d THE WORLD BANK FILE NVY POLICY PLANNING AND RESEARCH STAFF 4A/MV - 0O / 7 Environment Department The Causes, Nature and Rate of Soil Degradation in the Northernmost States of Nigeria and An Assessment of the Role of Fertilizer in Counteracting the Processes of Degradation Michael Mortimore July 1989 Environment Department Working Paper No. 17 This paper has been prepared for internal use. The views and interpretations herein are those of the author(s) and should not be attributed to the World Bank, to its affiliated organizations or to any individual acting on their behalf. This paper has been prepared by Michael Mortimore. He is a Honorary Research Fellow at the Centre of West African Studies, University of Birmingham and prepared the paper as a consultant for the Environment Department. The permission of the Director of the Institute for Agricultural Research, Samaru to use the Institute's library, the advice and support of Mr. Graham Quinn (FACU, Kaduna) and others who provided information are gratefully acknowledged. Departmental Working Papers are not formal publications of the World Bank. They present preliminary and unpolished results of country analysis or research that is circulated to encourage discussion and comment; citation and the use of such a paper should take account of its provisional character. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. Because of the informality and to present the results of research with the least possible delay, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. - ii - This study was carried out as part of a program of work on dryland management, and as a background for agricultural sector work in Nigeria, which had a particular emphasis on the role of fertilizer. There has been much concern expressed about soil degradation in semi-arid areas such as northern Nigeria, and in particular "desertification". The consensus is that desertification is proceeding steadily as reflected in the following indicators: declining rainfall, soil erosion by wind and water; deforestation, scanty vegetation, tree mortality, the invasion of dryland species and fodder scarcity; diminishing groundwater; falling farm yields; southward migration of livestock producers and/or farmers; and food scarcity. However, this consensus is heavily influenced by the forestry profession's view of environmental degradation (i.e., fewer trees) and the difficulties in afforestation programs. Declining trends in crop yields, although widely reported, are difficult to substantiate in reliable figures and impossible to consider independently from the effects of rainfall over the past decade. A large number of studies and reports were reviewed, but none attempted to demonstrate empirically the linkages between causes and effects. The effect of cultivation on the physical and chemical properties of soils are known and have been demonstrated under experimental conditions. However, few attempts have been made to monitor such effects under smallholder conditions. Intensive cultivation has been practiced in areas of northern Nigeria for many years (generally near population centers, especially Kano) without apparent degradation of soils. These areas are not typical of all northern Nigeria, but they do suggest that degradation is not an inevitable consequence of intensification and reduction of fallow. The study was not able to find time-series data or equivalent information monitoring changes in soil characteristics as land use intensified. The generation of such data should be a priority. CONTENTS 1 INTRODUCTION 2 RESOURCE MAPPING 2.1 Rainfall 2.2 Soils and degradation risk 2.3 Population and land use 3 EVIDENCES OF DEGRADATION 3.1 Rainfall 3.1.1 Drought 3.1.2 Monthly distribution 3.1.3 The August peak 3.1.4 Effects on crop production 3.1.5 Hydrological conditions 3.2 Desertification 3.2.1 The 'Desertification Consensus@ 3.2.2 Administrative problems 3.2.3 Affected areas 3.2.4 Evaluating the 'desertification consensus' 3.3 Vegetational degradation 3.3.1 Burning 3.3.2 Grazing 3.3.3 Woodcutting 3.3.4 Farming 3.3.5 Evaluating vegetational degradation 3.4 Soil erosion 3.4.1 Erosion hazard 3.4.2 Gully erosion 3.4.3 Soil crash 3.4.4 Dunes 3.4.5 Aeolian transport 3.4.6 Evaluating soil erosion 3.5 Soil degradation 3.5.1 Changes in soil chemical properties 3.5.2 Changes in soil physical properties 3.5.3 Degraded soils 3.5.4 Soils under irrigation 3.5.5 Evaluating soil degradation 4 SOIL PROUUCTIVITY AND REPLENISHMENT 4.1 Yield trends 4.1.1 The CCDE view 4.1.2 The ADP view 4.2 Nutrient removal in crops 4.2.1 Major nutrients 4.2.2 Minor nutrients 4.3 Nutrient replenishment from the atmosphere 4.3.1 Air particulates 4.3.2 Nitrogen fixation 4.4 Fallowing 4.5 Manuring 4.5.1 Benefits of FYN 4.5.2 Use of FYN 4.5.3 Supply of FYM 4.6 Fertilizers 4.6.1 Rates of use 4.6.2 Use efficiency 4.6.3 Soil properties 4.6.4 Prices and profitability 5. ADAPTATION 5.1 Farmer response to soil degradation 5.1.1 Abandoning land 5.1.2 Continuing as before 5.1.3 Land improvement 5.1.4 Degradation and poverty 5.2 Farmer response to drought 5.3 Fadama farming 6. THE POPULATION FACTOR 6.1 Population growth 6.2 Intensity of land use 6.2.1 Intensive systems 6.2.2 Less intensive systems 6.2.3 Extensive systems 6.3 Alternative pathways 6.3.1 Theoretical alternatives 6.3.2 Intensive systems 6.3.3 Less intensive systems 6.3.4 Policy 7. CONCLUSION 7.1 Summary 7.2 The role of inorganic fertilizers 7.2.1 Profitability of IF 7.2.2 Suitability of IF 7.2.3 Farmyard manure 7.2.4 Legumes 7.2.5 Other atmospheric additions 7.2.6 Soil degradation ACKNOWLEDGEMENTS REFERENCES NAPS APPENDIX I APPENDIX 2: TERMS OF REFERENCE LIST OF TABLES 3.1.2 Average monthly rainfall for June, July, August and September (1968-87). 3.1.4 Rainfall and estimated crop losses in Kano State, 1987. 3.3.4 Land use change at Dagaceri, NE Kano State 3.4.3a Comparison of eroded material vith original soil, Samaru. 3.4.3b Loss of nutrients by erosion, Samarn. 3.5.1a Effect of cultivation on amounts of organic carbon ia topsoil, Samaru. 3.5.1b Nitrogen loss under cultivation, Samaru. 3.5.1c Effect of cultivation on soil chemical properties at two sites in Senegal. 3.5.2 Changes in soil physical properties under continuous cultivation on farms near Zaria. 4.2.1 Nutrient removals in groundnut pods and grain. 4,4 Soil properties according to land use. 4.6.2 Fertilizers in Bauchi ADP Northern Zone, 1984. LIST OF MAPS 1 Mean Annual Rainfall 1968-1987 (mm). 2 Provisional Map of Soil Degradation Risks, 1973. 3 Intensity of Farming, 1977-78. 1. INTRODUCTION Soil fertility in the northern states of Nigeria has been maintained over long periods of time by the use of organic or farmyard manures (FYM) and fallowing. Under the impact of rapid population growth and the expansion of the market, these systems are coming under strain with a possible adverse impact on crop yields and soil conditions in the longer term. This study is intended to analyse the linkages between soil degradation and its causes, its nature, its impact on production, farmer response to the problem, and the role which inorganic fertilizers (IF) may play in stopping or reversing the loss of soil quality. The study is a preliminary review of available evidence. This evidence is considered to be inadequate to support firm judgements on many aspects of the problem, and more research is urgently needed, in particular to generate data on the nature and rate of soil degradation in the study area (for Terms of Reference, see Appendix 2). The area covered by the study corresponds to the following States: Kano Bauchi (the Northern Zone of BASIRDA) Borno (north of Lat. 110 N.) This area has been characterised as the Dry Zone of Nigeria (Mortimore, 1989:10), and corresponds to the semi-arid and arid bioclimatic zones (UNEP, 1977). - 1 - 2. RESOURCE MAPPING 2.1 Rainfall (Mao 1). The most critical parameter of climate in the Dry Zone of Nigeria a the rainfall, which diminishes northwards along with the length of the growing season. The synoptic weather stations of the Federal Meteorological Service recorded average annual rainfall ranging from 411mm (Nguru) to 967mm (Zaria) during the period 1968-1987. These averages are significantly lower than those received during earlier lecades. There has been a corresponding shortening of the average growing season. Drought (however de ined) has been a persistent characteristic of the rainfall regime since the 1970s. Rainfall is considered further in Section 3.1 (below). 2.2 Soils and degradation risk Over most ot the northern savannas of Nigeria, the soils are predominantly sandy in texture, posing problems of moisture control, and low in organic matter, N and P (Jones and Wild, 1975). They are neutral or moderately acid, and have low CEC values which are associated with a scarcity of clay. They therefore have low fertility status, but they are nevertheless capable of producing more than they do now. They may degrade rapidly under conditions of intensive rainfall. Water enters the soil rapidly and nutrients tend to leach down through the profile. The key to successful agricultural operations in the indigenous farming systems has been correct management of moisture and efficient use of nutrients, especially the N-flush which occurs after the commencement of the rains. Soils with a higher proportion of silts and clays, better fertility status and higher moisture levels (even flooding during the rainy season) are confined to the flood plains of river systems (past and present), old lake beds, and localised depressions such as interdunes. The risk of soil degradation, assessed on the basis of climatic parameters (UNESCO, 1977), is shown in Map 2. The recent geological history of the soils should be borne in mind. The great majority of the sandy upland soils of the study area are developed from parent materials of aeolian origin - dunefields or loessic plains - which became stable at different times in the past. Some of the soils are so young that they show little profile development. Nevertheless, crusting and sealing at the surface may affect even sandy soils, owing to the loss of structure, impeding cultivation and the growth of grasses and shrubs (Ahn, 1977). In the present study, a distinction is made between soil erosion (by wind or water action) and soil degradation in situ under agricultural management. Soil erosion is considered - 2 - further in Section 3.4 below and soil degradation in Section 3.5. 2.3 Population and land use It is impossible to map the distribution or density of population in the study area, owing to the absence of census data after 1963. The practice of projecting population distribution forwards from that year at an assumed standard rate of growth would be misleading for the purposes of the present study. Such a procedure assumes that rural population patterns have remained stable for 25 years, and enough is known about rural population growth and movements to call such an assumption into question. In addition to this objection, total population gives an increasingly inadequate indication of the pressure on agricultural land. The percentage urbanised is growing; the non farm sector in rural areas is increasing; patterns of short-term spatial mobility are increasing in complexity; and food imports from outside the study area are increasing in volume. It is therefore preferable to try to map land use intensity directly. Map 3 is derived from the Land Use and Vegetation Map of Nigeria, scale 1:250,000, which was based on side-looking airborne radar survey (FDF, 1977). Farmland was in three intensity classes - >60%, 30-60% and <30% - which provide a direct approximation of pressure on agricultural land. During the last ten years, and especially since the Structural Adjustment Programme is introduced in 1986, the growth of the large-scale farming sector has further reduced the relationship between population and land use, arising from the dependence of farming systems on residential hand labour. New levels of profitability, with partial mechanisation, are injecting a new and massive demand for agricultural land. This is satisfied in some areas by the consolidation of existing small holdings, but is commonly met by allocating, under Certificates of Occupancy, community land presently managed under fallow or woodland. This development is too recent to be reflected in Map 3. The population factor is considered further in Section 6 below. -3- 3. EVIDENCES OF DEGRADATION In both public awareness and scientific analysis it has proved difficult to separate the evidences of land degradation from the effects of diminished rainfall. It is therefore mandatory for the purpose of the present study to review the rainfall record, and its implications for surface and groundwater resources (Section 3.1), before examining other types of evidence, This other evidence will be grouped under the following heads: desertification or general environmental degradation (Section 3.2); vegetational degradation (Section 3.3 soil erosion (Section 3.4); and soil degradation (Section 3.5 . 3.1 Rainfall 3.1.1. Drought The persistence of drought, and a diminution o rainfall, have characterised rainfall regimes in the sub-Saharan zone of Africa since the 1960s (Farmer and Wigley, 1985; Hare, 1985) Some authorities argue that a secular trend can be discerned from the statistics (for example, Thambyahpillay, 1987, in respect of Lake Chad Basin). A more cautious view accepts the evidence of lower rainfall, without commitment to trends or cycles, which could be used as a basis for predicting future rainfall. The answer to this question is obviously important for land use management, and the absence of a proven method for forecasting injects unavoidable uncertainty into agricultural, livestock and forestry development plans. Some progress has been made in forecasting seasonal rainfall in the region as a whole on the basis of 'teleconnections' with global sea surface temperatures (Parker et al, 1987), but forcaasts for smaller areas, and a longer time-range, remain problematic. 3.1.2. Monthly distribution. Table 3.1.2 displays the average monthly rainfall for the four critical growing months of June, July, August and September at each of 10 synoptic stations in or near the study area. The averages for the period 1968-87 are shown as percentages of the averages for preceding periods (a mirimum of 19 years, usually ending in 1969: after Kowal and Knabe, 1972). At the six drier stations, June rainfall in 1968-87 averaged only 71% of that received during the preceding period; July rainfall, 84%; August rainfall 71% and September rainfall, 78%. These figures indicate the greatest reductions at the outset of the growing season, and in the August peak. Such reductions fundamentally alter the environment of farming systems, and have equally important implications for the growth of natural vegetation, on which the livestock economy and energy system largely depend. -4- TABLE 3.1.2 Average monthly rainfall for June, July, August and Setember (1968-873 as az of that received in the preceding period at 10 synoptic stations in the study area. Station Total (mm) June July August September Drier Stations 1. Nguru 411 60 93 72 68 2. Maiduguria 536 76 74 67 108 3. Katsina 553 65 84 70 87 4. Sokoto 562 75 83 75 61 5. Potiskumb 608 67 87 72 70 6. Kano 678 81 85 69 75 Wetter Stations 7. Gusaua 847 73 93 77 85 8. Yelwaa 913 80 99 97 81 9. Bauch 960 91 105 77 79 10.Zaria 967 73 113 93 62 a 19 years only b 18 years only: For locations see Map 1: For data, see Appendix 1. 3.1.3. The August Peak. During the early 1980s a decline of the August peak was noted at Kano (Adebayo and Mohammed, 1987; Mortimore, 1989: 142). This has been analysed in two ways. First, numerical analysis of the data for 1968-87 shows that July peaks increased from 27% of sta.-:3 in th. i-eriod 1964-75 to 36% in the period 1976-87 (a 33% increase). If the three relatively wet stations, Zaria, Yelwa and Gusau are omitted, the increase was from 21% to 43% (a 105% increase). Taken in conjunction with a persistent tendency for the rains to become established late, such a trend has negative implications for the longer-season crops. Second, graphical analysis of monthly rainfall departures during the period 1968-87 shows a clear negative trend at only half of the stations, and this trend terminated in 1987. In 1988, it is widely agreed, the August rainfall was everywhere exceptionally heavy (official data were still not available in February, 1989). It remains to be seen whether the relative decline of August rainfall turns out to have been a temporary aberration. 3.1.4. Effects on Crop Production. Agricultural drought has a direct impact on crop production; for example, Table 3.1.4 gives estimates of crop losses incurred in 1987 in Kano State, when recorded rainfall in the four KNARDA zones ranged from 63 to 86% of that received in the previous year, and the monthly distribution was also - 5 - unsatisfactory. Although comparison with the previous year does not give any indication of departure from the mean, such fluctuations from year to year are more real to many farmers than longer-term trends. Their size, as indicated in the table, and their economic importance, make it very difficult to establish the direction and scale of longer-term trends. TABLE 3.1.4 Rainfall and estimated crop losses in Kano State, 1987. KNARDA Annual Rainfall Estimated Crop Losses (%) Zone (as % of 1986) Sorghum Millet Cowpea Groundnut 1 83 15 15 10 10 2 63 70 50 60 40 3 86 25 15 25 20 4 66 65 75 80 75 State 76 38 39 44 25 Source: KNARDA (1987) In addition to their direct impact on primary productivity, agricultural droughts may provide favourable conditions for pest incursions, especially grasshoppers, locusts, birds and rodents. In 1987, in grasshopper- infested areas of NE Kano, insect populations of 113,000- 306,000/ha were recorded (KNARDA, 1988a), and in some areas there was virtually no harvest. The decline of groundnut production in the traditional growing areas (which was initiated by the rosette epidemic of 1975) has been attributed partly to a progressive decline in total annual rainfall since 1960 (Yayock and Owonubi, 1985); although the effects of reduced rainfall were compounded by increased labour costs and stagnating prices for the industrial crop. 3.1.5. Hydrological conditions. Rainfall affects both surface and groundwater resources, and this in turn affects the growth of crops and natural vegetation adversely, where the soil moisture falls to levels below the rooting zone of perennial plants. The magnitude of such changes, and their impact on vegetational degradation, has rarely been measured, one of the problems being the scarcity of surface height data against which groundwater depths may be accurately monitored. Measurements in parts of Kano State (WRECA, 1985) appear - 6 - to show substantial, but not universal, falls in the water table; but in almost every river basin, the surface and sub-surface flow had been affected by the construction of dams (Mortimore, 1989). The fluctuating level of Lake Chad, which forms the base level for surface drainage from nearly two thirds of the study area, provides an indicator of hydrological conditions in the Chad Basin from which conditions in the rest of the study area may be inferred. In 1984, following gradual decline since about 1970, this level fell below the intake of the South Chad Irrigation Project and pumping had to cease (Kolawole, 1987). In 1988-89 there was an improvement in the level of the south basin of the lake, but in the northern basin a temporary incursion of water in January, following the heavy rains of August, 1988, was quickly lost by seepage. Fluctuations in the discharge of major rivers such as the Hadejia - Komadugu Yobe system profoundly affect the livelihoods and management practices of riverain land users. 3.2 Desertification. 3.2.1. The 'Desertification Consensus'. There is a general consensus amongst government officers in the states visited, and especially among forestry officers, that general environmental degradation is occurring (see Sagua et al. 1987). It can be stated that this consensus reflects a view of environmental change widely held by responsible community leaders down to the village level, and by many individual resource managers. However the holders of these views freely admit a want of empirical data on the rate of progression of these processes. The desertification consensus, 'as perceived by the 11 Governors a4 the meeting', is summarised in a Report of the Consultative Committee on Desert Encroachment (CCDD, 1988). In a state-by-state survey of the northern states, the following indicators of desertification are cited (in diminishing order of frequency): (1) declining rainfall, with persistent drought and associated dust storms and low humidity; (2) soil erosion by wind and water, loose or fragile soils, and active sand dunes; (3) deforestation, scanty vegetation, tree mortality, the invasion of dryland species, and fodder scarcity; (4) diminishing groundwater, and the drying of surface ponds and streams; (5) falling farm yields (discussed below); (6) southward migration of livestock producers and/or farmers; and (7) food scarcity and increased human mortality. - 7 - 3.2.2. Administrative Problems Some insight into official perception of the causes of this desertification, and the interplay of environmental change, land use management and the shortcomings of government in failing to arrest such degradation, is provided by reports from the states of the problems encountered in afforestation efforts. In diminishing order of frequency, these are: (1) limited funds equipment and logistical support; (2) inadequate policing of forestry law; (3) overgrazing; (4) bush burning; (5) wood-cutting; (6) shifting cultivation; (7) land-grabbing by large-scale farmers and others; (8) inadequate land use policy; and (9) an assortment of problems mentioned only once: low survival rates of plantation trees, poor plantation sites, inadequate forestry extension, population pressure (Kano State), and ignorance of the environmental consequences of land use practices. The perceived failure of many government shelter belt and afforestation schemes is a significant part of the desertification consensus, and is firmly based in the experience of forestry officers. Since the 1940s the most serious failure was that of the Arid Zone Afforestation programme, commenced in 1978 and effectively abandoned a few years later. Weaknesses in the design, implementation, and sustainability of forestry programmes are not easily disentangled from rainfall and other environmental changes. 3.2.3. Affected Areas. On the basis of the review of forestry, agriculture and water resources in the 11 states, the Report identifies three zones: (1) Severely affected areas (Sokoto, Katsina, Kano, Borno and northern Bauchi States) where from 55 to 75% of the state areas are estimated to be subject to desertification; (2) moderately affected areas (southern Bauchi, Gongola, Plateau and Kaduna States (30-45% affected); and (3) slightly affected areas (Niger, Kwara and Benue States). No explanation, however, is given of how the area affected was estimated. This is a serious difficulty in view of the number of different indicators employed. 3.2.4. Evaluating the 'desertification consensus Studies of some small areas by undergraduates in university Geography departments tend to provide grassroot confirmation of the desertification consensus. Yari (1981), 1or example, described the dimensions of environmental degradation, as perceived by villagers in northern Katsina State, as: increasing shortages of wood fuel, construction timber, farmland, and grazing land; vegetational degradation and tree mortality; the movement of surface soil; localised gully erosion; and increased mobility of the human and livestock populations. - 8 - The desertification consensus is much influenced by the forestry profession's view of environmental degradation, and the difficulties experienced in afforestation programmes which, as the C.C.D.E. report admits, are the result of interacting administrative, rainfall and management factors. Declining trends in crop yields, although widely reported, are difficult to substantiate in reliable figures and impossible to consider independently from the effects of rainfall (see Section 4.1 below). Owing to the power of the consensual perception of desertification, none of the studies reviewed has attempted to demonstrate empirically the linkages between causes and effects. Since the public awareness of desertification at all levels is influenced by the media, as well as by experience, and diffused by means of social networks, it may be self-justifying and should be accepted with caution. The scarcity of data for measuring environmental change weakens the 'desertification consensus'. 3.3 Vegetational degradation. Ecological arguments for vegetational degradation centre on four pra.tices, all having direct or indirect links with population: burning, grazing, woodcutting and farming. 3.3.1. Burning. The effects of fire on savanna woodland are well known. Notwithstanding the widespread occurreas. of relatively stable plant communities under annual burning regimes, bush-burning receives much blame for vegetational degradation, and some states have tried to ban it. There is an ancient controversy concerning the effects and optimal management of bush burning in northern Nigeria, going back to middle colonial times. An entirely negative view of burning is not supported by the evidence. Some loss of organic matter (including N and S), increased water loss, and physical damage to the trees are compensated by positive effects: the regrowth of perennial species, increased availability of P, K, Mg and Ca in the soil, enhanced destruction of microbial-born diseases, the removal of dead vegetation (which may impede germination or cause fire risk), and improved control of the natural vegetation (Penning de Vries and Djiteye, 1982). Early burning is less destructive than later in the dry season. A balance sheet for burning in the study area cannot be attempted on the basis of available evidence. With the transfer of more land to permanent or annual systems of cultivation, the extent of bush burning is bound to decline, though crop residues may continue to be disposed of in this way. Burning also tends to diminish in frequency and intensity northwards in the savanna zone. - 9 - 3.3.2. Grazing Grazing may reduce biomass, species diversity, percentage grass cover, and average height of the vegetation, and increase the percentage of annual grasses, invasive dryland species or poorly nutritive species in the plant community. Allegations of overstocking are frequently made, sometimes on the basis of quite subjective estimates of carrying capacity. But annual variability of rainfall, and of primary productivity, reduce the value of such estimates for practical purposes. The migratory grazing circuits of many herds made it all but impossible to relate forage production on given areas of rangeland to livestock management units on a year-round basis. Awareness of deteriorating rangeland quality is often expressed by livestock producers, but it is more common for it to be attributed to rainfall diminution than to overstocking. The recuperative capability of rangeland in Sahelian environments, given protection is also inadequately known (Peyre de Fabregues, 19845. It seems premature to conclude irreversible damage from field evidence of rangeland deterioration in the short term. The agricultural value of land under heavy grazing pressure may however be expected to decline if the organic matter content of the soil is not maintained. 3.3.3. Woodcutting. Woodcutting, commonly described as 'indiscrmlnate '-reduces biomass and exports nutrients from the ecosystem. Its effects, however, only become noticeable where offtake exceeds the rate of natural regeneration of the species concerned. In the study areas, both conservationary and degrading woodland management regimes are found. Studies in the Kano region (Cline-Cole et al., in press) suggest t at rural populations having densities in excess of 200/km can supply their requirements for wood fuel from farm trees at densities of 12-15/ha, and that as population density increases, the rising value of land and of tree produce (including fuel) tend to ensure a transition from bush fallowing to a permanent farmed parkland. In the Kano Close-Settled Zone, traditionally the source of urban Kano's wood fuel, the tree density and timber volume actually increased between 1972 and 1981 (a period including two major drought years), according to air photo interpretation by Nichol (Cline-Cole et al., in press). Consumption of wood fuel (per capita) in urban Kano is low, relative to estimates for some other areas, reflecting its high price, but alternative energies have failed to replace it. The aggregate demand of such a city (population about one million) is still large (300-500,000 - 10- t/yr). Urban demand is deflecte4 away from the valuable multi-purpose trees of the farmed parkland to distant natural woodland and fallows. It poses a threat to the future of natural woodlanos, whose annual incremental growth may be less than lm /ha/yr (Jackson et al., 1983; Papka, 1984). It is in areas of lower (but rising) population density, therefore, that the greatest threat to the rural woodstock is found. Here intensive agro-forestry systems with livestoak cannot be supported by the family labour available. Increasing distances have to be walked to secure wood fuel supplies (e.g. Papka, 1984; Nggada, 1988). At the same time,the need for supplementary income induces cutting fallow and natural woodland for sale to urban wood fuel traders. Organised depredations in forest reserves are commonplace. In woodland in western Borno (Fuve LGA), for example, up to 80% of the mature trees have been cut, and much of the varied woody flora replaced by Guiera senegalensis scrub: a plant normally associated with tallows, whose regeneration is assisted by its unpopularity with grazing animals. Woodcutting is ineffectively controlled, and Borno's arbitrary ban, imposed in 1986, is a dead letter (Madaki, 1987). Incremental growth diminishes northwards. Papka (1984) projected a fuel wood scarcity in the Sahel Zone of Borno by the year 2,000. 3.3.4. Farmin&. Farm clearance is the principal cause of deforestation, although selected trees are normally protected on small holdings. Abundant material for the measurement of land use change is available in neglected air photography stored in Federal and State Government survey departments and other agencies, but very few attempts have been made to use it for this purpose. In a small area of 25 kmL in NE Kano State (Mortimore 1989: 105), the transfer of woodland, grassland and fallow to arable proceeded at an accelerating rate between 1950 and 1981 (Table 3.3.4). These data are representative of a larger area. In a study of two drainage basins in Gongola and Bauchi States, Patrick (1987) recorded a 21% increase in the cultivated area from 1964 to 1978 in the Lankoviri basin, Gongola State, and a 37% decrease in the area of savanna woodland; fallow was negligible. In the Kaltungo basin, Bauchi State, however, the cultivated area decreased slightly and savanna woodland increased. Such contradictory trends call for more measurements, and these examples show that the possibility exists of quantifying such trends. - 11 - TABLE 3.3.4. Land use change at Dagaceri, NE Kano State(%) 1950 1977 1981 Woodland, grassland, fallow 71.3 65.6 60.4 Arable 28.5 34.2 39.4 Other 0.2 0.2 0.2 Total 100.0 100.0 100.0 Population growth still contributes to the rate of farm clearance, but as important now are population movements from one area to another, and the expanding activities of capitalised large-scale farming enterprises. Clear felling is often practised by such farmers and may expose the soil to erosion, particularly at the beginning of the rainy season. The transfer of land from woodland to farmland does not necessarily indicate degradation in the sense of loss of productivity. Such degradation depends on the management. But the loss of trees in areas subject to rapid agricultural expansion makes a powerful visual impact on the public awareness of environmental change. It is not surprising, therefore, that deforestation tends to be equated with desertification in some official statements, in the media, and in public opinion. 3.3.5. Evaluating vegetational degradation. Burning, grazing, woodcutting and agricultural expansion are not incompatible with sustained productivity even under semi- arid conditions, given the right management. The purpose of the foregoing review is not to question the logic of the ecological arguments for degradation but to suggest that the cause-effect linkages which are widely accepted have often been oversimplified and have hardly ever been subjected to testing and measurement in representative field situations. Nevertheless, they cannot be ignored, because soil fertility is sustained by its vegetation. 3.4 Soil Erosion. 3.4.1. Erosion hazard. Erosion hazard assessment has been included in some regional soil inventories in the study area, whose main objectives however were the classification of the soils and capability or suitability assessments (e.g., FAO, 1969: LRD, 1972; LRC, 1979). Comments on erosion risk are included in the State Reconnaissance Soil Survey Reports of the FDALR (1985a, b, -12- m, d). But the assessment of erosion rates of loss over time at the regional scale is still 'at the stage of reconnaissance. Such reconnaissance surveys, some superficial, have been traced for: nort-western Nigeria Ologe, 1975); Borno (FDALR, 1986/87); Katsina State NCEP, 1988); Sokoto-Rima Basin (SRRBRDA, 1987); and Kano MANR, Kano, 1985). The existence of erosion by wind or water is generally accepted as self-evident, and no attempts to quantify it at the regional level have been traced, whether by Federal or State Government, by RBDAs or by the ADPs. Even the mechanical conservation projects undertaken by the FDALR, and by some RBDAs, rarely involve measuring rates of erosion, either at the feasibility study or the implementation stages. 3.4.2. Gully Erosion. Rates of gully erosion during one or more seasons have been measured in several small-scale studies (e.g. Ologe, 1972, 1974; Olofin, 1980; Patrick, 1987). In Gongola State (Patrick, 1987: 259, 262), rates of headcut advance ranged up to 32 during a3single season (1985), destroying an area of 126m and 149m of soil; and rateq of gully s4de recession ranged up to 1.4m, removing 70 m' and 140 m of soil. Gully erosion rates, however, are highly variable, subject to many controlling factors, and incapable of generalisation over large areas. Locally the loss of soil may be spectacular but no basis yet exists for a quantitative estimate of regional soil loss by gully erosion. 3.4.3. SW . 1. 'oil wash is considered to be more important than gully erosion at the regional scale. Kowal (1970) calculated surface runoff and soil loss on run-off plots on gentle slopes on an experimental. farm. The average loss was found to be 10.8t/ha/yr, representing 1cm in 15 years. Such a rate may be considered acceptable on deep soils, but not where a shallow sesqui-oxide-rich horizon is thereby exposed to hardening (Jones and Wild, 1975:63). Furthermore the eroded topsoil contains a large proportion of the fine material (Table 3.4.3a). Nutrients carried away from these fertilised plots in the runoff water exceeded those removed in the eroded soil (Table 3.4.3b) indicating that soil conservation cannot stop all losses. TABLE 3.4.3a Comparison of eroded material with original soil Samaru. Category Eroded material % Original soil % Sand 63.3 80 Silt ) 6 Clay & Humus ) 36.6 15 Source: Kowal (1970) quoted in Jones and Wild (1975:65) -13- TABLE 3.4.3b Loss of nutrients by erosion, Samaru (kg/ha) Na K Ca Mg N Runoff water and suspended load 1.6 6.3 4.9 2.1 7.4 Eroded soil 0.3 1.1 3.2 1.2 6.3 Sources as Table 3.4.3a. Few studies have been traced which permit these results to be tested under different conditions of soil, rainfall, and management, though it seems possible that on small- holdings, losses may be substantially greater. Three small-scale studies estimated average seasonal rates of soil loss of 13 t/ha in the Northern Guinea savanna (Ogunsanya, 1986), 7 t/ha in flat alluvial and sandy soils in the Hadejia River basin (Abenemi, 1986), and 3.6 cm (1.0 m in 30 years) in eroded soil near Samaru (Ilani, 1981). For comparison, in the Southern Guinea Zone at Bida, Zakari (1985) estimated average sediment yield from splash and run off to be 214 t/ha/yr, ranging from 62t on slopes of less than 10 to in 426t on slopes steeper than 190, 3.4.4. Dunes. Active dune systems are localised, notwithstanding claims to the contrary. The largest area affected is the Manga Grassland of north-west Borno, a grassland area of about 9,000 km4 overlapping into Niger. Moving dunes have been observed there since 1937, but under heavy grazing pressure, they increased by 31% between 1950 and 1969 (still less than 1% of the surface area), and by 1986 under the impact of reduced rainfall in the 1970s and 1980s, increased to as much as 20% of the surface area in severely affected localities. According to one interpretation (Mortimore, 1989: 157f.), this former dunefield was stabilised in very recent geological time. Grassland communities have not yet been followed by a woodland succession, and the soils have exceedingly weak profile development. It should not, therefore, be expected that neighbouring areas, where the presence of more organic matter in the surface soils and a developed woodland community indicate a longer history of stabilisation, will follow the same path, notwithstanding the grazing pressure and reduced rainfall. Dunes have also been observed north of Sokoto, where coarse sandy soils have developed over sedimentary sandstones. The recent increase in dune formation here also coincided with dry conditions, and is related to the loss of vegetation and to topographical conditions -14- affecting windflow. No detailed study of these formations has been seen, but the local variability of the geological and soil conditions suggests that these, too, may not be indicative of trends in the region as a whole. 3.4.5. Aeolian Transport. Rates of soil loss by aeolian deflation have not been estimated, either on cultivated land or on rangeland. Wind action is universal on unprotected surfaces during the dry season and during storms preceding rainfall. But the mobilised material may not travel far, before being deposited in heaps around obstructions (often the first indicators of deflation to be recognised). In a given area, therefore, the measurement of gross and net loss will be necessary before the impact of wind action can be accurately assessed. In some areas (such as Danbatta in northern Kano state), live field boundaries obstruct the transport of soil, and saucer shaped depressions are created in each field. But weak profile development in such soils, which are developed on aeolian sands, may disguise any evidence of truncation. Wind action is selective, removing the clay and silt fraction, but this fraction is already very low in most of the soils affected. It is certain that aeolian action redistributes large quantities of surface material, when soil humidities are low during the long dry season. 3.4.6. Evaluating soil erosion. Available studies may be indicative rather than representative of rates of soil loss by water or wind erosion, and lacking quantitative survey data for the study area, the impact of erosion processes, and the projection of their impact in the future, must be guesswork. However the intensity of erosion is easily observable in seriously affected localities and clearly indicates the urgency of a regional assessment. 3.5 Soil Degradation. The effect of cultivation on the physical and chemical properties of soils are known and have been demonstrated under experimental conditions at the IAR, Samaru (see Jones and Wild, 1975). Few attempts have been made to monitor such effects under smallholder conditions, where a formidable number of management variables need to be controlled, including crop mixtures and rotations, manuring and fertilizer treatments, mulching and residue management, and fallowing. With the exception of some studies of irrigated soils (3.5.4 below), such attempts have been based on comparisons between cultivated and control sites. The core of the present study ought to be a review of time-series data, but no such data has been found for the non-irrigated soils of the study area. The generation of -15- time-series data for monitoring soil fertility should, therefore, be a priority. 3.5.1. Changes in soil chemical properties. Under cultivation nutrients may be lost from the topsoil by erosion and in runoff water, by leaching, through volatization on burning vegetation, and by crop removals. Erosion losses have been discussed in 3.4.3 (above). Leaching losses are reduced under vegetation cover owing to increased transpiration, plant uptake and root interception, and so depend on the nature of the crop mixture. Burning a fallow may lose up to 20 kg N/ha. Losses of organic C and N have received most attention. Hot, dry soils cleared for cultivation are liable to enhanced oxidation and mineralization when wetted by the first rains (Jones and Wild 1975: 97-8, 143, 73). Jones (1971) reported the losses of organic C from the topsoil at Samaru under a variety of cultivation regimes, deduced from comparison with 'local bush' soils (Table 3.5.1). In these experiments all residues were removed. The same study reported an annual loss of 25 kg N/ha under continuous cultivation (Table 3.5.1b). Grass fallows (Andropogon gayanus) modified losses of both C and N. Since such losses depend entirely on management, the results from experimental stations give us no quantitative indicators of what is happening on smallholdinis. In Senegal, a study of topsoils (0-10cm) from farmers fields on two soil types showed that organic matter, CEC, exchangeable Ca, K and Mg, and soil water pH, all declined with increasing years of cropping (Table 3.5.1c). No comparable data has been found for the study area, although intuitive expectations would be similar. TABLE 3.5.1a Effect of cultivation on amounts of organic carbon in topsoil (O-15cm), Samaru. Management Years' cultivation %C Percent of control Local bush 0.60 100 No manure or fertilizer, mono- cropping or rotation 20 0.22 37 No manure, some DNPK 10 0.26 43 15 0.21 35 18 0.17 28 Gamba grass, fallow: cropping years ratio 2:3 19 0.24 40 3:3 19 0.30 50 6:3 19 0.41 68 Source: Jones'(1971) quoted Jones and Wild (1975):72 - 16 - TABLE 3.5.1b Nitrogen loss under cultivation, in topsoil (0- i3cm) samaru, % N in soil Annual loss of N Initial After Percent kg/ha or cultivation of Control Control 11th & 18th years of cultivation 0.026 0.017 4.3 25 3 years after clearing 6 yrs grass fallow 0.046 0.034 8.7 99 3 yrs grass fallow 0.036 0.026 9.3 75 2 yrs grass fallow 0.027 0.026 1.2 7 Source: as Table 3.5.ia TABLE 3.5.1c Effect of cultivation on soil chemical proporties at two sites in Senegal, Slightly Ferralitic Leached Ferruginous Years since clearing: 5 16 80 2 15 50 Soil property Organic matter, % 2.4 1.5 1.3 1.5 1.3 1.1 CEC, me/100g 4.7 3.5 2.6 3.5 3.2 2.3 Exch.Ca,me/100g 2.9 1.7 1.1 2.1 1.5 0.7 Exch.K,me/lOOg 0.09 0.07 0.08 0.10 0.08 0.04 Exch Mg.me/100g 1.0 0.9 0.5 0.8 0.6 0.4 pH, water 6.5 6.4 6.0 6.6 6.2 5.6 Source: as Table 3.5.ia 3.5.2 Changes in soil physical properties. In theory, any changes in soil chemical properties can be corrected by appropriate fertilizer management. The same is not true of changes in physical properties under cultivation, which may be pronounced in northern savanna soils:' when the soil is cultivated, physical conditions deteriorate as compared with zero tillage. This has now been reported for the following properties: soil bulk density (except close to the surface), infiltration and soil aggregate stability. Tractor-powered systems have the largest effect on physical conditions, the manual system has the smallest effect while the bullock system is intermediate' (IAR 1981:22; 1982:4). Physical changes have only rarely been investigated under smallholder conditions. Ologe et - 17 - al. (1988) compared soils on aeolian drift 45 km NE of Zaria, under the three management regimes: natL-al woodland (forest reserve), hand tillage and mechanical cultivation. The cultivated plots had been opened up less than ten years before. Compared with the woodland soils, those under mechanical cultivation were found to have coarser texture, lower percentages of water-stable aggregates, lower bulk density and lower equilibrium infiltration rates, at both 0-15 cm and 20-30 cm depth (Table 3.5.2). Mechanical cultivation, however, is a necessary condition for effective crop residue incorporation, which by increasing the organic matter content of the topsoil, enhances exchangeable base content and soil structure. Jones and Wild (1975:60) suggest that infrequent deep ploughing is less damaging than frequent shallow ploughing. TABLE 3.5.2 Changes in soil physical properties under continuous cultivation on tarms near 7aria. Woodland Hand Mechanical Cultivation Cultivation Top Soil (0-15cm) coarse sand (%) 25 26 31 Fine sand (%) 29 28 43 Silt (%) 25 28 14 Clay (%) 21 19 12 Water stable aggregates 0.25mm (%) 58 38 25 Bulk density (g/cm 3 ) 1.4 1.3 0.9 Equilibrium infiltration rate (mm/hr) 58 52 30 Sub soil (20-30cm) Coarse sand (%) 24 22 19 Fine sand (%) 25 24 40 Silt (%) 20 23 14 Clay (%) 33 31 28 Water stable aggregates 0.25mm (%) 56 58 47 Bulk density (g/em 1.4 1.4 1.7 All values significant at 0.05% Source: Ologe et al1 (1988). 5 185- 3.5.3. Degraded Soils. Abandoned farmland occurs in heavily cultivated areas and especially in Kano, where it was the subject of some early experiments in rehabilitation by the Ministry of Agriculture (Dennison, 1956). Such soils are commonly devoid of vegetation owing to surface sealing or hardpan. More recently, some degraded sandy soils have been identified at the Danbatta Farm Centre, and experiments with dung and inorganic fertilizers have been repeated over a number of years (KNARDA, 1983, 1984, 1985, 1986, 1989). The soils have responded only - and then slowly - to dung treatments.The poor response to IF except in the presence of FYM, and the need for several years' treatment before normal yields can be achieved, have potential relevance for fertilizer policy on such soils. The significance of this study for our present purposes, however, is reduced by an absence of information on the history of management before the soils were abandoned and on their representativeness for small farm conditions in the area. The latter may be doubted, since neighbouring farmers' fields were found to be more fertile than the soils of several Government farm centres here and elsewhere (IAR, 1984, 1987)! 3.5.4. Soils under Irrigation. Some soils under irrigation are subject to salinization and to physical degradation. Reviews of salt-affected soils are given by Allison (1962), FDALR (1980) and Maurya (1982), including soils in the Sokoto-Rima, Hadejia, Komadugu-Yobe and Lake Chad basins. Soils on the Yau Irrigation Scheme in northern Borno have been seriously affected by salinity since the 1960s, but it has not yet been found necessary to withdraw them from cultivation. Insufficient irrigation or poor drainage have been blamed for salinity problems. Daniel (1987) and Daniel and Essiet (1989) found that on irrigated soils in the Tomas scheme in northern Kano State, the effect of irrigation was to reduce the percentages of silt and clay, and with them, many nutrients, notably exchangeable K, Ca and Mg, CEC and organic C. Increases in available P and K resulted from fertilizer application, but leaching was intensified by the increased sandy texture of the soil. Essiet (1989), in a study of irrigated soils at six locations in southern Kano State, found that while the clay and silt percentages had fallen under irrigation (by 61 and 38%), increases were recorded in exchangeable Na, K, Ca, Mg and CEC. A third study by Kodiya (1988) of soils on the South Chad Irrigation Project showed no significant changes in physical or chemical properties after ten years of irrigation. These soils had a texture dominated by clay. Such differences suggest that there is considerable variability amongst irrigated soils, not merely in their -19- starting properties but in their management (especially fertilizers) and response to irrigation, and caution should be exercised in drawing conclusions. 3.5.5. Evaluating soil degradation. From the results of experiments and from what is known of soil management under smallholder conditions, it appears probable that the deterioration of soil chemical and physical properties in the study area is significant, but variable. Of the two, changes in physical properties are potentially of greater importance because they are not amenable to correction by IF management. - 20 - 4. SOIL PRODUCTIVITY AND REPLENISHMENT 4.1 Yield Trends. 4.1.1. The CCDE View. The Consultative Committee on Desert Encroachment (CCDE, 1988) reported declining crop yields in six states (Sokoto, Katsina, Kano, Bauchi, Kwara and Gongola) and for all the major grain crops (sorghum, millet and maize). In 1987 (a drought year) yields were estimated to be down by 30% in Bauchi (millet), 65% in Gongola (maize), 85% in Sokoto (sorghum), 40% in Kwara (maize), and 37-39% in Kano (sorghum, millet and maize), and production of all crops by 80% in Katsina. The datum for these comparisons, and others like them, is 'normal', 'former', 'expected' or 'average' yields. Given the difficulties of obtaining reliable yield data under Nigerian conditions, and the reliance on 1987, the CCDE estimates cannot be accepted as indicative of trends. Such estimates do not provide an objective basis for statements about declining farm yields per hectare - independent of the effects of rainfall. 4.1.2. The ADP View. Some statements from the ADPs have reported increased production of major crops during the 1980s. For example,on the Gombe Integrated Development Project, yields of maize with IF increased from an average of 740 kg/ha in 1977 to 1414 kg/ha in 1980, and millet yields from 415 to 799 (according to APMEPU data quoted in Adesiyan,1985:160), and yields without IF also increased. On less productive soils, the Northern Zone of Bauchi ADP reported a 16% increase in total grain output between 1983 and 1984, of which 12% was due to increased hectarage, implying that yields (about 1 t/ha) rasL s.,titly ",3ADP, 1985). These years were both dry (80% and 93% of average rainfall at Bauchi and 86% and 76% at Potiskum). But the CCDE (1988) reported that millet yields in Bauchi State have fallen by 30% from 704 kg/ha formerly to 477 kg/ha at present (probably 1987, also a dry year). It is not possible to reconcile conflicting representations of yield trends. The ADP reports are calculated from agronomic survey data and serious doubts have been cast on the reliability of such data. In any case, individual crop yields are difficult to evaluate in view of the practice of growing crops in mixtures. Rainfall fluctuations disguise trends. ADP performance cannot yet be assessed in terms of yields. The information available on yield trends gives no reliable indication of the rate of soil degradation. Kano (KNARDA) data is more ambiguous. Estimated sorghum yields in Kano State averaged 704 kg/ha in 1982, 409 kg/ha in 1983 (a drought year) and 562 in 1986; and those of millet, 774, 503 and 600 kg/ha. By comparison, average - 21 - yields of sorghum on KNARDA seed multiplication farms were 660 kg/ha in 1983, and 1,500 kg/ha in 1986; and of millet, 680 and 1,500 kg/ha. 4.2 Nutrient Removal in Crops. 4.2.1. Major Nutrients 'One form of biological degradation seldom considered is the total loss of nutrients from a country through the export of its agricultural products' (FAQ, 1974: 20-22). For nutrient loss from a particular farming syste-n, it is sufficient that crops are exported from the locality. The amounts of nutrients contained in the grain, residues and roots of the major crops are known,if inadequately (Jones and Wild, 1975:165). Using assumed yields, nutrient balance sheets have been calculated for different fertilizer regimes in Senegal (Charreau and Fauck, 1970: Tourte, 1971; reported in Jones and Wild, 1975: 166, 211). Given reliable estimates of areas, yields and sales, it should be possible to construct, for northern Nigeria, a 'nutrient audit' in respect of particular crops or farming systems. In an early study, Watson (1964) calculated that the amount of phosphorus in crops sold off the farm in northern Nigeria was 2 kg/ha for a 500 kg crop of groundnuts, 1 kg/ha for a 250 kg crop of seed cotton, and 3 kg/ha for a 1,000 kg crop of grain; and that about 25,000 tons of single superphosphate were then needed annually to replace the phosphate removed in northern Nigeria s exported crop of groundnuts. This is consistent with estimates of nutrient removal (2.7-4.0 kg P/t of kernels) reported by Jones and Wild (Table 4.2.1), but Harkness and Yayock (1979) recommend a fourfold increase to 100,000 t of single superphosphate, on a similar area - about lm ha of groundnuts - presumably to allow for removal of residues and for basal applications for soil improvement. This is equivalent to I million t of rock phosphate. While groundnuts are still sold off the farm (though no longer exported from the country), the residues are also sold as fodder. Lombin (1985) calculated that on average, the kernels contain about 58% of N, 68% of P, 19% of K, 5% of Ca and 22% of Mg while the haulms contain 27% of N, 23% of P, 64% of K, 83% of Ca and 60% of Mg. Complete removal of both kernels and haulms will therefore deplete soil cation reserves, unless these losses are compensated by the use of fertilizers. The implications of the shift from groundnuts to grain marketing, which began with the rosette epidemic in groundnuts in 1975, are suggested in Table 4.2.1. Some - 22 - cereals appear to remove less N,K, and S, but N-fixation does not compensate (as with groundnuts) for removals. Unless this is achieved by increased emphasis on leguminous cowpea (also a market crop) in mixtures or rotations, it must call for increased fertilization in the long run. TABLE 4.2.1 Nutrient removals in groundnut pods and grain Crop N P K Ca Mg S Groundnut (pods) 30-50 2.7-4.0 9 0.5 2 2.6 Millet (grain) 26 3 10 0.6 1.6 2.8 Sorghum (grain) 19-22 2.0-2.4 3.5-4.2 0.5-1.0 1.9-2.7 0.5-0.8 Maize (grain) 9-13 1.9-3.0 2.5-4.0 0.1-0.4 0.7-1.0 - Source: Jones and Wild (1975): 165 A range indicates more than one measurement reported. Yayock et al. (1978) and Balasubramanian et al. (1978), using data on yields and nutrient uptake from experimental stations, the Federal Office of Statistics' estimates of the areas under the major savanna crops (for 1970-71), and suggested application rates from IAR experiments, were able to approximate the total requirements of N, P and K for millet, sorghum, maize, wheat, rice, groundnuts and cowpeas. The yields used were higher than those generally obtained on small farms, the estimated areas under the crops now require up-dating (and doubts have been cast on the reliability of FOS data), and no distinction could be made between nutrients recycled as incorporated residues or dung and those exported off the farm. The estimates were also limited to the three major nutrients. Some imvrovement in the data available will be necessary before a nutrient audit' can be achieved for farming systems in the area as a whole. 4.2.2. Minor nutrients. More detailed studies have been carried out on some other nutrients. For example Yayock and Lombin (1981), in a review of work at the IAR on sulphur in northern Nigerian soils, report that S- deficiency is becoming widespread because of inherently low values and the loss of organic matter following cultivation. Sorghum, millet and cotton may each remove 5-8 kg/ha (Bromfield, 1975). But restorative bush fallows are becoming less feasible owing to land shortages, and FYM is not available in adequate quantities to restore S. The scarcity of organic matter in the soil is also related -23- to low Zn levels reported in the northern savannas (Lombin, 1983b), which are also deficient in Mg (Lombin, 1979). On the other hand, Cu and Mn are considered to be adequately supplied (Lombin, 1983a). There is a need for these technical studies to be linked to economic appraisals, based on estimates of total production and the percentage sold off the farm, in respect of the major crops in the study area. According to Jones and Wild (1975:124),' the frequency and severity of occurrence of micro-nutrient deficiencies may be expected to increase as the intensity of agriculture increases'. 4.3 Nutrient replenishment from the atmosphere. Research on atmospheric sources of plant nutrients has concentrated on (1) the chemical content of air particulates, and (2) the fixation of nitrogen. 4.3.1. Air particulates. Bromfield (1974, a,b,1975) measured the deposition of S in dust and rainwater at 10 stations in northern Nigeria. Including the absorption of gaseous-S directly by the soil, the mean estimated annual accession of S was 1.5-2.5 kg/ha, equivalent to the amounts removed by low yielding crops (see Table 4.2.1 above), but, in view of probable leaching losses, low in relation to total crop requirements especially at higher yields (Jones and Wild, 1975:116; Yayock and Lombin, 1981). However, all the S is soluble soon after the first rains and is available when crops are germinating. Beavington and Cawse (1979) analysed the trace elements contained in air particulate at Bagauda (56 km SSE of Kano) and compared total deposition in dust and rainwater with the removal of the same elements in a 680 kg/ha crop of sorghum. They found that the amounts of Cl, In, Mg, K, Mo and Rb deposited were from 3 to 10 times the amounts removed in the grain, and of Br, Ca, Co, Cr, Cu, Mn, Ni, Sm and Zn (and others), even greater. However, the losses by leaching and erosion, and incorporation into leaf and stem tissue, were not known, and the solubility of the elements was variable. More research is needed into the contribution made by these trace elements to plant growth. Using a different (wet) collecting technique, the deposition of Harmattan dust was measured during the months October-April, 1978-86 at Kano (McTainsh, 1980; McTainsh and Walker, 1982; Mortimore, 1989:150). Average annual deposition, corrected for secondary deposition locally remobilised dust) was 830 kg/ha. Wilkes et al. 1984) carried out chemical analyses of the dust. Using their results, annual deposition rates for the -24- macronutrients are obtained as follows: 1.70 kg P/ha, 24.96 kg K/ha, and 3.13 kg N/ha. (These estimates do not take account of nutrients deposited outside the Harmatton season; and the weekly removal of samples prevented their remobilisation). Such amounts are significant when compared with nutrient removals in crops (Table 4.2.1 above) but less so when leaching and total crop requirements are taken into account (for example, only 12- 15% of K-uptake in maize is contained in the grain; Jones and Wild, 1975:120). The Harmattan dust must be supplemented by other methods of soil improvement. 4.3.2. Nitrogen Fixation. Because the N content of savanna soils is low (Jones and Wild, 1975), the dynamics of N during the short growing season are critically important. During the dry season, microbial activity is suppressed, but within 10 days of the first rainfall, N reaches its peak in the top 20 em of the profile (Wild, 1972). Thereafter, leaching carries it increasingly beyond the rooting zone in sandy soils. Early planting is essential. Under experimental conditions, millet (Pennisetum typhoides) takes up 22 kg N/ha in the first 28 days of the season and 49 kg N/ha in the next 21 days, to yield 1,930 kg grain/ha (Blondel, 1971). To maintain agricultural yields in such soils, biological fixation of N is necessary. In two years under fallow vegetation, N may increase by 200-350 kg/ha, and under cultivation, it may decrease by 100-700 kg/ha (Adu and Nnadi,1987).But fallowing is giving way to annual cultivation, and non- symbiotic fixation by free-living microbes consequently becomes secondary to symbiotic fixation by leguminous crops. Legumes may fix up to 60 kg N/ha. The advantages of legumes such as groundnut and cowpea in rotations with the grain crops have been acknowledged under experimental conditions (e.g. Lombin, 1981a). Small farmers, however, mix legumes with grain crops, and there is some uncertainty as to whether current transfers of N oc,ur between crops in the same mixture (as opposed to residual benefits). Forage legumes are being advocated in mixtures with cereals for farming systems with livestock, improving the feed value of residues (Mohammed-Saleem et al., 1988a); Nnadi and Haque, 1986), and also in fodder banks now being introduced in some grazing schemes in northern Nigeria (Mohammed-Saleem et al; 1986b), where livestock owners are encouraged to plant cereal crops in following years. Legumes have potential value for controlling soil degradation (Nnadi, 1983). The introduction of pasture legumes enriches the N content of soils too deficient in N to support good grass growth, and too low in productivity to justify fertilization. Legumes have been used to rehabilitate degraded soils in experiments in Kano (Dennison, 1956). Leguminous trees may fix 40-50 kg N/ha. -25- Under the canopies. of such trees as Acacia albida, total N organic matter, available P and K and CEC are all higher than outside the canopy (Nnadi, 1983; Papka 1984; Umar, 1981; Baita, 1986; Abdu, 1987) - though the lack of N-fixing nodules on the root systems remains an enigma (Jones and Wild, 1975). It seems certain that as the demands made on the N- deficient soils of the study area continue to increase, the management of legumes will become even more important. Leguminous plants occur in the natural flora which can be used to enrich rangeland; leguminous crops are employed in existing farming systems; and leguminous trees are prominent among economic species conserved on farmland. Raising the productivity of semi-arid soils may depend on improving the efficiency of legume management to minimise losses of N (Fada and Rayar, 1988; Rayar and Haruna, 1985). 4.4 Fallowing. According to data from Ghana (FAO, 1971: 37) the beneficial effects of a 7-10 year fallow may last no more than two years and amount to the equivalent of one year's extra grain output. Most fallows in the study area are shorter than this. Land under short fallow may be little more fertile than that under annual cultivation. Singh and Balasubramanian (1977) reported values for surface soil carbon content under different management regimes (Table 4.4). From these they conclude that short grass fallows are not effective in restoring soil organic matter to the level found in virgin soils prior to clearing, and that continuous cultivation with fertilization maintains organic matter at a higher rate than on adjoining grass fallows. Analysis of LRC (1979) data (Dirisu, 1983) according to land use gave results shown in Table 5. According to this analysis, cropland has on average a higher organic matter content than woodland, grassland or fallow, although another study of the same data suggested a different conclusion (Bello, 1985).Another study in the Danbatta area (Tayyib, 1981) also showed that short fallows are little different from cultivated soils. - 26- TABLE 4.4 Soil properties according to land 2se. A Land use Carbon B Land use Organic (percent) Matter (percent) 1 'Virgin Land' 1.03 1 Shrub savanna 1.07 Woodland 0.54 2 Grass fallow 0.40 2 Grassland 0.58 3 Agriculture with fallows 0.58 3 Fallow 0.47 4 Continuous Cultivation with fertilizer 0.50 4 Cropland 0.62 Sources: A Singh and Balasubramanian (1977) B LRC (1979) analysed by Dirisu (1983) The balance of these conclusions suggest that there is not much to lose by transferring to a system of annual cultivation in semi-arid areas where fallow vegetation regenerates relatively slowly. In any case, there are reports of bush fallowing systems breaking down. Fallow needs additional labour to bring into use, and labour costs have risen rapidly in recent years. Many poorer farmers have no access to fallow land, being obliged to cultivate all the land over which they have rights. Planted and fertilized fallows, though technical alternatives, seem unlikely to gain wide acceptance. 4.5 Manuring. 4.5.1. Benefits of FYM. According to Lombin (1988) 7.5-10 t/ha of FYM are necessary to meet the N requirements of most crops. Such quantities are not available on a general basis, though applications on individual plots may reach or exceed this level. The value of FYM is universally appreciated in the study area, and low rates of application are considered worthwhile. Various sample surveys report that the percentage of farmers using FYM ranges from 55 to 100, and of those who do not, a number are unable to procure it. A long series of studies at Samaru has investigated the effects of FYM (for references, see Jones and Wild, 1975). Twenty years' application at 12.5 t/ha/yr raised mean soil carbon content four times higher than that of control plots (Jones, 1971). Its beneficial effects on crop yields were attributed to the provision of nutrients and of P in particular (Heathcote, 1971). It may effect a -27- temporary improvement in physical structure and, more permanently, augment the exchange complex and buffer the soil against acidification (Jones and Wild, 1975: 138). FYM has residual effects for at least two years (Baker et al., 1977) and fertility builds up with time under continuous application (Lombin and Abdullahi, 1977). It is superior to mineral fertilizers at higher rates of application (Abdullahi and Lombin, 1978). FYM demonstrated superior restorative effects on degraded soils at Danbatta, compared with mineral fertilizers (KNARDA, 1988). 4.5.2. Use of FYM. Under small farm conditions in south- western Kano State, Hill (1985a) found that 34% of 810 plots received FYM only, and 20% both FYM and IF (24% received IF only and 22% were not fertilized at all). On plots receiving FYM, 55% received 14 t/ha or less and 27% received 15-30 t/ha. Sorghum and millet yields increased with the amount of FYM applied to a peak of 16-20 t/ha. Inorganic fertilizers are most often used in conjunction with FYM, and are considered to carry a greater risk of burning the plants. Essiet (in press) has examined the effects of FYM, collected from fields in the Kano Close- Settled Zone, on soil properties under laboratory conditions. Compared with control samples taken from the fields (which may still show residual effects of FYM applications in previous years), the major benefits of the addition of FYM were an improvement in soil structure and water retentive capacity. Increase in nitrogen was not noticeable. FYM is considered to be more beneficial in the long term than IF, in view of its effects on physical 4.5.3. Supply of FYM. The major constraint on the ;A,-, f FYM is the supply. Small livestock populations increase with the human population density (Hendy, 1977), and household refuse is a constituent of FYM. Nevertheless it does not appear possible to raise FYM production to the levels needed to sustain desirable crop yields on annually cultivated land, owing to limitations on the supply of fodder and the scarcity of land for growing forage crops. In the Kano Close-Settled Zone, where browse from trees contributes to feeding livestock, an integrated system of crop, livestock and tree husbandry has developed over many years. It can no longer support the rising human populations although its environmental sustainability is admirable (Mortimore, 1988). Livestock can only recycle nutrients, and significant losses from FYM occur by volatisation. But elimination of this component of the nutrient cycle and incorporating the residues directly into the soil would also be affected by deterioration during the intense dry season, as well as removing the economically valuable livestock sector from the farming -28- system. FYM is exported from cities to the countryside and its value justifies labour-intensive transport and sorting operations. High prices for agricultural output, together with an increase in the prices of IF, will further increase the pressure on the supply of FYM. 4.6 Fertilizers. 4.6.1. Rates of use. The latest estimates of IF use per hectare are those of Wedderburn (1988). These are derived from gross consumption data on a state basis, and estimates of (a) cultivable and (b) cultivated areas in each state. The derivation of these estimates is not known (FMAWRRD data), but they provide upper and lower limits of 15.3 kg nutrients/ha (cultivated land) and 3.1 kg/ha (cultivable land) for Nigeria as a whole. This is low by standards elsewhere. The percentages of the cultivated areas fertilized (p.31) in some states relevant to the present study are given as follows: Bauchi, 38, Kano, 22; Sokoto, 31 (Borno and Katsina were not analysed). The rates of nutrients used (kg/ha of fertilized land) in 1987 were: Bauchi, 35; Kano, 94; Sokoto, 30. An alternative approach is through farm surveys. Hill's (1985b) study of fertilizer use in SW Kano State provides a series of conclusions on smallholder practice in that State. In terms of the number of farmers using it and the area fertilized, IF was still less important than FYM. Of the plots that received no FYM only 24% received IF, and the remainder were not fertilized at all. When IF was used, on 40-50% of occasions it was in conjunction with FYM. When so combined, IF gave poorer crop response than when used alone; but the plots gave higher total yields, particularly at lower levels of application. IF is applied after thinning and concurrently with weeding; and is probably applied selectively to mixtures. Crop response does not correlate regularly with rates of application, either of FYM or IF. But in 10 villages representing Kano State as a whole, the mean yield of threshed grain on 372 fertilized plots was 1,769 kg/ha, compared with 1,434 kg/ha for unfertilized plots (Hill, 1986). Although the farm survey method avoids the major weakness of the gross consumption approach (controlling for inter-seasonal storage, storage losses inter-state movements and exports), it is still difficult to control for variations in the method and time of application, stand density, crop mixtures, residual effects of previous years fertilization, combinations with FYM, and rainfall (Wedderburn et al., n.d.) Wedderburn (1988: Annex 7, Table 1) reports that 30% of farmers purchased IF in Sokoto, 33% in Kano and 46% in - 29 - Bauchi. There is much variability. In five areas of the northern zone of the BSADP, 26, 39, 64,66 and 67% of households purchased fertilizer (BSADP, 1987). Various small scale studies conducted by students (which may have the advantage of independence from project management) indicate percentages ranging from 30% to over 90% in different parts of the country. Such variability is to be expected, in view of the differences between villages in accessibility, innovativeness, wealth and other variables, and in the preferences for growing and fertilizing different crops. 4.6.2 Use efficiency. These surveys also reveal numerous factors that reduce the efficiency of fertilizer use: such as the fear of burning seedlings, especially when rainfall is deficient; ignorance of the correct quantities to apply; complaints of increased weed growth; lack of sufficient labour time to follow the recommendations forbasal and split maintenance applications; inadequate rates of application; supply problems; and a belief that fertilizers increase the quantity of fadama crops (such as sugar cane) at the expense of quality. The type of fertilizer applied is influenced by supply as well as by recommendations, and the proportions purchased vary noticeably among states (Wedderburn, 1988: Annex 2). NPK, for example, constitutes 53% of sales in Kano, 46% in Kaduna and 28% in Sokoto States; whereas BSP constitutes 1%, 22% and 10% respectively (Jayaraman, 1988). It seems doubtful if the distribution system works well enough to ensure optimal nutrient application in more than a few areas. For example, fertilizer sales in the Northern Zone of Bauchi (BSADP, 1984) suggest that shortages of NPK probably constrained sales (Table 4.6.2), while the under- disposal of stocks of alternative fertilizers may reflect consumer resistance as well as the poor rains and currency change-over experienced during that year. -30- Table 4.6.2 Fertilizers in Bauchi ADP Northern Zone, 1 Jan- 31 Aug. 1984, Sales Stock Sold ( Percent of (Percent) families)* NPK 43 96 SUPA/BSP 24 26 CAN 18 19 DAP 9 24 UREA 7 22 TOTAL 100 34 * N - 516 (3% of farming families). A study of fertilizer use and benefits in Bauchi State (BSADP, n.d.) calculated the quantities of nutrients applied from the tonnages of fertilizers sold between 1982 and 1986, and compared the N:P:K ratio with those recommended for the different crops. The study concluded that had the N:P:K ratio been 2.0:1.0:0.5 instead of 1.39:1.0:0.5, the incremental benefits during the period would have increased by 32%. This agrees fairly well with the finding of Singh et al. (1983) that the optimum applications of nutrients for sole crop millet are 60 kg N/ha, plus 30 kg P/ha (with 30 kg K/ha recommended to maintain soil fertility). Further inefficiencies in fertilizer use result from under-, over- or untimely applications. Hill (1986b) suggests that in SW Kano farmers are not receiving reasonable benefits from high rates of IF application, probably because FYM is being used in combination. 4.6.3. Soil properties. Efficiency from an economic standpoint affects returns, but more seriously in the longer term, warnings have been issued about inappropriate chemical additions to the soil. Continuous use of ammonium sulphate for example, increases the acidity of soils (whose pH may change from 4.5 to 3.4 after three years with 280 kg N/ha), and N fertilizers also reduce exchangeable Ca and Mg (Jones, 1974; Mokwunye, 1977). Use of FYM, on the other hand, lowers the components of soil acidity. Increased use of major nutrients alone will also lead to S, B, Zn and other deficiencies (IAR, 1977). Adding fertilizers on an ad hoc basis to increase yields is therefore hazardous, and may accentuate nutrient deficiencies. Certain nutrients should be recommended as a soil insurance policy, even though they give no -31- immediate increase in yields (Jones, 1976; Lombin, 1981b) But few farmers recognise soil conservation as an objective of IF management. Finally it should be emphasised that IF does little directly to improve the physical properties of the soil, which tend to deteriorate under cultivation, especially mechanical. FYM, or the incorporation of crop residues, however, can correct such deterioration.(IF may, by increasing biomass, enhance the density of roots in the soil and hence organic matter and moisture retention.) 4.6.4. Prices and Profitability. The profitability of subsidized Fertilizers in northern Nigerian conditions (Balcet and Candler, 1982), though rather taken for granted, has been questioned .n relation to the West African semi-arid zone as a whole by Matlon (1987), on the grounds of low response rates of local varieties; high risk; negative long-term impact on soil quality; infrastructural and foreign exchange costs. Singh et al. (1983) calculated that the added income from the use of N:P:K: (at 60:30:30 kg/ha) on sole millet was N149 /ha at subsidised prices and N108/ha at unsubsidised prices (1983). Wedderburn's (1988) analysis showed a deterioration in profitability between 1983 and 1987, owing to poor crop prices and major increases in the cost of fertilizer. Under a range of scenarios it was concluded that most crops could not stand a further major increase in fertilizer prices, the exception being rice. (No analysis was carried out for wheat.) Even under the marginal increase in fertilizer prices, at best a third of the crops analysed would remain profitable. Since this study was carried out, food price inflation has reached new levels. There are reports of farmers paying up to 4 or 5 times the official price for IF as recently as 1988. A small survey of farmers' responses to increased fertilizer prices in Kano State (KNARDA, 1988b) was carried out in May, 1988. One third of the respondents said that they would respond to a further reduction in the fertilizer subsidy by reducing their heetarage; one third said that they would buy less fertilizer, and the rest would use other methods of soil improvement. Given a range of possible prices per bag, the respondents indicated that as the price increases, both the number of the fields fertilized and the rate per hectare will decrease; falling from 100% of fields at 470 kg/ha at a price of N5/bag to 49% of fields at 203 kg/ha at a price of N30/bag. These results support the expectation that major price increases, resulting from the removal of subsidies, will undo many of the achievement-i - !.-.a 1.L)s, eliminating resource-poor farmers from participating, and slowing or reversing the overall growth of output. In -32 - 1985, Hill (1985b) found that in south-western Kano State the vast majority of fertilized plots and the greater part of the fertilized area received less than 150 kg/ha (the recommended level) of IF in 1985 - before prices rose. An improvement in use efficiency of IF can only partly compensate for such a trend. - 33 - 5. ADAPTATION 5.1 Farmer response to soil degradation. Once bush Fallowing systems have given way to permanent, annual or bi-annual cultivation systems, three options are available to the farmer who becomes aware of the degradation of his land: (1) to abandon it, (2) to continue using it as before, and (3) to invest in rehabilitating or improving it. The choice of option is of course determined by the economic resources of the farmer. 5.1.1. Abandonin Land. Land abandonment occurs, but to an unknown extent. in the Kano Close-Settled Zone there are pieces of land known to have been once under annual cultivation and now virtually worthless. In a system of annual cultivation where manuring is the condition of substainability and alternative land has, by definition, been eliminated, it may be concluded that abandonment is evidence of economic failure, extreme poverty. It carries the penalty of loss of access to productive resources. The costs of rehabilitation of exhausted land (which may be guessed from Dennison, 1956; KNARDA, 1989) are such that the effort is unlikely to be made except by Government experimental institutions. Much more needs to be known about the rate and causes of land abandonment if, as suggested here, it is the terminal state of the process of degradation. 5.1.2. Continuing as before. The second option, to continue using the land as before, is really identical to option (1) since only poverty (or foolhardiness) can compel a resource manager to deny the evidence of his own eyes. Surveys of IF and FYM use (e.g. Hill, 1985a, b) always establish the existence of a residual proportion of plots that are not fertilized. Some of them may be awaiting their turn in a cycle of infrequent manuring. Hill (1985b) suggests (without citing evidence) that they are the more fertile plots, needing FYM least. But FYM, for many farmers, is a purchased input. It seems possible that a number of unfertilized plots are degrading by reason of the economic incapacity of their owners. , 5.1.3. Land Improvement. The third option, investment in land improvement, raises the questions of (1) economic inequality (demonstrated in the enormous differences in rates of manure or fertilizer application among plot- holders) , and (2) the transfer of resources into agriculture from other sectors, which is the motor of capitalistic agriculture, whether on the urban periphery or further afield. Farmer surveys which have traditionally confined themselves to questions of farming technology and agricultural economics - such as the FRADYS and Micro-Agronomic Surveys - cannot illuminate this *034- question adequately since it is multisectoral. The capitalisation of agriculture is, of course, a bigger question than that of land improvement alone. But more needs to be known about the manipulation of land as capital, the operation of land improvement as an economic investment, and related questions (Labaran, 1987; Mortimore et al., 1987). With regard to the technologies of land improvement, there is good reason to believe that, like manuring, the advantages of such practices as the incorporation of crop residues, mulching, run-off control by appropriate ridging systems and avoidance of steep slopes, and windbreaks, are recognised, but they may not always be implemented owing to economic constraints (such as the alternative use value of crop residues, or labour shortages). Surveys have suggested a high level of awareness of soil erosion (e.g. Bello, 1983; Gurumpanwa, 1983; El-Buba, 1985). There is certainly scope for conservation extension, but attention needs to be paid to the rationale for indigenous practices. For example, Liman (1988) found that the most important divergence between indigenous and scientific assessments of land capability rankings in Ganye District (Gongola State) is the farmers' preference for cultivating erosion-prone hillslopes, which are believed to give higher yields relative to the effort needed to clear them, longer yield cycles, and need less weeding. Farmers in the Kano Close-Settled Zone, who maintain many trees on their farms, were subjected for years to forestry extension designed to further the planting of exotic trees and communal woodlots, but their own system of woodstock management and values provides a sounder base for protecting and improving the woodland (Cline-Cole et al., in press). 5.1.4. Degradation and poverty. Alternatively, the explanation for degrading management practices may be sought in socio-economic inequality, individual appropriation of wealth and the diminishing power of the poor to conserve their own productive resources. By this argument, the intensified inequality that has followed the introduction of ADPs (e.g. Funtua: Mabogunje and Gana, 1981) may possibly contribute to selective land degradation in the longer term. 5.2 Farmer Response to Drought. There has been a number of studies of adaptation among farming and pastoral communities in northern Nigeria to agricultural drought, especially the Drought of 1972-74 (Abdu, 1976; Ahmed, 1976; Aliyu, 1976; Dauda, 1976; Kura, 1976; Morm -,, 1976; Ndaks, 1976-all summarised in Van Apeldoorn, 1981; Abdu 1984' Watts, 1983; Mohammed, 1985; Ibrahim, 1986). Mortimore (1989 , drawing on data collected throughout the -35- period 1973-86, has systematised a general descriptive model which distinguishes between adaptation to drought, adaptation to poverty, and adaptation to hunger. The first consists in agro-pastoral management techniques (such as varying plant spacings, crop varieties, grazing circuits, herd composition) in order to use reduced rainfall more efficiently. The second consists in multi-sectoral economic strategies, sometimes involving short or long-term spatial mobility, and intended to supplement incomes as families become more dependent on the market for food. The third consists in mobilising alternative foods from the ecosystem, substituting amongst market foods, or going without. The intensity of use of these options is governed by the level of food shortages from year to year and from household to household. It is common for such adaptive behaviour to be attributed to environmental degradation. However, this view is incorrect. When the rain returns, there is a return to agriculture and a reunion of migrants. The observed behaviour is a response to an economic crisis accentuated by the imperfections of the market - food price inflation and the depression of livestock and property prices being the normal sequels to harvest failures. Farmer response to drought is only relevant to land degradation in the sense that individual poverty may be induced by the loss of land productivity, and a major harvest failure brings this situation to a crisis. But, as suggested above, the loss of land productivity (or degradation) may rather be induced by individual poverty. It is preferable, therefore, to remove the question of farmer response to drought from the analysis of land degradation. 5.3 Fadama Farming. This provies an example of the complex interactions involved in the question of land degradation and the farmers' response to it. Small-scale irrigation, and the production of perennial crops in river valleys and depressions with a high water table, has always been primarily a response to market opportunity, since the crops produced - most of them in the dry season - are usually sold. But under conditions of low rainfall and periodic drought, fadama farming (given market opportunity) acquires added significance as an alternative to rainfed farming. Water supply is less limiting and soil fertility is above the average for the locality. Labour demand does not conflict with rainfed farming. For the same reasons, fadamas are intensively developed in densely populated areas, where land supply and productivity are insufficient to support the subsistence needs of the resident population in all years. Such was so in the Sokoto Close-Settled Zone in the 1960s, for example (Goddard et al, 1975), where fadama farming provided an alternative to dry season mobility as a source of supplementary income. - 36 - In the 1980s, the demand for dry season vegetables, rice, and recently, wheat has expanded rapidly, with dramatic increases in some prices. At the same time, small-scale irrigation has received official approval as a supplementary developmental strategy to rainfed farming in drought-prone areas of West African countries. Support was provided for fadama farmers by some state ADPS in the form of subsidised pumps, washbores, inputs and technical advice (BSADPq 1985). It is as unreasonable to attribute the expansion of fadama farming in recent years solely to degradation in the rainfed farming sector as it is to credit it entirely to the ADPs. In the 1988-89 dry season, the fadama farming sector became a major target of the Federal Government's policies to boost wheat output, following the cessation of imports. The fadama farming sector is characteristically dynamic. Between 1974 and 1984 at Ringim, in the Hadejia River Valley, there was a shift out of sugar, tobacco and wheat into perishable vegetables (see Erhabor, 1982), and more recently back into wheat. There was a swing from FYM to IF and, of course, from shadiifs to pumps (Ubale, 1985). But it has been reported that the high economic returns from fadama farming in Bauchi encouraged farmers to neglect soil conservation practices (BSADP, 1986). The rush into fadama farming may be accentuated by the removal of the fertilizer subsidy, since these enterprises appear likely to remain profitable, notwithstanding the increased cost of pumps. -37 - 6. THE POPULATION FACTOR 6.1 Population Growth. The estimated population of Nigeria in 1984 was 93.7 million and the annual rate of growth is believed to lie between 3.0 and 3.5 per cent (WFS, 1984). About half the population is under 15 years of age and current fertility is high (Total Fertility Rate, 6.34). Infant mortality is high, though declining (84.4 per thousand live births) and child mortality also (144.5 per thousand live births). Whether these rates are representative of the northernmost states is not known. Bradley et al. (1982a,b) calcuIat !1:1s of natural increase of 2.9 per cent and of infant mortality of 170 per thousand in a population of over 40,000 at Malumfashi, in southern Katsina State, during the early 1970s. From what is known of mortality, health service delivery and levels of living in rural areas north of 11 0 N, it seems probable that mortality is higher and natural increase lower in the study area than the national average. 6.2 Intensity of Land Use Because of the complex relationships between population and land (referred to in Section 2.3 above), as well as the absence of reliable and compatible estimates of population at district or LGA level, a map of farming intensity is more appropriate as the starting point for an analysis of the population factor (Map 3). This map shows the approximate distribution of three classes of intensity: (1) over 60% (intensive farming systems); (2) 30-60% (less intensive systems); and (3) less than 30% (extensive systems). These correspond roughly to (1) permanent, annual or bi-annual cultivation systems; (2) shrub or short bush fallowing systems, and (3) long bush fallowing systems and uncultivated areas. 6.2.1 Intensive Systems The most important characteristic of these systems is their great spatial extent within the study area. Their distribution is markedly influenced by: (a) major historic political centres and close-settled zones (Sokoto; Kano-Katsina-Zaria; Azare and Potiskum). The configuration of the Kano Close-Settled Zone appears to have been partly controlled by the distribution of semi-arid brown soils (Mortimore, 1968) and its extensions westwards and eastwards occur on soils with similar properties; (b) major river-systems and associated hydromorphic soils (the Sokoto-Rima system, the Hadejia- Komadugu Yobe-Jama'are-Komadugu Gana system, and localised systems in Borno). The spatial continuity of these systems is interrupted by forest reserves, notably those between Katsina and Gusau. The intensive systems, generally having recourse only to short grass fallows (or none), are dependent on manuring or fertilization to sustain minimal farm yields. - 38 - 6.2.2. Less intensive systems. Such systems occur widely in areas of recent population growth and in-migration, such as the Gumel-Nguru area, and areas west of Gusau and of Argungu. Long fallows are increasingly difficult to sustain when farming intensity generally exceeds 30%, and upland soils are becoming dependent on some degree of manuring or fertilization in order to sustain minimal farm yields. 6.2.3 Extensive systems. Many of the areas shown correspond to soils of exceptionally low fertility, notably the weakly developed dune sands of northern Borno and Bauchi States, and the sandstone uplands of Sokoto. The map does not distinguish extensively farmed from uncultivated areas or forest reserves, but the majority of areas shown are farmed on long fallowing systems. Within such areas, however, there are localised intensive or less intensive systems in operation in river valleys and in the vicinities of towns or villages. Manuring or fertilization are, therefore, locally important. Because of the poverty of some of the soils in question, it cannot be assumed that fallowing generally sustains minimal farm productivity. 6.3 Alternative Pathways. 6.3.1 Theoretical alternatives. The debate about the relations between population growth and agricultural intensification continues to be dominated by the opposed theoretical positions given early expression by Boserup (1965) and Allan (1965). Essentially, an extensive land use system supporting a growing population and running short of free land is faced with a choice between two options: (1) an intensification and (2) a degradation pathway. The first is achieved by means of increased labour inputs, and the extension of land-conserving technologies to an increasing proportion of the land; the second by holding to a fixed technology and constant labour inputs. The question arising from this theoretical debate has much practical significance for the land use systems of semi-arid Nigeria. Will the present systems break down under the rates of population growth presently experienced in the area, or will they accomplish a transition to a higher level of intensity? If the first, land may be expected to degrade further; if the second, ecological sustainability may be achievable. 6.3.2. Intensive Systems. Few studies have been conducted with the objective of zesolving this issue, and those that have tend to rely on indirect evidence rather than close monitoring of relevant variables over time. A case has been made for the Kano Close-Settled Zone, and other areas like it, as exemplars of the intensification pathway (see * 39 - Mortimore, 1989 for references). In a recent review of the evolution of this system during the last two decades (Mortimore, in press), it is argued the Kano triple system of crop, livestock and tree husbandry has survived intact the pressures of urbanization and the oil boom; that the functional relationships between these three elements are stable ecologically and economically; and that rather than damage the sustainability of the system, the increasingly dense population has chosen to diversify out of primary production. The importance of the Kano Close-Settled Zone rests in its validity as a guide to the future course of development of land use systems elsewhere. The critical importance of access to land under conditions of dense and growing population is expressed in a tenure system that allows small family holdings to confer rights on increasing numbers of people (Morris, 1981). Tenurial relations become increasingly complex. The involution of land ownership has increased the participation of the population in the competition for land (Hill, 1977; Ross, 1987). Such conditions must favour land conserving and soil improving technologies. Population density in the Kano Close-Settled Zone increases with proximity to Kano. Hill (1986) found that mean plot sizes in five villages in SW Kano diminished towards Kano (from 0.41 ha to 0.14 ha) and mean application rates of FYM increased (from 3,950 kg/ha to 18,900 kg/ha). Preliminary agronomic survey results for 1985 gave a correlation coefficient of rm-0.81 between plot size and the rate of application of organic fertilizer (FYM). 6.3.3 Less Intensive Systems. Lower intensity systems based on bush tallowing or shifting cultivation are increasingly subject to land shortages. Fallow periods are cut down and an increasing proportion of land is cultivated annually. Woodland status provides an indicator of change. Woodland fallow is normally characterised by short, young and dense woodland that is frequently cut; and annual cultivation, by mature economic trees at lower densities (farmed parkland). Analysis of sequential air photographs provides evidence that such a transition is widespread. This is born out by field observations (Pullan, 1974; Cline-Cole et al., in press). If the low intensity bush fallowing systems were to break down there should be evidence of this in abandoned land (labour inputs would be uneconomic and would be withdrawn), depopulated settlements, and permanent out- migration - and in physical evidence of land degradation. No such analysis has been reported however. Notwithstanding numerous major droughts during the last two decades, there is no evidence that large numbers of villages have ceased to be viable productive units. Soil - 40 - loss from arable land needs to be reduced, tree planting needs to be increased, and field boundaries stabilised in many areas. But the evidence suggests that an intensifi- cation rather than a degradation pathway is being followed. 6.3.4 Policy. The nature of the problem, if this analysis of the population factor is correct, is not that remedial action is urgently needed to prevent a degradational disaster, but rather that policies supportive of the adoption of an intensification pathway should be pursued. While the intensification objectives of the ADPs are consistent with such a goal, and IF has a part to play, it should be noted that the increasing power of the foodstuffs market introduces a new set of stresses that may soon become more important than those created by the growing population. -41- 7. CONCLUSION 7.1 Summary. The main points arising from Sections 3-6 are summarised below (using the same paragraph numbers). 3. EVIDENCES OF DEGRADATION. 3.1 Rainfall. The facts of reduced rainfall, altered monthly distribution, its impact on crop production and hydrological conditions in the last two decades are inescapable. The impact on the degradation of land is very difficult to assess in distinction from management factors. 3.2 Desertification. The 'desertification consensus' is widely accepted, but rests on an inadequate empirical base and until a better empirical foundation can be provided, should be accepted with caution. Holders of this view admit that rainfall and administration as well as management are involved. 3.3 Vegetational degradation. Ecological arguments against burning, overgrazing, woodcutting, and agricultural clearance are quite often misinformed or overstated and the evidence that links these practices with degradation is uncritically used. Nevertheless, vegetation contributes to soil productivity, and must retain a central place in degradation analyses. 3.4 Soil Erosion. Soil erosion by water (gullies and soil washJand 1y ind (dune mobilisation and aeolian action) is widespread, but has not been subjected to measurement in enough places in the study area to provide a basis for regional assessments of soil loss. 3.5 Soil degradation. The data will not allow any conclusion on soil degradation at the regional level, though the effects of cultivation are known under experimental and farm conditions. The degradation of the physical properties of the soil under mechanical cultivation or irrigation is the most urgent aspect of the problem. 4. SOIL PRODUCTIVITY AND REPLENISHMENT 4.1 Yield trends. There appears to be a contradiction between the 'desertification consensus' view of low and declining yields and some ADPs' reports of improving and higher yields. A systemati inv2stigation of available data on yield trends is needed. *42 - 4.2 Nutrient removal in crops. Considerable work has been done on the nutrient requirements of the major crops as a basis for fertilizer recommendations. But a 'nutrient audit' for the study area needs to take into account the amounts removed from the farm and should be extended to all nutrients known to be limiting. Economic data on crop output is still hardly adequate to support such an audit. 4.3 Nutrient replenishment from the atmosphere. The atmospFere is a significant source ot S, P, K, and N. The Harmattan may be particularly useful as a supplier of P. Symbiotic N-fixation, using leguminous crops, will continue to increase in importance as the contribution of non-symbiotic fixation decreases with the gradual disappearance of fallows. Legumes have potential in crop mixtures and rotations, in forage-cereal mixtures on mixed farms, in fodder banks, for land rehabilitation, rangeland improvement and in farm tree systems. Legume management will become more critical as more demands are made on the inherently N-poor soils of the study area. 4.4 Fallowing. Fallows are becoming shorter as land becomes scarce, and short fallows are little superior in their soil properties to cultivated (and fertilized) land. Breaking fallows needs extra labour. Under these conditions there is little to lose in transferring from fallow systems to systems of annual or permanent cultivation. 4.5 ManurinR. FYM's proven value, under both experimental and farm conditions, is universally appreciated, even at low rates of application. It is generally superior to IF, but its principal advantage is its beneficial effects on soil physical properties, which are adversely affected by cultivation. The main constraint on the use of FYM is its supply, which is linked to forage production and land scarcity. 4.6 Fertilizers. IF has not taken over the position of FYM but is a valuable supplementary resource; not all farmers use it and average rates of use are low by world standards. Inefficiencies arise from misuse, from suboptimal nutrient ratios, and from supply bottlenecks affecting preferred types. There is real risk of accentuating nutrient deficiencies in the soil by unjudicious ad hoc application, especially of N fertilizers which raise soil acidity. IF does nothing to restore soil physical properties under cultivation. The profitability of IF is diminishing, and will deteriorate further if the subsidy is removed, causing reduced rates of application, reduced fertilized areas, and reduced output of less profitable crop enterprises. -43- 5. ADAPTATION 5.1 Farmer response to degradation. Conservation makes economic sense, and therefore management that causes degradation must be explained either in terms of conflicting rationale or economic incapacity. On the first, indigenous management rationales need to be better understood; and on the second, poverty as a cause (rather than a consequence) of degradation needs to be investigated and corrected. 5.2 Farmer response to drought. Response to drought is response to a short-term food and poverty crisis, and is not a direct indicator of long-term degradation. 5.3 Fadama farming. Recent expansion has resulted from market growth, poor returns from rainfed farming, the need for an alternative to dry season mobility as a source of income, and ADP or Government support ?rogrammes. It illustrates the difficulty of isolating farmer response to degradation'. 6. THE POPULATION FACTOR 6.1 Population growth. The rate of growth in the study area is probably less than the national average (recently estimated to be 3.35%). 6.2 Intensity of Land Use. Map 3 shows the approximate distributions of intensive, less intensive and extensive farming systems. The spatial extent of these systems should be considered in conjunction with the degree of their dependence on manuring or fertilization. 6.3 Alternative Pathways. The question arising from theoretical considerations is whether the land use systems will follow an intensification pathway - like the Kano Close-Settled Zone - or a degradation pathway. If, as most studies suggest, land use systems are moving along an intensification pathway, policy should be designed to support intensification. This appears consistent with the main objectives of the ADPs. 7.2 The role of inorganic fertilizers 7.2.1 Profitability of IF. Under present conditions of food price inflation, and notwithstanding the rising cost of other inputs (such as pumps), it remains to be demonstrated that IF can become unprofitable for all enterprises. A more realistic scenario is that high value crops (such as wheat and rice), and highly productive systems (such as small-scale irrigation) will continue - 44 - indefinitely to use IF. On the less profitable enterprises there will be increased demand for FYM and probably a reduction in output. 7.2.2 Suitability of IF. Technically, IF can be used to correct any nutrient deticiency, but in practice the match between nutrient requirements and nutrient application is a clumsy one, carrying the risk of accentuated nutrient imbalance in the soil. IF may not improve soil physical properties significantly under cultivation (especially mechanical cultivation). Therefore IF can not be a complete answer to the needs of the soil under agricultural intensification. 7.2.3 Farmyard manure. It seems that experimental and on- farm experience agrees on the superiority of FYM, especially with respect to its beneficial effect on the physical properties of soils under cultivation. In the context of removing the fertilizer subsidy, FYM is the best alternative source of P and K, as well as some other nutrients. But under present conditions, the supply of FYM is incapable of meeting the requirements of annual cultivation carried on over increasingly large areas, even at moderate yield levels. The introduction of more legumes into crop mixtures could increase forage output, and thereby support more livestock, but forage is traditionally regarded as a byproduct of the food crops. Research is needed into the possibility of raising the productivity of linked crop-livestock-tree husbandry systems such as that of the Kano Close-Settled Zone. 7.2.4 Legumes. N-fixation offers the possibility of an external input to the nutrient system, either to counteract nutrient losses through crop removals, or to raise the system's total productivity. Legumes already have an important place in the cropping systems. To increase N-fixation to support intensification should therefore be a major area for research. In the context of removing the fertilizer subsidy, N-fixation appears the most practicable alternative source of N. 7.2.5 Other atmospheric additions. Although more research is needed, enough is known about air particulates to support the view that dust contains several elements of agricultural significance and that, even though the quantities deposited each year are small, the residual and long-term effects of such deposition must be significant. 7.2.6 Soil Degradation. Soil erosion by water and wind, although not quantified at the regional level, are evidently important in the study area. Soil degradation under cultivation, although demonstrated experimentally and inferred from comparisons of soil properties under - 45 - differing management regimes on farms, also has not been quantified. Its importance for the future of agriculture in the area remains an article of faith rather than a measured fact. The reason for this is the absence of time- series data on soil properties at control sites. This is a research priority. A beginning can be made with fresh analyses of the soils at sites sampled in north-central Nigeria by the LRC before 1976. Both analytical data and sample material from the original work is available in Kano, and the time depth may now be sufficient to test the possibility of significant change. Older soil data may be recoverable for other sites in the area. Only when exact assessments of soil degradation have been made, can correct specifications of nutrient needs (for soil maintenance, rather than crop yields) be made, whether IF or another source is intended. Acknowledgements The permission of the Director of the Institute for Agricultural Research, Samaru to use the Institute's Library, the advice and support of Mr. Graham Quinn (FACU, Kaduna), and the patience of all who gave information are gratefully acknowledged. -46- REFERENCES ABDU, P.S. 1976 Drought-caused migration around Illela. B.A. Dissertation, Department of Geography, Amadu Bello University, Zaria. ABDU, H.M.S. 1987 The influence of Acacia albida on some soil properties in a Nigerian savanna: M.Sc Thesis (Arid Zone Ecology), University of Maiduguri. ABDULLAHI, A.; LOMBIN, G. 1978 'Long term fertility studies at Samaru - Nigeria: III Comparative effectiveness of separate and combined applications of mineral fertilizers and farmyard manure in maintaining soil productivity under continuous cultivation in the savanna'. Samaru Miscellaneous Paper 75, Institute for Agricultural Research, Ahmadu Bello University, ABENEMI, J.F. 1986 Estimating soil erosion rate and productive lifespan: experiences from the Hadejia River Basin. B.Sc bissertation, Department of Geography, Ahmadu Bello University. ADEBAYO, S.I.; MOHAMMED, I. 1987 'Rainfall, crop failure and drought. An appraisal for 1983-84 experience in relation to normal in Kano State'. In: V.O.Sagua et al. (eds.), Ecological disasters in Nigeria: drought and desertifi- cation, Federal Ministry of Science and Technology, Lagos: ADESIYAN, J.0. 1985 Regional development alternatives: a critical assessment of Gombe Integrated Development Project. Ph.D Thesis, Ahmadu Bello University. ADU, J.K.; NNADI, L.A. 1987 'Biological nitrogen fixation in in the tropics: a review with particular reference to savanna zones of Nigeria'. Samaru Miscellaneous Paper 120, Institute for Agricultural Research. AHMED, M.L. 1976 Socio-cultural organisation and the sharing of impact of a natural hazard. B.A. Dissertation, Department of Geography, Ahmadu Bello University, Zaria. AHN, P.M. 1977 'Soil factors affecting rainfed agriculture in semi-arid regions with particular reference to the Sahel Zone of Africa',Technical Communication 8, University of Nairobi, Department of Soil Science. ALIYU, M. 1976 Modifications of agricultural techniques to the drought situation in llela village. B.A. Dissertation, Department of Geography, Ahmadu Bello University, Zaria. - 47 - ALLAN, W. 1965 The African husbandman. Edinburgh: Oliver and Boyd (reprinted 1977 at Westport: Greenwood Press). ALLISON, L.E. 1962 'Salinity status of irrigation schemes in Northern Nigeria including an appraisal of artesian borehole waters for irrigation'. Consultant Report 26, USAID, Nigeria. BAITA, R. 1986 Some aspects of the ecology and distribution of Kapok tree in Minjibir, Kano State. B.sc. Dissertation, Department of Geography, Bayero University, Kano. BAKER, F.F.I.; LOMBIN, G.; ABDULLAHI, A. 1977 'Long term ferti- lity studies at Samaru: I. Direct and residual effects of single superphosphate and farmyard manure on yield of cotton, sorghum and groundnuts grown in a rotation'. Samaru Miscellaneous Paper 67, Institute for Agricultural Research, Ahmaau Bello University. BALASUBRAMANIAN,V.; NNADI, L.A.; LOMBIN, L.G.P; YAYOCK, J.Y. 1978 'Fertilizer use in Nigeria. I. Trends in consumption'. Annual Conference, Nigerian Society of Chemical Engineers. BALCET, J.-C.; CANDLER, W. Farm technology adoption in northern Nigeria. Vol 1, Main Report. The World Bank, Washington DC. BEAVINGTON, F.; CAWSE, P.A. 1979 'The deposition of trace elements and ma4 or nutrients in dust and rainwater in northern Nigeria . The Science of the Total Environments 13: 263-74. BELLO, A.S. 1983 Soil degradation and soil conservation. A case study at Zogirma area Sokoto State. B.Sc Dissertation,Ahmadu Bello University, Zaria. BELLO, S.N. 1985 Relationships between soil colour and soil fertility in Kano and part of Raduna plains. B,Sc Dissertation, Department of Geography, Ahmadu Bello University, Zaria. BLONDEL, D. 1971 'Development of mineral nitrogen in tropical ferruginous soil under millet crop (Pennisetum typhoides)'.STRC/OAU Seminar, Dakar. BOSERUP, E. 1965 The conditions of agricultural growth. London: Allen and Unwin. - 48 - BRADLEY, A.K.; MACFARLANE, S.B.J.: MOODY, J.B.: GILLES, R.M.: BLACKER, J.G.C.; MUSA, B.D. 1982(a) 'Malumfashi Endemic Diseases Research Project XIX. Demographic findings: population structure and fertility'. Annals of Tropical Medicine and Parasitology,76:381-91. BRADLEY, A.K. et al., 1982(b) 'Malumfashi Endemic Diseases Research Project. XX. Demographic findings: mortality'. Annals of Tropical Medicine and Parasitology. 76:393-404. BROMFIELD, A.R. 1974a 'The deposition of sulphur in dust in northern Nigeria'. Journal of Agricultural Science, Cambridge. 83:423-5. BROMFIELD, A.R. 1974b 'The deposition of sulphur in the rain- water in northern Nigeria'. Tellus, 26:408-11. BROMFIELD, A.R. 1975 'The deposition of sulphur on soil in Nigeria'. Journal of Agricultural Science, Cambridge, 83: 567-8. BSDAP, 1984 Input Survey report, 1984. Azare: Northern Zone Evaluation Unit, Bauchi State Agricultural Development Project. BSDAP, 1985 Agronomic survey report, 1984. Azare: Northern Zone Evaluation Unit, Bauchi State Agricultural Development Programme. BSDAP, 1986 Fadama survey (1985/86) report. Planning, Monitoring and Evaluation Unit, Bauchi State Agricultural Development Programme. BSDAP 1987 Report on the 1987 large scale reconnaissance survey for Bauchi State ADP (Northern Zone2. Azare: Plannifng, Mon1toring and EvaLuation Unit, Bauchi State Agricultural Development Programme. BSDAP n.d. Fertilizers and benefits in Bauchi State. Bauchi; State Agricultural Development Programme. CCDE 1988 Report on desertification and desert encroachment in eleven northern states of Nigeria. Kano: Consultative Committee on Desert Encroachment. CHARREAU, C. and FAUCK, F. 1970 'Mise au point sur l'utilisition a7ricole des sols de la region de Sefa (Casamance)' Agronomie Tropicale, 25:151-91. CLINE-COLE, R.A.; FALOLA, J.A,; MAIN, H.A.C.; MORTIMORE, M.J.; NICHOL, J.E.; O'REILLY, F.D., in press. Wood fuel in Kanos Final Report of the Kano Rural Energy Research Project (United Nations University. - 49 - DAUDA, I. 1976 Modifications of aaricultural techniques to the drought situation in Mungurun village, B.A. Dissertation, Department of Geography, Ahmadu Bello University, Zaria. DANIEL, M.B. 1987 'The significance of post-implementation monitoring in land development projects'. In: Mortimore, M. et al. (eds.) Perspectives on land administration and development in northern Nigeria, Department of Geography, Bayero University, Kano: 198-209. DANIEL, M.B.; ESSIET, E.U. 1989 'Soil loss and effects on fertility in irrigated agriculture, Kano State*. Workshop on Land Administration and Development in Northern Nigeria, Bayero University, Kano. DENNISON, E.B. 1956 'Problems of reclaiming certain infertile soils in the Sudan zone of Northern Nigeria'. World Crops, 8:131-3. DIRISU, E.N. 1983 Land use as a factor affecting soil properties of the top-sol in Kano and Kaduna plains. B.Sc Dissertation, Department of Geography, Ahmadu Bello University, Zaria. EL-BUBA, S. 1985 Incidence of desertification in Geidam and its environs. B.Sc Dissertation, Department of Geography, University of Maiduguri. ERHABOR, P.O. 1982 'Efficiency of resource use under small- scale irrigation technology in northern Nigeria'. Technical Report 148, Purdue University Water Resources Research Center: West Lafayette, Indiana. ESSIET, E.U. 1989 'The impact of irrigation on soil properties in Kano, Nigeria'. Workshop on Land Adcinistration and Development in Northern Nigeria, Bayero University, Kano. ESSIET, E.U. in press 'The contribution of manure to soil fertility in Kano.' FADA, B.; RAYAR, A.J. 1988 'Agricultural management strategies in an arid environment' Journal of Arid Agriculture (University of Maiduguri), 1/1:1-21. FAO 1969 Soil and water resources survey of Sokoto Valley. IlReport (5 vols.). Rome: Food and Agriculture drganisation of the United Nations. FAO 1971 Report on the Third Conference on soil fertility and fertilizer use in Africa A3dis Ababa, 1970. Rome: Food and Agriculture organisation of the United Nations. - 50 - FAO 1974 Report of an expert consultation on soil degradation held in Romeq 10-14 June 1974. Rome: Food and Agriculture Organisation of the United Nations. FARMER, G.; WIGLEY, T.M.L. 1985 Climatic trends for tropical Africa. A research report for the Overseas Development Administration. Climatic Research Unit, University of East Anglia. FDALR 1980 Evaluation of soil fertility status of northern Nigeria soils. Preliminary report. Federal Department of Agricultural Land Resources. FDALR 1985(a) The reconnaissance soil survey of Sokoto State, Nigeria. SoiIs Report. Federal Department of Agricultural Land Resources. FDALR 1985(b) The reconnaissance soil survey of Borno State, Nigeria. Soils Report (2 vols.). Federal Department of Agricultural Land Resources. FDALR 1985(c) The reconnaissance soil survey of Kano State, Nigeria. Soils Report. Federal Department of Agricultural Land Resources. FDALR 1985(d) The reconnaissance soil survey of Bauchi State, Nigeria. Soils Report. Federal Department of Agricultural Land Resources. FDALR 1986/87 Report on land use and land degradation in Borno State. Federal Department of Agricultural Land Resources. FDF 1977 Nigeria, Vegetation and Land Use, scale 1:250,000. Federal Department of Forestry, Ibadan. GODDARD, A.D.; MORTIMORE, M.J.; NORMAN, D.W. 1975. 'Some social and economic implications of population growth in rural Hausaland', in: J.C. Caldwell et al. (eds.), Population growth and socio-economic change in West Africa, Columbia University Press, New York: 321-36. GURUMPANWA, H.B. 1983 The problems and awareness of soil degradation and conservation in Uurumpanwa areaoF Gongola State. B.A. Dissertation, Department of Geography, Ahmadu BeTI University, Zaria. HARKNESS, C.; YAYOCK, J.Y. 1979 'Changes in groundnut growing'. .eeti, on ce; abilitation of groundnut growing, Samaru, 1 7 9. HARE, F.K. 1985 Climate variations, drought and desertifi- cation: WMO Report 653. Geneva : World Meteorological Organisation. - 51 - HEATHCOTE, R.G. 1971 'Soil fertility under continuous culti- vation in Northern Nigeria. I. The role of organic manure.' Experimental Agriculture 6:229-37. HENDY, C.R.C. 1977 'Animal production in Kano State and the requirements for further study in the Kano Close-Settled Zone'. Land Resource Report 21, Tolworth: Land Resources Division. HILL, J. 1935(a) Working notes on fertilizer use and crop response to inorganic and organic fertilizer in south- western Kano State. Kano State Agricultural and Rural Development Authority. HILL, J. 1985(b) Plot yield response estimates to inorganic fertilizer for selected crop mixtures in south western Kano State. Kano State Agricultural and Rural Development Authority. HILL, J. 1986 Kano ADP fertilizer study results (Draft). Kano State Agricultural and Rural Development Authority. HILL, P. 1977 Population, prosperity and poverty. Rural Kano 1900 and 1970. Cambridge University Press. IAR 1977 Cropping Scheme Meetingg 1978: Notes on the soil fertility, crop nutrition and environmental programme. Institute For Agricultural Research, Ahmadu Bello University. IAR 1978 Cropping Scheme Meeting, 1978. Notes on the soil fert1lity, crop nutrition and environmental programme. Institute For Agricultural Research. IAR 1981 Cropping Scheme Meeting 1981. Soil fertility, plant nutrition and environmental studies, and soil survey programme. Institute for Agricultural Research. IAR 1982 Cropping Scheme Meeting 1982. Notes on the soil and crop environment programme. Institute for Agricultural Research. IAR 1984 Cropping Scheme Meeting 1984. Soil and crop environ- ment programme. Institute for Agricultural Research. IAR 1987 Annual Report of the Institute for Agricultural Researcht-AhSmadu Bello University, 1983184. IBRAHIM, M.A. 1986 Trends in the livestock economy in Birnin Kebbi Local Government Area. B,Sc. Dissertation, Department of Geography, Ahmadu Bello University, Zaria. - 52 - ILANI, Y.S. 1981 Rates of soil wash in a badland area under savanna climate Zaria. B.Sc Dissertation, Department of Geography, AhmadU Bello University, Zaria. JACKSON, J.K.; TAYLOR G.F.; CONDE-WANE, C.1983 Management of natural forest of the Sahel region. Club du Sahel Paper D(83)232, Paris: Organisation for Economic Cooperation and Development. JAYARAMAN, P. 1988 National fertilizer and agro-chemicals con- sumption study. Executive summary and analysis. Kaduna: Agricultural Projects Monitoring Evaluation and Planning Unit. JONES, M.J. 1971 'The maintenance of soil organic matter under continuous cultivation at Samaru, Nigeria'. Journal of Agricultural Science, Cambridge 77:473-82. JONES, M.J. 1974 'Changes in soil properties resulting from fertilizer use and their effect on crop yields'. Samaru Conference Paper 2, Institute for Agricultural Research. JONES, M.J.; WILD, A. 1975 Soils of the West African savanna. The maintenance and improvement of their Fertility. Harpenden: Commonwealth Agricultural Bureau. KNARDA 1983 Agricultural adaptive research in 1982. Working Paper. Kano state Agricultural and Rural Development Authority. KNARDA 1984 Agricultural adaptive research in 1983. Draft research report for IAR Cropping Scheme Meetingst February 1984. Kano State Agricultural and Rural Development Auti 3rity. KNARDA 1985 Rainfed agricultural adaptive research in 1984 Part 1. IAR Dratt Croeping Scheme Report, February 1985. Kano State Agricultural and Rural Development Authority. KNARDA 1986 Rainfed agricultural adaptive research in 1985. Part 1. March 1986. Kano state Agricultural and Rural Development Authority. KNARDA 1987 Report on the effect of drought and grasshopper infestation on field crops in Kano State. Kano State Agricultural and Rural Development Authority. KNARDA 1988(a) 1987 rainfall records and isohyets and crop performance. Land Use Planning and Agricultural Departments. Kano State Agricultural and Rural Development Authority. - 53 - KNARDA 1988(b) Survey on fertilizer pricing policy in Nigeria. Monitoring and Evaluation Department, Rano State Agricultural and Rural Development Authority. KNARDA 1989 TechnoloRy adoetion and 1988 adaptive research pro ramme Cropp ng Scheme Meetin Februar 1 Kano State Agricultural and Rural Development Authority. KODIYA, H.M. 1988 Effects of irrigation on some soil charac- teristics in the South Chad Project area Maiduguri, ieria. M.Sc Thesis (Land Resources), Bayero University, Kano KOLAWOLE, A. 1987 'Environmental change and the South Chad Irrigation Project (Nigeria)'. Journal of Arid Environ- ments, 13:169-76. KOWAL, J.M. 1970 'The hydrology of a small catchment basin at Samaru, Nigeria. IV. Assessment of soil erosion under varied land management and vegetation cover'. Nigerian Agricultural Journal, 7: 134-47. KOWAL, J.M.; KNABE, D.T. 1972. An agroclimatological atlas of the northern states of Nigeria. Zaria: Ahmadu Bello University Press. KURA, A.G. 1976 Social and economic adaptations to the drought in Bedde Division, B.Sc Dissertation, Department of Sociology, Ahmadu Bello University. LABARAN, A. 1987 'Land appropriation for capitalised farming in the Sokoto region: some preliminary findings'. In: Michael Mortimore et al. (eds.), Perspectives on land administration and development in northern Nigeria, Department of Geography, Bayero University: 46-57. LIMAN, A.T. 1988 The relationship between land capability and land use in Sugu District. Ganye Local Government Area of Gongola State. N.Sc. Thesis (Geography), University of Maiduguri. LOMBIN, L.G. 1979 'Evaluation of the magnesium supplying powers of the Nigerian savannah soils'. Soil Science and Plant Nutrition, 25:477-92. LOMBIN, L.G. 1981(a) 'Continuous cultivation and soil produc- tivity in the semi-arid savannah: the influence of crop rotation'. Agronomy Journal 73:357-63. - 54 - LOMBIN, L.G. 1981(b) 'The effects of continuous fertilization on nutrient balance and crop yield in the northern Nigerian savannah: a preliminary assessment'. Canadian Journal of Soil Science, 61:55-65. LOMBIN, L.G. 1983(a) 'Evaluating the micronutrient fertility of Nigeria's semi-arid savanna soils: I. Copper and manganese'. Soil Science,135:377-84. LOMBIN, L.G. 1983(b) 'Evaluating the micronutrient fertility of Nigeria's semi arid savannah soils: 2. Zinc'. Soil Science.136:42-47. LOMBIN, L.G. 1988 'Agronomic management of nitrogen fertilizers in the semi-arid savannah for enhancing their N-supplying efficiency'. International Symposium on Scrub SavannaStudies, Bauchi. LOMBIN, L.G.; ABDULLAHI, A. 1977 'Long term fertility studies at Samaru Nigeria: II Effect of farm-yard manure on monocropped cotton, sorghum and groundnuts and a rotation of the three crops under continuous cultivation'. Samaru Miscellaneous Paper 72, Institute for Agricultural Research. LOMBIN, L.G.; SINGH, L. 1986 'Fertilizer responses of ground- nuts (Arachis hpogaea L.) under continuous intensive cultivation in the Nigerian savannah'. Fertilizer Research, 10:43-58. LRC 1979 The Land resources of central Nigeria. Land Resource Study 29, Land Resources Development centre, Tolworth. LRD 1972 The land resources of North East Nigeria. Land Resource Study 9. Land Resources Division, Tolworth. MABOGUNJE, A.L.;GANA, J.A. 1981 Rural development in Nigeria: case study of the Funtua integrated rural development project, Kaduna State, Nigeria. University of Ibadan. MADAKI, A.I. 1987 Combatting desertification in Kukawa Local Government Area of Borno State: a critical appraisal. B.Sc. Dissertation, Department of Geography, University of Maiduguri. MANR, KANO 1985 Letter from Ministry of Agriculture and Natural Resources, Kano State to General Manager, Hadejia-Jama'are River Basin Development Authority, July. MATLON, P.J. 1987 'The West African semi-arid tropics', in J.W. Mellor; C.L. Delgado; M.J. Blackie (eds.), Accelerating food production in sub-Saharan Africa, Johns Hopkins University Press: 59-770 - 55 - MAURYA, P.R. 1982 'Development of salinity and alkalinity in irrigated soils of Nigeria'. Fourth Afro-Asian Conference of ICID, Lagos: Vol.ii, 1-17. McTAINSH, G.1980 'Harmattan dust deposition in northern Nigeria'.Nature, 286,587-8. McTAINSH, G.: WALKER, P.H. 1982 'Nature and distribution of Harmattan dust Zeitschrift fur Geomorphologie, 26/4:417-35. MOHAMMED, A.S. 1985 Some effects of the 1983 drought on crop production and farm sizes In Bosso, Niger State, Nigeria. B.Sc. Dissertation, Department of Geography, Bayero University, Kano. MOHAMMED-SALEEM, M.A.; OTSYINA, R.; VON KAUFMANN, R. 1986 'Some methods for improving fodder by incorporating forage legumes in cereal cropping systems in the Nigerian subhumid zone'. In: I. Haque; S. Jutsi; P.J.H.Neate (eds.), Potentials of fozage legumes in farming systems of sub-Saharan Africa, ILCA, Addis Ababa: 363-718. MOHAMMED-SALEEM, M.A.; SULEIMAN, H.; OTSYINA, R.M. 1986 'Fodder Banks: for pastoralists or farmers.' In: I. Haque; S.Jutsi; P.J.H.Neate (eds.), Potentials of forage legumes in farming systems of sub-Saharan Atrica, ILCA, Addis Ababa: 420-37. MOKWUNYE, A.U. 1977 'Role of inorganic fertilizers in the chemical degradation of Nigerian savanna soils'. Expert consultation on methodology for assessing soil degradation. Food and Agriculture Organisation of the United Nations, Rome; Samaru Conference Paper 149 Institute for Agricultural Research. MORMONI, Z.L. 1976 The effect of the 1972/73 drought on the economic role of women in Hurumi haMlet Kausanit ano State. B.A. Dissertation, Department of Geography, Ahmadu Bello University, Zaria. MORRIS, J. 1981 'Agrarian structure implications for develop- ment. A Kano (Nigeria) case study.' Oxford Agrarian Studies.10:44-69. MORTIMORE, M.; OLOFIN, E.A.; CLINE-COLE, R.A.; ABULKADIR, A. 1987 (eds.) Perspectives on land administration and development in northern Nigeria. Department of Geography, Bayero University, Kano. . 56 - MORTIMORE, M. 1968 'Population, settlement and soils in Kano Province, Nigeria, 1931-62' In: J. C. Caldwell et al. (eds.) The population of tropical Africa, Longman: 298- 306. MORTIMORE, M. 1988 'The resilience of human populations in the tropics: an example from Nigeria'. Tropical Agriculture Association, London, 22 September. MORTIMORE, M. 1989 Adapting to drought: farmers, famines and desertification in West Atrica, Cambridge University Press. MORTIMORE, M. in press 'The intensification of peri-urban agri- culture: the Kano Close-Settled Zone, 1964-86'. In: B. L. Turner et al. (eds.), Population growth and agricultural change in Africa. NCEP 1988 Report on the tour of flood and erosion affected areas of Katsina State, October. National Committee on Ecological Problems. NDAKS, G.D. 1976 The effects of the 1972-73 drought on the agricultural community of Sandamu in Daura Emirate. B.A. Dissertation, Department of Geography, Ahmadu Bello University, Zaria. NGGADA, A.Y. 1988 Population density and deforestation in Sakwa District, Biu. B.Sc. Dissertation, Department of Geography, University of Maiduguri. NNADI, L.A. 1983 'Utilization of legumes in arid soils'. International Workshop on uses of microbiological processes in arid lands for desertification control and increased productivity, Albuquerque/Sante Fe. NNADI, L.A.; HAQUE, I. 1985 'Forage legume-cereal systems: improvement of soil fertility and agricultural production with special reference to sub-Saharan Africa.' In I.Haque; S.Jutsi; P.J.H.Neate; (eds.), Potentials of forage legumes in farming systems of sub-Saharan Africa, ILCA, Addis Ababa: 330-62. OLOFIN, E.A. 1980 'Gully erosion consequent on the construction of the Tiga Dam in the Kano Basin'. 23rd Annual Conference of the Nigerian Geographical Association, Abstracts:219-22. OLOGE, K.0. 1972 'Gullies in the Zaria area: a preliminary study of headscarp recession'. Savanna, 1/1:55-66. OLOGE, K.0. 1974 'Kudingi Gully: headscarp recession, 1969-73'. Savanna.3/1:87-90. - 57 - OLOGE, K.O. 1975 'A quick preliminary survey of soil erosion in north-western Nigeria'. Department of Geography, Ahmadu Bello University, Zaria. OLOGE, K.O.; JAIYEOBA, J.A.; SAM, H.M. 1988 'Soil degradation under hand-hoe and mechanical tillage: a case-study from the Zaria area'. Department of Geography, Ahmadu Bello University, Zaria. OGUNSANYA, 1.0. 1986 Soil erosion rates under Northern Guinea Savannal Northern Nigeria: an evaluation based on-the Universal Soil Loss Equation. B.A. Dissertation, Department of Geography, Ahmadu Bello University, Zaria. PAPKA, P.M. 1984 The role of forestry in integrated land use potential in Borno State Nigeria, M.Sc. Thesis, University College of North wales, Bangor. PARKER, D.E.; FOLLAND, C.K.; WARD, M.N. 1987 'Sea surface tem- perature anomaly patterns and prediction of seasonal rainfall in the Sahel region of Africa'. International Geographical Union Study Group on recent climatic change, Sheffield. PATRICK, S. 1987 Gully erosion in Gongola and Bauchi States, Niseria.Ph.D Thesisq University of London. PATRICK, S. 1989 'Agricultural practices and land degradation (soil erosion) in parts of Gongola State - Nigeria'. Workshop on Land Development and Administration in Northern Nigeria, Bayero University, Kano. PENNING DE VRIES, F.W.T.; DJITEYE, M.A. 1982 La productivite des paturages Saheliens. Centre For Agricutural Publishing and Documentation, Wageningen. PEYRE DE FABREGUES, B. 1984 'Quel avenir pour 'elevage au Sahel?' Revue d'Elevage et Medecine Veterinaire des Pays Tropiques, S7hOd-. PULLAN9 R.A. 1974 'Farmed parkland in West Africa'. Savannat 3/2:119-52. RAYAR, A.J.; HARUNA, B.U. 1985 'Studies on distribution of total and available nitrogen in the soils of South Chad irrigation Project area of Borno State'. Annals of Borno, II:105-11. ROSS, P.J. 1987 'Land as a right to membership: land tenure dynamics in a peripheral area of the Kano Close-Settled Zone'. In: M.Watts (ed), State, oil and agriculture in Nizeria. Institute of International Studies, University of Ca=ornias 223:47. - 58 - SAGUA, V.0. et al. (eds.),1987 Ecological disasters in Nigeria: drought and desertification. Federal Ministry of Science an Technology, Lagos. SINGH, L.: BALASUBRAMANIAN, V. 1977 'Effect of continuous application of chemical fertilizers on the organic matter levels of soils at Samaru, Nigeria'. 13th Annual Conference of the Agricultural Society of Nigeria, Zaria; Samaru Conference Paper 17, Institute for Agricultural Research. SINGH, L.; ZANGO, U.A.; EGHAREVBA, P.N.; OGUNLELA, V.B.; BALASUBRAMANIAN, V. 1983 'Proposed new fertilizer recommendations for sole crop millet'. Institute for Agricritural Research. SRRBRDA 1987 'Memorandum on soil erosion and flood control measures for the consideration of Sub-committee on Erosion and Flood Control'. Sokoto-Rima River Basin and Rural Development Authority. TAYYIB, H.M. 1981 Soil management in Northern Sudan Zone. A comparative study of soil tertility status under differing management systems in Danbatta area. B.Sc. Dissertation, Department of Geography, Bayero University Kano. THAMBYAHPILLAY, G.G.R. 1987 'Methodological and climatological perspective of drought and desertification in the Lake Chad Basin; Sahelo-Sudan Nigeria'. International Conference, Water Resources of the Lake Chad Basin. Lake Chad Basin Commission, N'Djamena. TOURTE, R. 1971 'Themes legers. Themes lourds. Systemes intensifs. Voies differentes ouvertes au developpement agricole du Senegal'. Agronomie Tropicale, 26: 632-71, UBALE, A. 1985 An analysis of recent changes in fadama farming in Ringim. B.Se. Dissertation, Department of Geography, Bayero University, Kano. UMAR, A.0. 1981 The relationship between interception loss and soil properties under savanna trees. B.Sc Dissertation, Department of Geography, Ahmadu Bello University. UNEP, 1977 World Map of desertification. United Nations Environment Programme, Nairobi. VAN APELDOORN, G.J. 1981 Perspectives on drought and famine in Nigeria. London: Allen and Unwin. WATSON, K.A. 1964 'Fertilizers in Northern Nigeria. Current utilization and recommendations for their use'. African Soils, 9:5-20. - 59 - WATTS, M. 1983 Silent violence. Food, famine and the peasantry in northern Nigeria. University of California Press. WEDDERBURN, S.G. 1988 Nigerias: key issues in formulating ferti- lizer price policy: an evaluation of multiple impact of alternative price levels. Agricultural Projects Monitoring and Evaluation Unit, Kaduna. WEDDERBURN, S.G.; BAMIDELE, J.; OKONJO, C. n.d. 'Crop yield response to fertilizer use at selected ADPs'. Agricultural Projects Monitoring Evaluation and Planning Unit, Kaduna. WFS 1984 The Nigerian Fertility Survey, 1981-82. A summary of findings. World Fertility Survey, London. WILD, A.1972 'Mineralization of soil nitrogen at a savanna site in Nigeria'. Experimental Agriculture, 8:91-7. WILKE, B.M.; DUKE, B.J.; JIMOH, W.L.O. 1984 'Mineralogy and chemistry of Harmattan dust in northern Nigeria . Catena, 11/91-96. WRECA, 1935 Groundwater monitoring in Kano State. Kano: Water Resources and Engineering Construction Agency. YARI, K.M. 1981 Desertification in northern parts of Kaduna State: case studies of YardaJe and Rijiyar Tsamiyar in Daura Emirate. B.Sc. Dissertation, Department of Geography, Bayero University, Kano. YAYOCK, J.Y.; LOMBIN, L.G.; NNADI, L.A.; BALASUBRAMANIAN, V. 1978 'Fertilizer use in Nigeria. I. Trends in consump- tion'. Nigerian Society of Chemical Engineers Annual Conference. YAYOCK, J.Y.: LOMBIN, G.L. 1981 'Sulphur deficiency and fertilization requirements of Nigerian savanna soils'. International Symposium on the use of Sulphur for Development and Modernization of Agriculture; Mexico City Samaru Conference Paper 38, Institute for Agricultural Research. YAYOCK, J.Y.; OWONUBI, J.J. 1985 'Weather-sensitive agricul- tural operations in groundnut production: the Nigerian situation'. International Symposium on Agro-meteorology of groundnut, Niamey. ZAKARI, A. 1985 Assessing erosion hazard through slope and erodibility indices, Eniko Basin, Niger State. M.Sc. Thesis (Land Resources), Bayero University, Kano. - 60 - APPENDIX 1 Average monthly rainfall at 10 synoptic stations in the study area, 1950-69 (a) and 1968-87 (b) mm. Station June July August September (a) (b) (a) (b) (a) (b) (a) (b) 1 Nguru (19) 75.7 45.7 139.5 129.7 210.3 152.2 82.8 56.5 2 Maidtiguri(15) 91.4 69.8 203.2 150.3 251.7 168.1 95.5 102.8a 3 Katsina (40) 96.8 62.8 192.0 161.1 268.5 188.4 116.3 100.7 4 Sokoto (20)105.4 78.7 185.2 153.5 249.7 188.5 133.6 82.0 5 Potiskum (30) 98.8 66 .3a 216.2 1 87 .5 a 275.6 1 9 9 .4 a 137.7 96 .9b 6 Kano (41)132.8 107.5 210.8 178.5 313.9 218.0 132.8 99.1 7 Gusaua (19)157.0 114.3 217.7 201.5 303.8 233.1 195.8 167.0 8 Yelwa (18)159.0 127.3 206.8 205.3 236.5 223.7 213.4 173.7 9 Bauchia (48)150.1 136.4 253.5 267.1 246.2 267.8 185.2 147.2 10 Zariab (42)165.6 120.3 221.5 250.3 281.4 260.8 230.4 143.6 a 19 years only Col. (b) b 18 years only Col. (b) (18) No. of years used in calculating mean in Col.(a) * usually ending 1969. APPENDIX 2 Terms of Reference for a Study into the Causes, Nature and Rate of Soil Degradation in the Northernmost States of Nigeria and an Assessment of the Role of Fertilizer in Counteracting the Processes of Degradation 1. Soil fertility in the Northern states of Nigeria has been maintained over long periods of time by the use of organic manures and fallowing. Under the impact of rapid population growth these systems are coming under strain in some areas with a possible adverse impact on the long term soil conditions and on crop yields. This study is intended to analyse the linkages between increasing population and soil degradation, define its nature and its impact on production, assess farmer response to the situation and the role which fertilizer may play in reversing a loss of soil quality. The Area of the Study 2. The study will cover the following states: Sokoto Katsina Kano N. Bauchi N. Borno It will provide a brief description of the major soil, vegetation and climatic zones of this region with quantitative data on the extent of each of the zones. The Evidence of Soil Degradation and Its Nature 3. The study w4 11 provide evidence from existing data that soil degradation is taking place in this region. It will analyse evidence of the nature of that degradation in terms of changes in soil quality including: Soil organic carbon content; Total nitrogen content of the soil; pH Soil fauna Permeability It will seek to disaggregate the impact of cultivation and cropping on soil quality in situations with and without significant erosion hazards. - 62 - The Rate of Degradation and Impact on Yields 4. The study will gather evidence of the rate of degradation resulting from declining fallow periods and per ha availability of manure in the major soil and climatic zones covered by the study. It will relate this to evidence of declining yields with quantitative data on the losses involved for the major annual crops of the zone. Changes in Farming Patterns Resulting from Degradation 5. The study will record available evidence of farmer response to declining fallow and soil degradation. This will include any changes in cropping patterns or in farming practices and in the comparative importance of non-crop production activities. The Extent of the Problem 6. The study will review the available evidence relating to the levels of population density at which traditional systems break down under the major soil and climatic conditions of the region. It will link this to the available population information on the states and thereby provide a broad assessment of the scale of the problem. Ameliorating Soil Degradation 7. The study will review the available evidence of the role of inorganic fertilizer in slowing down the rate of soil degradation under annual cropping systems. It will relate this to any other practicable and proven farming practices which can enhance the role of fertilizer in checking degradation. It will specifically detail the impact of this intervention on the various factors involved in soil degradation in the region as outlined in para 4 above. The Study Process and Product 8. It is expected that the study will take a total of six weeks to prepare, that it will rely on material available internationally and in the research stations and universities in the study area. It is anticipated that at least part of the study will be carried out in Nigeria. The document will be of approximately forty pages and will include three maps of the study region covering: Major soil types; Major climatic zones; and Population distribution. - 63 -