsoils. Soil Science Society of America Journal 50:28-34.

Jagtap, S. 1991. The Resource Information System (RIS, Computer Software). IITA Agroclimatology Laboratory. Ibadan, Nigeria.

Kanake, P.J.K. 1982: Semi-detailed Soil Survey of Kimorigo/Kamleza Irrigation Schemes (Taita-Taveta District). Semi-detailed Soil Survey Report No. 56, Kenya Soil Survey. Nairobi.

Knox, E.G. 1981. The role of soil surveys in the decision-making:Process for development planning. In: SMSS Technical Monograph No. 1. Soil Resource Inventories and Development Planning. Proceedings of Workshops at Cornell University 1977-1978.

Lal, R. 1986. Soil surface management in the tropics for intensive land use and high and sustained productivity. In: Advances in Soil Science, Volume 5. (Steward, B.A. Ed.), pp. 2-97. Springer-Verlag New York Inc., New York.

Lavelie, P. 1988. Earthworm activities and the soil system. Biology and Fertility of Soil 6:237-251.

McClellan, G.H. 1991. Use of East and Southeast African indigenous agrominerals for fertilizer production. Fertilizer Research 30: 125-126.

Mochoge, B.O. 1993. Fertiliser trial results of maize and maize-bean intercrop trials in Kenya. In: Soil Fertility Research for Maize and Bean Production Systems in the Eastern African Highlands. CIAT Network on Bean Research in Africa, Workshop Series 21:5978.

NilI, D. and Nill, E. 1993. The efficient use of mulch layers to reduce run off and soil loss. In: Dynamics of Organic Matter in Relation to Sustainability of Agricultural Systems. John Wiley and Sons, Chichester, U.K. pp. 331338.

Oades, J.M. 1984. Soil organic matter and structural stability: Mechanisms and implications for management. Plant and Soil 76:319-337

Okigbo, B.N 1990. Sustainable agricultural systems in tropical Africa. In: Sustainable Agricultural Systems. Edwards, C.A. et al. (Eds.), pp. 323-352.

Post, W.M. and Mann, L.K. 1990. Changes in soil organic carbon and nitrogen as a result of cultivation. In: Soils and the Greenhouse Effect. Bouwman, A.F. (Ed.), pp. 401-406. John Wiley and Sons, New York.

Russell, E.W. 1973. Soil Conditions and Plant Growth. lOth Edition. pp. 284-326. Longman Group Ltd. London and New York.

Sanchez, P.A. 1990. Deforestation reduction initiative: an imperative for world sustainability in the Twenty-First Century. In: Soils and the Greenhouse Effect. Bouwman, A.F. (Ed.), pp. 375-382. John Wiley and Sons Ltd. Chichester, UK.

Sanchez, P.A. 1976. Properties and Management of Soils in the Tropics. John Wiley and Sons, New York. 618 pp.

Sanchez, P.A., Palm, C.A., Szott, L.T., Cuevas, E. and Lal, R. 1989. Organic input management in tropical agroecosystems. In: Dynamics of Soil Organic Matter in Tropical Ecosystems. Coleman, D.C., Oades, J.M. and

Uehara, G. (Eds.), pp.125-152. NifrAL Project, Paia, Hawaii, USA.

Sant-Anna, R. 1985. Problem soils of Sub-Saharan Africa. World Soil Reports 58: 14-20.

Smaling, E.M.A. 1993. An Agro-ecological Framework for Integrated Nutrient Management, With Special Reference to Kenya. Doctoral Thesis, Agricultural University, Wageningen, The Netherlands. 250 pp.

Swift, M.J. and Anderson, J.M. 1992. Biodiversity and ecosystem function in agricultural systems. In: Biodiversity and Ecosystem Function. Schultz et al. (Eds.), Berlin, German. Springer-Verlag.

Swift, M.J., Kang, B.T., Mulongoy, K. and Woomer, P. 1993. Organic-matter management for sustainable soil fertility in tropical cropping systems. In: Evaluation for Sustainable Land Management in the Developing World. Volume 2; Technical Papers. lBSRAM Proceedings 12:307-326.

Van Kauwenberg, S.J. 1991. Overview of phosphate deposits in Easi and Southeast Africa. Fertilizer Research 30:127-150.

Woomer, P. 1993. What beyond sustainability? In: Soil Fertility Research for Maize and Bean Production Systems of the Eastern African Highlands.

Wortmann, C.S. and Ransom, J.K. (Eds.), pp. 91-100. Network on Bean Research in Africa, Workshop Series No. 21.

Woomer, P.L., Martin, A., Albrecht, A., Resck, D.V.S. and Scharpenseel, H.W. 1994. The importance and management of soil organic matter in the tropics. In: Biological Management of Tropical Soil Fertility. Woomer, P. and Swift, M.J. (Eds.), pp. 47-80. John Wiley and Sons, Chichester, U.K.

Woomer, P.L. and Ingram, J.S.I. 1990. The Biology and Fertility of Soils: TSBF Report 1990. TSBF, Nairobi, Kenya.

Yamoah, C.F., Burleigh, J.R. and Malcolm, M.R. 1990. Application of expert systems to study of acid soils in Rwanda. Agriculture, Ecosystems and Environment 30:203-218.

Copyright 1996 The African Crop Science Society