Soil Use

An Assessment of the Soil Resources of Africa in Relation to Productivity

Hari Eswaran *, Russell Almaraz, Evert van den Berg, and Paul Reich

World Soil Resources, Soil Survey Division, USDA Natural Resources Conservation Service, Washington D.C. 20013. Received 28 February, 1996.

Corresponding author (heswaran@usda.gov).

Abstract

Africa, with a total land mass of about 30.7 million km² and a population exceeding 746 million persons, has generally lagged behind in agricultural development. Sub-Saharan Africa (excluding South Africa) is the poorest developing region, with 29 out of 34 countries being some of the poorest in the world. The purpose of this study is to develop a Soil Taxonomy map, based on the FAO Soil Map of the World, which together with other data, is used to make continent-level assessments of land productivity and sustainability. Prime land occupies about 9.6% of Africa and the lands with high potential occupy an area of about 6.7%. The medium and low potential lands, which together occupy 28.3% of the area have major constraints for low-input agriculture. Resource poor farmers who live on these lands have high risks and generally, the probability of agriculture failure is high to very high. The remaining about 55% of the land consists of deserts or other lands with major constraints even for low-input agriculture. The desert margins have nomadic grazing which with increasing animal population is stressing the environment. A soil quality analysis and an evaluation of sustainable production, based only on biophysical considerations, suggest the need for major investments to enhance the productivity of the soil resources of this continent.

Index words: Africa, land productivity, land quality, soil constraints

An Assessment of the Soil Resources of Africa in Relation to Productivity

Africa is a vast continent with a tremendous resource endowment and offers great potential for increased agricultural productivity (FAO, 1993). If agricultural performance is measured in terms of per capita food production, there has been an alarming decline in the last few decades according to The World Resources Institute (WRI, 1994 p.292). Where food production in Africa from 1980-82 and 1990-92 fell short of the demand, per capita food production dropped by 5 percent from the benchmark years of 1970-72. There are many reasons for this, including political and socioeconomic, but the associated consequence is a decline in the quality of the land resource base in many African countries, which has negative longer term impacts (Eswaran and Dumanski, 1994). Research and development must address these linkages to find solutions to balance the natural ecosystems with the human social systems.

With the exception of a few countries such as Kenya, detailed information on the soil resource base is generally inadequate for most developmental purposes. In most countries, farm-level information and detailed soil maps are non-existent. When this is coupled to other socioeconomic constraints, including land titling and availability of capital for land management investments, it provides one explanation for the lack of progress in poverty alleviation and food security (Cleaver and Schreiber, 1994).

The purpose of this study is to develop a Soil Taxonomy (Soil Survey Staff, 1975, 1994) map at an Order 5 level using specifications detailed by the Natural Resources Conservation Service (NRCS) of the U.S. Department of Agriculture (Soil Survey Staff, 1993) of the African Continent which together with other data, can be used to make continent-level assessments of constraints to productivity. This analysis is then used to assess soil quality and thereafter, aspects of agricultural sustainability. Many assessments have been made of the land resource base of Africa. This assessment uses recent soils information and a significantly larger soil climate database than previous studies.

Materials and Methods

The Soil Map of the World (FAO-UNESCO, 1977), digitized and made available by the Food and Agriculture Organization (FAO) is the starting point for the assessment. The office of World Soil Resources of the Natural Resources Conservation Service (NRCS) has a pedon data base with more than 400 pedons from Africa and this, together with published national soil survey reports, provide the recent information for the translation of the legend of the Soil Map of the World into Soil Taxonomy. The FAO soil map of Africa at a scale of 1: 5 million has over 6,000 polygons. The FAO soil unit designator for each polygon is systematically converted to a Soil Taxonomy great group and for this process, only the dominant unit in the association is considered. A minimum map delineation size of about 5,000 km²represents the smallest polygon which can be made, at the map scale of 1:12 million, without reducing map legibility (Soil Survey Staff, 1993).

An associated database of rainfall and temperature of more than 2,500 stations is assembled, and using an unpublished model of NRCS, modified later by Van Wambeke (1982), estimates of the soil moisture and temperature regimes (SMR and STR) are made for each station. The basic data set is from Wenstedt (1972). For some countries, this data set was substituted with more recent data when available. Thus the data set is not uniform with respect to temporal coverage. An evaluation was done at a few locations by on-site visits and in most cases by comparisons of the computed SMR and STR with other climatic and vegetation maps. Geographical patterns were the main criteria for assessment, and if within a given area a station has an aberrant rainfall and/or temperature condition, this station was eliminated from the database. Each polygon on the FAO map is then assigned a soil climate code. This is necessary because the Suborder and Great Group categories of Soil Taxonomy include soil climate information. The validation of the final Soil Taxonomy map of Africa, produced at a scale of 1:12 million, is made using the pedon database, published information, and the field experience of the authors.

Soil Climatic Conditions

Assessment of soil moisture and temperature conditions

Knowledge of soil moisture and temperature regimes (SMR and STR) for land use and other applications of climatic data, is as useful as air temperature and rainfall information. SMRs and STRs are studied by meteorologists, climatologists, ecologists, agronomists, soil scientists, civil engineers, farmers, and many others who use soils. Both SMR and STR affect physical, chemical, and biological processes in soils. Growth, multiplication, and activities of soil-borne organisms are influenced by soil climate. Seed germination is determined by both the availability of adequate moisture and by a limited range of soil temperature. Soil temperatures above or below critical limits severely inhibits seed germination even if there is adequate soil moisture. The life-cycles of many soil-borne pests and diseases are controlled by SMR and STR.

The definitions of SMRs and STRs provided in Soil Taxonomy (Soil Survey Staff, 1994) have some deficiencies and an attempt is made to modify some of the definitions. Only changes relevant to Africa are presented here:

we have introduced megathermic and isomegathermic regimes with a mean annual soil temperature (MAST) of 29^oC or higher.
we have deleted isofrigid as we have no station designated as such in our database of 27,000 stations around the world. The only possible location for isofrigid temperature regime in Africa would have been at the top of Mt. Kilimanjaro.

In Table 1, the pedo-climatic domains of Africa are shown. Geographers and ecologists often do not use the terminology of Soil Taxonomy and so common names, that serve as close equivalents such as temperate, tropical, Mediterranean etc. are included in Table 1. The subdivisions of the SMRs are according to Van Wambeke (1982) with a few additions to complete the system.

The tropical regions are demarcated by the iso-temperature regimes -- isomesic, isothermic, isohyperthermic, and isomegathermic. The temperate regions are demarcated by the mesic, thermic, hyperthermic, and megathermic STRs. The mesic STRs are confined to small areas on the Atlas mountains in North Africa and the delineations are too small to be shown on the map in Fig. 1.

The Soil Climate domains of Africa are depicted on a map produced using PC ARC/INFO Geographic Information System at 1:12 million scale ( the original maps in color are available from the authors and Fig. 1 is a reduced, simplified version). The soil climate domains of the climate stations were geographically plotted as a separate map coverage overlying the soil polygon coverage. Soil climate classifications were assigned to the soil polygons manually according to similarity or dominance of climate classes within the polygon and classes in close proximity to the polygon. Estimated classifications were made for polygons without climate stations. For those polygons between stations having contrasting climates, a gradient was visually estimated to identify the transition of soil climate.

Table 1 lists the areas occupied by each soil climate class defined by the combination of SMR and STR. The hyperthermic and isohyperthermic STRs together occupy more than 60 % of the land mass of Africa. The new isomegathermic STR occupies about 16 % and the megathermic STR, about 5 %. The megathermic and isomegathermic STRs only occur in the arid and semi-arid areas. In the humid areas, mean annual soil temperature is reduced by the cloud cover in the hot season. The tropical deserts occupy about 10 % of the area and are largely in the Sahelian region. The temperate deserts occupy about 37 % with the larger extent in the Sahara and smaller areas in Namibia, Botswana, and South Africa. About 2 % of Africa has a xeric SMR or the classical Mediterranean climate and over 95 % of this is in North Africa. There is a small area around Cape Town in South Africa where the SMR is xeric. Much of tropical Africa (about 35 %) is semi-arid or has an ustic SMR. The semi-arid area in temperate Africa is largely in South Africa. About 12 % of tropical Africa is humid and distributed mainly in the Congo Basin, the Ethiopian and East African highlands, and parts of West Africa. The humid regions in temperate Africa are mainly in the western part of North Africa and in Madagascar.

Soil Resources

Distribution of major soil classes

Two of the early efforts of compiling the soil resources of Africa, were those of Marbut (1923) and Shokalskaya (1944). These maps at a scale of 1:30 million are based on geology and phytogeography and had minimal field soil information. Many of the recent concepts of tropical weathering and soil formation originated in Africa during the colonial period (Ollier, 1959; King, 1962, and Moss, 1968). In 1953 an Interafrican Pedological Service was created with its Administrative Headquarters at Yangambi, Belgian Congo (Zaire). The Council of this Service decided in 1955 to make a soil map of the continent which was eventually realized in 1964 (D'Hoore, 1964). This soil map and accompanying report was perhaps the last effort in Africa until the FAO-UNESCO Soil Map of the World was published a decade later. In retrospect and particularly realizing the conditions under which soil survey investigations were conducted during the fifties and sixties in Africa, the map and report is a most commendable effort. As will be shown later, soil names may have changed and polygon boundaries refined, but the basic information is still very reliable. As an example, the Vertisols were reported to occupy about 3.28 % of the area; in the present study, the area is 3.23 %.

Fig. 2 illustrates the distribution of soil Orders in Africa. The scale does not permit a map at a more detailed categoric level. However, a map at scale 1:12 million showing the distribution of soils at the Great Group level is available with the authors. Table 2, provides the areal distribution of the Great Groups of soils in Africa. Information on the corresponding Suborders and Orders is also given in Table 2.

Histosols (Eswaran, 1986) is one of the smaller soil Orders in terms of areal distribution, occupying about 0.05 % of the land mass. They are confined to the coastal plain of West Africa and Madagascar where they are associated with acid sulphate soils (Sulfaquents and Sulfaquepts) and other Aquepts and Aquents. Spodosols are also present as small enclaves in two locations. A large extent is with the wet sandy soils present in depressions in southern Zaire. The sands are wind-blown material from the Kalahari and the soils form a catenary sequence with deep Psamments on the plateaus, Arenic Palustalfs on the slopes leading to the Aquods and Humods, associated locally with Tropofibrists. A small area of Spodosols (Haplohumods) is also found east of Cape Town in S. Africa on wind-blown sands. Andisols are the third group of soils found in small areas and occupy less than 0.16 % of the area. The upper slopes of Mt. Cameroon, Kilimanjaro, Mt. Kenya, and parts of the Ethiopian Highlands have Andisols.

Oxisols (Eswaran et al., 1986) occupy about 3.75 million km² or 14.3 % of the land area and are confined to the Congo Basin and adjoining areas. Much of the early information on African Oxisols was provided by the soil survey program of the Belgians (Tavernier and Sys, 1965). Both in the legend of the Belgian maps and that of the FAO-UNESCO Soil Map of the World (FAO-UNESCO, 1977), Oxisols are referred to as Ferralsols, with a definition very similar to the Soil Taxonomy definition. The southern extension of Oxisols in Zambia is not well defined due to the Holocene cover of the Kalahari sands which are of varying thickness. When the cover is thicker than 150 cm, the soils are classified as Alfisols, Ultisols, or Entisols. This distribution of Oxisols is evident in figure 2 where, in the south, the tongue of Psamments is seen invading the zone of Oxisols with a large area on the west (in Angola) and the remaining on the eastern part of the tongue in Zaire. Madagascar has a very large extent of Oxisols on the eastern piedmonts and much of the central highlands. These are formed on old basement rock complexes and weathering may extend to several tens of meters. The organic matter rich Oxisols are located in the highlands around Lake Kivu (Rwanda and Burundi) and the north eastern part of Zaire, in the Ituri Province. Many of them have more than 16 kg/m/m² of carbon qualifying them for humic subgroups. Many also have a sombric horizon which has not been described in other parts of the world. These high organic matter isothermic Oxisols are very productive and support some of the highest population densities in the world.

The Vertisols (Wilding and Puentes, 1988) are distributed along the rift valley from Sudan in the north to South Africa in the south, with sporadic occurrences in other parts of Africa. In North Africa, particularly in Morocco, the Xererts or Tirs have been used for agriculture since the Roman period. They are extremely difficult soils for tillage (Eswaran et al., 1988) both in the wet and dry season and until the recent introduction of high energy equipment, much of the cultivation was confined to the post-rainy period. On the African Rift, the Vertisols of Sudan on the Gezira Plains are the best known and are most intensively used under irrigation. There are also large contiguous areas in South Africa, Zambia, Kenya, and Somalia, and occur locally in West Africa.

Aridisols (Eswaran et al., 1993) occupy about 26.4 % of Africa and associated with these soils are the 'torric' great groups of Entisols and some other Orders. The Torripsamments and Torriorthents (Table 2) together occupy about 15 %. The remaining 46.2 % of the areas with aridic SMR of Africa are comprised of rockland, salt pans, sand dunes, and minor components of other 'torri' great groups. Thus for most practical purposes, about half of Africa has land unsuitable for low-input agriculture.

Closely related to the Oxisols are the Ultisols occupying about 6.2 % of the land area (Table 2). In Zambia and the western part of Angola, many of the Ultisols are actually buried Oxisols buried under the Kalahari sands and thus characteristically have very sandy top soils and a low activity clayey subsoil. A similar feature is seen in the Alfisols, which occupy about 10.5 % of the area and are extensive in the Sahelian part of Africa. In the Sahel, which borders the Sahara, the wind-blown sand has buried many of the former Oxisols and Alfisols/Ultisols. In Central Africa, the Alfisols and Ultisols are the typical soils of the mid- and end-Tertiary plateaus. Most of the soils are reworked and some have stone-lines composed of quartz or petroplinthic gravel (Ruhe, 1956). Typically, the geomorphic surfaces occupy specific elevations and at the edge of the plateaus are bands of re-cemented petroplinthite. These are the Petroferric subgroups of some Alfisols and Ultisols. The petroferric contact is an impermeable layer to both roots and water.

The Mollisols are more dominant in the areas with xeric SMR, with large extents in Morocco and the coastal areas of Algeria and Tunisia. In sub-Saharan Africa, the Mollisols are confined to the isothermic areas on recent base rich materials. In these areas, a Mollisol-Alfisol association is common.

Inceptisols occupy about 7.8 % and Entisols, about 24.5 % of the land mass of Africa. About 50 % of the Entisols have an aridic SMR and are formed on sandy or loamy deposits. Another about 5 % of the Psamments are present as interfingerings of the Kalahari or Sahara in zones with ustic or udic SMRs. A large proportion of the Inceptisols are shallow soils on dissected or mountainous lands. They are generally not used for agriculture.

Soil Quality

Constraints to agricultural use

Soil quality is the ability of the soil to perform its functions in a sustainable manner. For detailed land use assessments, quantitative soil quality assessments can be made. As the intent here is to make qualitative continent-level assessment, only factors that reduce the ability of the soil to perform in an optimal manner for agriculture are considered. Constraints to sustained use of soils are multiple and include socioeconomic constraints which are not considered here. Only the biophysical constraints are evaluated with the objective of having some continent-level estimates. Soil moisture stress, as indicated earlier is perhaps the overriding constraint in much of Africa. As shown in Table 1, only about 14 % of Africa is relatively free of moisture stress. The other major stresses discussed here relate to soil properties and are considered individually, though the same soil may have more than one of these constraints. A more detailed GIS analysis than the one attempted here, will provide details of areas with multiple stresses.

Moisture stress is not only a function of the low and erratic precipitation but also of the ability of the soil to hold and release moisture. Table 3a, provides estimates of the area of soils with different available water holding capacities (AWHC). About 10 % of the soils have high to very high AWHC. These are mainly the Mollisols, Vertisols, and other clayey soils with 2:1 layer lattice clays. The 29 % of soils with medium AWHC are mainly the Alfisols and Ultisols and some loamy Inceptisols and Entisols. The low AWHC class soils are the Oxisols and other sandy loam soils. Despite their clayey textures, Oxisols have low AWHC. The very low AWHC class soils are the sandy soils such as Psamments and other sandy and sandy loam soils.

The potential of soils to fix phosphorous is difficult to estimate. High free iron content, which is reflected in the red colors of the soil, due to the nature of the parent material or the weathering stage, is employed as an indicator and table 3b provides estimates occupied by the different classes of P-fixing soils. The high P fixing soils are mainly the sesquioxide rich Oxisols and Ultisols. P is immobilized as Fe- and Al- phosphates in these soils. This is a crude estimate of P in soils and a refined evaluation would require determination of the activity of P in the soil.

In the arid and semi-arid parts of Africa, salinity and alkalinity is a major problem affecting about 24 % of the continent (Table 3c). These are included in the soils designated as having a pH >8.5 in table 3d. The extremely acid soils, which are mainly the acid sulphate soils (both potential and actual) occupy a small area around the Niger delta and occur sporadically along the coastal plains of West Africa. The soils which have high aluminium problems, occupy about 15 % of the continent and are mainly in the moist parts of the semi-arid zones and the sub-humid areas. Many of the Ultisols and some Alfisols have acid surface and subsurface horizons which, coupled to the moisture stress conditions, makes these soils extremely difficult to manage under low-input conditions. In West Africa, the annual additions of dust from the Sahara brought by the Harmattan winds, raise the pH of the surface horizons and so the problem is less acute but subsoil acidity remains.

Soil depth is frequently understood as being depth to rock or an impermeable layer. The term effective soil depth (Table 3e) is used here to include chemical barriers which reduce the volume of the soil for root exploitation. Effective soil depth is a problem in more than 50 % of the soils and this reduces the potential of the soil for crop production.

Wind and water erosion is extensive in many parts of Africa. The potential for such erosion is shown in Tables 3f and g. Excluding the current deserts which occupy about 46 % of the land mass, about 25 % of the land is prone to water erosion and about 22 %, to wind erosion. High intensities of these erosion forms are mainly in the semi-arid and sub-humid areas. The soils in Central Africa are largely low activity Oxisols and Ultisols and are less susceptible to water erosion, unless severely mismanaged. These estimates include human-induced erosion, good estimates for which are available in Oldeman et al., 1992.

Based on this analysis and with an understanding of the soils and their distribution, it is possible to make an estimate of the lands which are susceptible to desertification, in its broadest sense. Table 3h, indicates that about 30 % of Africa, mainly in the semi-arid and sub-humid parts, is highly susceptible to desertification. From this point of view, these are fragile ecosystems and considerable damage can be done with low-input agriculture.

Assessment of sustainable development potentials

Soil properties, including soil climate, provide some preliminary information to address soil quality. Soil quality is used here to indicate the ability of the soil to perform its function of sustaining agriculture under the current low-input system of agriculture. There are areas, particularly in South Africa and some other countries, where the agriculture is characterized by medium or high input systems and a separate assessment can be made for such situations. Low-input implies that large-scale irrigation is absent, use of fertilizers, and pest and weed control is minimal, and soil management does not require high energy mechanized equipment. The soil classification name provides adequate information to make this preliminary assessment on a continent scale. The properties incorporated in the soil name are empirically related to crop performance and a judgment is made on the potential of the soil for sustaining agriculture. A more refined assessment would require a much larger database.

Each polygon on the soil map at 1:12 million scale, is assigned one of the six classes of Table 4. Prime land comprises those soils with deep, permeable layers, with an adequate supply of nutrients, and generally do not have significant periods of moisture stress. The high potential lands are similar to the prime lands but seasonal moisture stress is a major constraint for agricultural use. The medium potential lands have a major soil constraint and one or more minor constraints which normally can be corrected through management. The low potential lands have several major constraints which are not easily corrected through management. The unsustainable class of lands are those which are considered to be fragile, easily degraded through management, and in general are not productive or do not respond well to management. They are generally highly erodible and generally require very high investments for any kind of agriculture. A new interpretive map is then drawn (Fig. 3) with these classes and the area for the classes is computed.

In Table 4, prime land has soils which are generally Mollisols and Alfisols. The soils are deep, without impermeable layers, textures are loamy to clayey with good tilth characteristics, and the land is generally level to gently undulating. These relatively young soils have a good and balanced nutrient supply and the water holding capacity is more than 150 mm per metre. They have a wide range of agricultural uses, crop performance is generally good and more importantly, their response to management is very high. They also have good resilience and if degraded slightly, can be brought back to their near original level of performance with good management. They occupy about 9.6 % of Africa and as shown in Fig. 3, they occupy significant areas in West Africa south of the Sahel, in East Africa mainly in Tanzania, and in Southern African countries of Zambia, Zimbabwe, South Africa, and Mozambique.

High potential lands are those with some minor limitations. The limitations may be an extended period of moisture stress, sandy or gravelly materials, or with root restricting layers in the soil. Generally, there is adequate moisture during the rainy season to obtain one good crop, however stored moisture for the second crop is not always reliable. The high potential lands (Table 4 and Fig. 3) occupy an area of about 6.7 % and have the inherent potential to be productive if proper land management is practiced. They can be severely stressed under low-input systems. Their resilience properties are not expected to be as good as the prime lands and thus may be permanently damaged through mismanagement.

The medium and low potential lands, which together occupy 28.3 % of the surface have major constraints for low-input agriculture. Resource poor farmers who live on these lands have high risks and generally, the probability of crop failure is high to very high. The constraints include adverse soil physical properties including surface soil crusting, impermeable layers, soil acidity and specifically subsoil acidity, salinity and alkalinity, and high risks of wind and water erosion. The desert margin areas, which includes the southern part of the Sahel and the zones where the ustic SMR is grading to aridic, are examples of this condition. The large contiguous areas of Central and West Africa, are considered as medium potential, due to the presence of acid soils and soils which fix high amounts of phosphates. With an inherently low soil quality, low-input agriculture can be equated to potential soil degradation. These are some of the priority areas for technical assistance and the implementation of appropriate soil management technologies.

The soil quality analysis and the evaluation of sustainable production provide a first assessment for Africa as a whole. Such a continent-wide assessment is not useful for a country which may have soils of variable quality and this points to the need for detailed national assessments.

Productivity of African lands

A few other observations from this study are:

there is a concentration of high quality lands in West and East Africa, including parts of the Central African Highlands. These lands also have high population densities. Population pressure and land degradation are major problems which go in tandem and both must be addressed together to continue sustainability. The major thrust of national and regional strategies must focus on maintaining the high productivity of these lands, which calls for assisting the farming community move up to medium-input systems. These lands will be most responsive to introduction of high yielding cultivars and modern pest and weed management techniques;
the least populated parts of Africa, the Sahara and Namibian Deserts and the Horn of Africa, have the most unsuitable (for agriculture) and fragile soils. However, domesticated animal population is increasing with concomitant land degradation. Many ecosystems in these fragile lands are under onslaught. The desert margins are increasingly being exploited and require new research and developmental efforts;
much of the land within the tropical belt is biophysically capable of supporting some type of sustainable agriculture system. Much of their sustainability is dependent on the extent of inputs to maintain soil quality. In many of these developing countries, though application of costly inputs in order to achieve sustainability is not realistic, expansion of low-input agriculture systems on these soils is likely to result in soil degradation and loss of productivity. There is thus an urgent need to provide alternate options for a more equitable use of the soil resources until some form of capital intensive systems can be implemented;
food security is becoming or is already of paramount concern in many of the African nations. Traditional agriculture systems and declining 'aid' imports may not supply the needs of a region which has some of the highest birth rates (Kesseba, 1993; Okigbo, 1993). A declining resource base will eventually contribute to civil unrest due to uncertain food supplies;
marginal and unsuitable lands, identified by the assessment, should be managed to preserve their quality as they may be too uneconomical or unsustainable to support more than very low intensity traditional systems. These areas may be of benefit to other types of land use such as wetlands needed by breeding fish stocks and wildlife refuges. These marginal lands have been shown to be income producers when managed for wildlife. Kenya and Zimbabwe have been successful in attracting tourists to view wildlife in 'marginal' land set aside as wildlife refuges. Tourism contributes many millions of dollars to the local and national economies of both countries.

As the population growth and the demand for suitable agricultural land increases, there is need for a planned assessment of agricultural potential in Africa. Soil Taxonomy provides a basis for assessments of soil resources, especially using small scale regional maps and more detailed national and local assessments. The information, when incorporated into a Geographic Information System, becomes more useful by providing greater analytical capabilities and facilitating the creation of interpretive maps. Data can also be manipulated to give a 'best guess' where data is scarce or of unknown quality, as is often the situation in Africa. With socioeconomic and other data layers, realistic national assessments and appropriate strategic plans can be developed.

Acknowledgements

The authors acknowledge the Land and Water Division of FAO for providing the digital database of the FAO-UNESCO Soil Map of the World. Support of this work was from the US Agency for International Development through the Soil Management Support Services Program.

References

Cleaver, K.M. and G.A. Schreiber. 1994. Reversing the Spiral: The population, agriculture, and environment Nexus in Sub-Saharan Africa. The World Bank, Washington D. C. 293 pp.

D'Hoore, J.L. 1964. Soil Map of Africa, Scale 1:5,000,000. Joint Project No. 11. Commission for Technical Cooperation in Africa Publ. Ceuterick, Louvain, Belgium, 205 pp.

Eswaran, H., 1986. Classification and management of Histosols. In: Eswaran et al. (Eds.) Proc. of 2nd. Int. Symposium on Peat land soils., Thailand/Malaysia. Publ. Dept. of Land Development Thailand. 15-26.

Eswaran, H., H. Ikawa, and J.M. Kimble. 1986. Oxisols of the World. Proc. Symposium on Red Soils. Beijing, People's Republic of China. Publ. Elsivier Amsterdam, 90 - 123.

Eswaran, H., J. Kimble and T. Cook. 1988. Properties, genesis and classification of Vertisols. In Trans. Int. Workshop Swell-shrink Soils. Publ. Oxford & IBH Publ. House. New Delhi, India. 1 - 22.

Eswaran, H., E. Van den Berg, and R. Almaraz. 1993. Global distribution of Aridisols and their characteristics. In: Proc. Intl. Workshop on Classification and Management of Arid-Desert Soils. Publ. China Science and Technology Press. 388-399.

Eswaran, H., J. Dumanski. 1994. Land Degradation and Sustainable Agriculture: A Global Perspective. In Press. Submitted to 8^th Conference of the International Soil Conserv. Organization, New Delhi, India.

FAO-UNESCO. 1977. Soil Map of the World, Africa. Publ. FAO-UNESCO, Paris, 299 pp.

FAO. 1993. Agriculture Towards 2010. Publ. FAO, Rome.

Kesseba, A.M. 1993. Strategies for developing a viable and sustainable agriculture sector in sub-Saharan Africa: Some issues and options. In: J. Ragland and R. Lal (ed.), Technologies for Sustainable Agriculture in the Tropics. Publ. Amer. Soc. Agron. Madison, WI., ASA Special Publ. 56:211-244.

King, L.C. 1962. Morphology of the Earth. Publ. Oliver and Boyd, Edinburgh.

Marbut, C.F. 1923. The vegetation and soils of Africa. American Geographical Society.

Moss, R.P. 1968. Soils, slopes, and surfaces in tropical Africa. In: (Moss, R.P. Ed.) The Soil Resources of Tropical Africa. Cambridge Univ. Press., Cambridge, 29-60.

Okigbo, B.N. 1993. Sustainable agriculture in sub-Saharan Africa: Perceptions, scope, complexity, and challenges. In: J. Ragland and R. Lal (ed.), Technologies for Sustainable Agriculture in the Tropics. Publ. Amer. Soc. Agron. Madison, WI., ASA Special Publ. 56:245-256.

Oldeman, L.R., R.T.A. Hakkeling, and W.G. Sombroek. 1992. World map of the status of human-induced soil degradation: an explanatory note. Wageningen, Netherlands: International Soil Reference Center. 34 pp.

Ollier, C.D. 1959. A two-cycle theory of tropical pedology. J. Soil Sci. 10:137-148.

Ruhe, R.V. 1956. Geomorphic surfaces and the nature of soils. Soil Sci. 82:441-455.

Shokalskaya, Z.V. 1944. A new soil map of Africa. Pochvovodenye, 9:424.

Soil Survey Staff. 1975. Soil Taxonomy. A basic system of soil classification for making and interpreting soil surveys. Soil Conservation Service, USDA. Agr. Handbook No. 436. Washington D.C.

Soil Survey Staff. 1993. Soil Survey Manual. USDA Handbook No. 18. US Govt. Print. Office, Washington D.C. 437 pp.

Soil Survey Staff. 1994. Keys to Soil Taxonomy. Sixth Ed. 1994. US Govt. Print. Office, Washington D.C., 306 pp.

Tavernier, R. and C. Sys. 1965. Classification of the soils of Congo. Pedologie (Ghent, Belgium). Special Issue No. 5:91-136.

Van Wambeke, A. 1982. Calculated soil moisture and temperature regimes of Africa. SMSS Tech. Monograph No. 3. Washington D.C.

Wenstedt, F.L. 1972. World Climatic Data. Climatic Data Press. Lemont, Pennsylvania, USA.

Wilding, L.P. and R. Puentes. (Eds). 1988. Vertisols: Their distribution, classification, and management. Publ. Soil Management Support Services, Technical Mono. No. 18. 193 pp. USDA-NRCS, Washington DC.

World Resources Institute. 1994. World Resources 1994-95. Oxford Univ. Press, pp. 400.

< Back to Manuscript and Papers Index