Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

14: Land use and cover change in the southern forest-savanna transition zone in Ghana: A sequence model

Study area
Conceptual basis
Land use and cover sequences
Food cropping on abandoned land
Land use and cover sub-sequences
General indications and future trends

G. T. Agyepong and S.K Kufogbe


It has become urgent to understand land use and cover changes that have taken place in diverse environments and which have intensified over the past 100 years or so in many tropical countries, particularly those of Sub-Saharan Africa. These changes occur essentially at the spatial scale of the farm field and the locality. The cumulative effects in terms of biogeochemistry, abundance and composition of terrestrial species, heat and water fluxes, the atmosphere and human living are, however, regional and global. Hence the increasing global concerns about land use and cover changes.

The comprehensive understanding of these complex biophysical and human relations form the basis of a major project proposal to relate land use and cover changes to global environmental change (Turner et al. 1993).

Some of the key questions to be addressed concern how land use and cover have changed in the last 300 years and how these will change in the next century or so. In order to address these relevant questions it is suggested, among other things, that it is necessary to obtain a comprehensive understanding of the dynamics of land use change, and how these changes influence land cover over time, using models. It is believed that this will enable the development of land use and cover change typologies and the demarcation of situations. An interdisciplinary investigation has been undertaken by the Ghana (West Africa) cluster of PLEC in the forest-savanna ecotone in Ghana to examine the conditions of environmental change. A sequence of land use and cover changes has been observed.

These observations form the basis of this chapter, which seeks to abstract and rationalise the essential human and physical factors and processes of the land use and land cover changes in the study area in a descriptive model.

Study area

The southern forest-savanna transition zone in Ghana constitutes a broad ecological area between the dry coastal savanna and the dry semi-deciduous forest zone (Hall and Swaine 1976).

The in-depth field study which constitutes the basis for this paper was concentrated at three sites between 60 km and 120 km inland from Accra. The sites are Sekesua in the migrant Krobo district, Yensiso in Akuapem, the cradle of Ghana's cocoa industry, and Amanase, in the historic southern Akyem cocoa district (see fig. 5.1, p. 39). Each study site comprises a cluster of farming villages/communities which are distinguished on the basis of ethnicity and migration history.

Within the broad framework of the United Nations University (UNU) Population (now People), Land Management and Environmental Change (PLEC) programme, data were collected through interdisciplinary and collaborative efforts from archival sources and field observations, including traverse measurements as well as interpretation of available 1974 aerial photographs.

Conceptual basis

Successional development is an important concept in ecological studies. Ecological succession comprises the sequential changes that occur in the structure and functioning of a biotic community at a particular location in water or on land. Thus, the pioneer plant and animal communities that colonise bare ground and water bodies undergo replacements that may be observed in the species composition, physiognomy and functions of the communities until a dynamically stable condition is achieved that is consistent with the biophysical environment and the factors that predominate, for example climate, soils or human influences.

The concept of ecological succession provides a useful tool for the study and analysis of the development of plant and animal communities. The concept, which was originated by Cowles (1899) and Clements (1916, 1936) during the late ninteenth and early twentieth centuries, has undergone considerable debate, but the basic principles of succession remain valid. (See, for example, Whittaker 1953).

Land use and cover sequences express change in the use of land, particularly for agricultural production, as the result of changes in technology, the socioeconomic and biophysical environment. Technologically induced changes include the introduction of new crop species that may be superior to existing ones. Socioeconomically induced changes occur as the result of changes in demand and price responses or policies. Deteriorating environmental conditions arising from natural causes, such as decreased soil fertility, drought or pests, may lead to a change in land use and cover conditions. The mix of influencing factors ensures that, though general prediction of land use and cover change is possible in a given natural environment, it can only be done within a range of human choices.

Land use succession may be said to progress in the reverse direction of vegetational succession in terms of process and end state. The end states are the varying degrees of modifications or changes in the environment and the productive potential of the land.

Both sequences are characterisedby change sequences of energy flow and material cycling with stages determined by dominant physical and human environmental factors. Human forces dominate the land use sequences, which may thus be said to progress in a direction opposite to that of the natural ecological sequences.

It is proposed here that the changes in land cover that have taken form especially in the study sites may be modelled within the concept of ecological succession. This chapter presents an initial sequence model for the rationalisation of the findings of a multidisciplinary study in the forest-savanna ecotone in Ghana (Gyasi et al. 1994).


A small number of studies on land use and cover sequence have been undertaken in West Africa (Akin 1958; Clayton 1958), while extensive studies have been conducted on nutrient cycling under shifting cultivation (Nye and Greenland 1960).

The general observations suggest that the ecological succession concept is applicable. Clayton (1958) described the secondary vegetational sequence from forest to savanna near Ibadan in Nigeria. He observed three trends of cultivation succession:

  1. normal rotation of cultivation-thicket-secondary forest-recultivation;
  2. degeneration in the nature of the fallow as a result of frequently repeated cultivation;
  3. decline of cocoa plantations as the canopy opens by die-back as inter cultivation commences.

Ahn (1958) has described land use and cover development in relation to vegetational sequences, bushfires and bush fallow cultivation in the western forest areas of Ghana. Rose-Innes (1964) has also described the land use processes leading to land degradation in the savannas of northern Ghana (fig. 14.1). In these studies the vegetational sequences are emphasised rather than the land use sequences.

Land use and cover sequences

Land use succession as identified in the present study area and as may be used for the biophysical, socio-economic and technological analysis of land use change in the forest areas of Ghana is described in figure 14.2.

A sequence involving cocoa cultivation as an export cash crop, food cropping as both commercial and subsistence activities, oil-palm and other tree crops as cash crops is shown (Akin 1958; Dickson 1969; Dickson and Benneh 1988; Field 1943; Gyasi et al. 1994; Hill 1963; Howard 1978). The technological and socio-economic conditions associated with each of these land uses as well as the factors influencing them provide the basis of analysis and characterisation of the various stages of the development of the land use and cover sequences.

Figure 14.1 General Land Use and Cover Change Model in the Savannas of Northern Ghana (Source: Rose-Innes 1964)

Figure 14.2 Land Use and Cover Sequences in the Yensiso-Sekesua-Amanase Study Areas

(r) Decisions influenced by internal factors of population, urbanisation, transportation and policy

Mainly external factors of international production demand and prices

Cocoa Cultivation

Cocoa cultivation on virgin forest land involves the slashing of the undergrowth of shrubs, saplings of emergent tree species and of the stems of climbers that may reach the tree canopy. Selective removal of trees is carried out by felling or by burning and the debarking of the trees at the base above the ground to open up the canopy to allow enough light in for growth of the cocoa trees and the broad-leaved food crops, for example cocoyam and plantain. A modified environment and habitat are created. In the study areas forest with varying degrees of canopy degradation or fallow regrowth occupied between 38 per cent and 56 per cent of the study sites. Closed canopy occupied 0.2 per cent. At maturity, the cocoa plants provide the lower canopy of the modified land cover and habitat.

The Abandoned Cocoa Land and Land Use Choices

The bearing life span of the cocoa plant is about 30-40 years, depending upon the soils. The trees may die of disease, fire or age. The land use is then abandoned, although the harvesting of fruits and food crops such as cocoyam, yam and plantain may continue. The farmer's decision as to the use to which the abandoned cocoa land is put depends upon several factors, including principally the relative prices and profitability of cocoa, other tree crops such as oilpalm, coffee, citrus and food crops of cocoyam, plantain, maize, cassava and vegetables. The relative prices of these crops are influenced by population growth, urbanisation and the demand for food, industrialisation and the demand for agricultural raw materials, the external production conditions and markets, such as for cocoa and oil-palm, and finally upon government policies. External factors for Ghana have included increasing competition from other countries which also produce these commodities.

Within these choices, specific cropping patterns have been adopted to offset the changing conditions in the biophysical environment: for example, the increasing preference for maize and cassava as compared to plantain and cocoyam, and the preference for varieties of the same crops, such as cassava, that may be more suitable to the changing environmental conditions (Gyasi et al. 1994).

The sequence of uses and cover is, therefore' not unidirectional, and depends upon important factors. Within any choice the technology adopted is important as regards the end state of the environment and the productivity of the land. In the study area, the change from tree crops to food crop cultivation involves change from simple slash and burn with machetes and fire and no tilling to the increasing use of more disruptive methods, e.g. stumping and hoeing.

Cutting of New Forest

The normal reaction of the farmer to an old or diseased, non-bearing cocoa farm is to search for new forest land locally or outside the locality. This was possible in the past up to the 1960s, when migrant cocoa farming was an important phenomenon in the forest areas of Ghana (Arhin 1985; Hill 1963). This land use choice is no longer available, with only 0.48 per cent of Ghana's high forest area of 86 million ha remaining outside the reserve system (Ghana 1987). Estimates based on aerial photographs of land under closed canopy forest in the study sites are insignificant in both Yensiso and Amanase, where such land accounts for less than 1 per cent in both areas (Gyasi et al. 1994; see also chap. 7, this volume).

Secondary Regrowth Forest

Under conditions of abundant forest lands, the abandoned cocoa farm would be left to develop into a secondary regrowth forest. The development of regrowth vegetation in the forest has been described by Ahn (1958). The use of such land includes the gathering of non-timber forest products (NTFP), such as canes, fruits, leaves and animals. Commercial timber is also cut. The regrowth forest may eventually be cultivated to cocoa. This choice is hardly available now because the cocoa land can hardly be left for a sufficiently long period for forest conditions to develop, a period generally above 25 years. This used to be the felling cycle recommended by the Forestry Department in the reserved forests.

Replanted Abandoned Lands

Abandoned lands may be planted in cocoa, oil-palm, citrus or other tree crops. Cocoa farming in the study sites (and indeed in southern Ghana in general) was devastated by the swollen shoot disease beginning in the late 1930s (Dale 1962). A project to rehabilitate and replant old diseased cocoa farms by the Ministry of Agriculture was started in 1948 in parts of the country that include the present study areas. The results of the various phases of the project indicate that replanting is yet to be widely developed, though it is considered important in a situation where no more forest land is available.

It has been observed that only 12 per cent of those operating the generally old cocoa farms had attempted to rehabilitate them (Gyasi et al. 1994). The problems of replanting may partly arise from the loss of income to the farmer while rehabilitation and replanting take place. The modified environmental conditions resulting from the first cycle of cultivation may also pose problems, for example of fire. Replanting may be done under existing modified forest cover conditions of the old farm or without a forest cover, in which case broadleaved food crops are used to protect the young cocoa plants. Oil-palm cultivation also requires complete conversion of the forest cover. Experiments are being conducted at the Cocoa Research Institute at Tafo to find solutions to the problems of replanting.

Food cropping on abandoned land

Abandoned cocoa farm land may be put to food cropping under one of two systems: (1) a bush fallow system; (2) a technologically improved system. The bush fallow system relies on natural fallows for the regeneration of soil fertility. Fire is an important tool for the clearing of land. Fire and the axe eventually eliminate the tree and other woody cover, which leads to the dominance of grasses and increasingly the obnoxious weed Chromolaena odorata (Hall et al. 1972). This process is accelerated by the increasing population and urban concentrations, raising the demand for food and putting pressure on the land and thus leading to a reduction in the fallow period and soil fertility. The present study areas are located in the supply hinterlands of major urban centres: Accra (1,179,484), Koforidua (60,675), Nsawam (25,983) and Suhum (26,436). The soil organic carbon percentage had decreased from an average of 2.7 per cent for uncultivated virgin sites at all sites to an average of 1.2 per cent at sites cultivated in maize and cassava (Gyasi et al. 1994).

Continued pressures lead to the introduction of crop types and varieties that are more efficient in the utilisation of water and minerals, but which may require a more thorough disruption of the cover. In time, the woody root systems in the soil are again removed and hoe cultivation is introduced, which increases the hazards of erosion and further deterioration of the productive capacity of the environment. Hoeing (generally characteristic of the savanna cultivation systems) was common, for example, in the Amanase area.

Over extensive areas, the end state moves in the direction of the dominance of grasses. This is already observable in several locations at all the study sites.

The alternative conversion of abandoned cocoa land into a food crop area may be undertaken with improved technologies, such as an agroforestry system and the use of mineral fertilizers. Official extension policy has introduced and encouraged these technologies. This stage in the development of the land use is yet to be systematically adopted. Indications from the PLEC study, however, show adoption ratios of these strategies at 19.5 per cent for chemical fertilizers and 3.2 per cent for agroforestry (Gyasi et al. 1994).

Land use and cover sub-sequences

The land use and cover sequence observed in the forest-savanna ecotone is not unidirectional. A number of sub-sequences may be identified on the basis of the technologies used, with important implications for sustainability of land use and cover. These include the migrant cocoa farmer sub-sequence (boxes 1, 2, 3 in fig. 14.2). As has been shown earlier, this sub-sequence is no longer available because no more forest remains. Three sub-sequences are considered critical to the future development of land use and cover in the southern forest-savanna ecotone and, indeed, the whole forest area of Ghana (fig. 14.2):

  1. the technologically unimproved food cropping sequence comprising boxes 2,3,5 and 6;
  2. the technologically improved food cropping sequence comprising boxes 2, 3,7; and
  3. the cocoa and other tree crop replanting comprising boxes 2, 3, 4.

General indications and future trends

The general indications of the field data are that:

The future conditions of land use and cover, however, depend upon the extent to which the tree crop replanting and the unproved food cropping subsequences will develop to provide land use and cover conditions that are productively and environmentally sustainable. Both sub-sequences are poorly developed, although in the case of cocoa replanting, the study areas particularly were important sites for a major cocoa rehabilitation project in the 1970s and are currently the subject of research at the Cocoa Research Institute of the Council for Scientific and Industrial Research (CSIR). Improved food cropping is a major goal of current multifaceted extension programmes involving the Crop Research Institute of the CSIR, the Crop Services and the Extension Service departments, the Forestry Department and Non-governmental Organizations (NGOs). It will be necessary to subject the land use systems in these sequences to systematic analysis as a PLEC objective to determine the possibilities and constraints to development in the immediate and long-term perspectives.


Ahn, P.M. 1958. Regrowth and swamp vegetation in the western forest areas of Ghana. J. W. Afr. Sci Ass. 4: 163-73.

Arhin, K. 1985. The Expansion of Cocoa Production: The Working Conditions of Migrant Cocoa Farmers in Central and Western Regions. Legon: Institute of African Studies.

Clayton, W.D. 1958. Secondary vegetation and the transition to savanna near Ibadan, Nigeria. Journal of Ecology 46: 217-37.

Clements, F.E. 1916. Plant succession: an analysis of the development of vegetation. Carnegie Institute of Washington Publication, No. 242.

Clements, F.E. 1936. Nature and structure of the climax. Journal of Ecology 24: 25284.

Cowles, H.C. 1899. The ecological relations of the vegetation on the sand dunes of Lake Michigan. Bot. Gaz. 27.

Dale, W.T. 1962. Diseases and pests of cocoa. In: J.B. Wills, ea., Agriculture and Land Use in Ghana. London: Oxford University Press.

Dickson, K.B. 1969. A Historical Geography of Ghana. London: Cambridge University Press.

Dickson, K.B. and G. Benneh. 1988. A New Geography of Ghana. London: Longman.

Field, M.J. 1943. The agricultural system of the Manya-Krobo of the Gold Coast. Africa 14(2): 54-65.

Ghana, Forestry Department. 1987. Annual Report. Accra.

Gyasi, E.A. , G.T. Agyepong, E. Ardayfio-Schandorf, L. Enu-Kwesi, J. S. Nabila and E. OwusuBennoah. 1994. Environmental Endangerment in the Forest-Savanna Zone of Southern Ghana: A Research Study Report. Tokyo: UNU.

Hall, J.B., R. Kumar and A.A. Enti. 1972. The obnoxious weed Eupatorium odoratum (Chromolaena odorata). Ghana Journal Agric. Sci. 5: 75-78.

Hall, J.B. and M.D. Swaine. 1976. Classification and ecology of closed canopy forest in Ghana. Journal of Ecology 64: 913-51.

Hill, P. 1963. The Migrant Cocoa Farmers of Southern Ghana: A Study in Rural Capitalism. London: Cambridge University Press.

Howard, R. 1978. Colonialism and Underdevelopment in Ghana. London: Croom Helm.

Nye, P.H. and D.J. Greenland. 1960. The Soil Under Shifting Cultivation. Farnham Royal, Bucks: Commonwealth Agricultural Bureau.

Rose-Innes, R. 1964. Ecology in Land and Water Survey in Upper and Northern Regions, Ghana. Rome: FAO.

Turner, B.L., R.H. Moss and D.L. Skole. 1993. Relating land use and global land cover change: a proposal for an IGBP-HDP Report, No. 5.

Whittaker, R.H. 1953. A consideration of climax theory: the climax as a population and pattern. Ecol. Monogr. 23.

15: Women, environmental change and economic crisis in Ghana

Background to the research: Economic crisis and structural adjustment
Environmental degradation in North-Eastern Ghana
Gender and agricultural systems in North-Eastern Ghana
The gender division of labour
Structural adjustment and its impact on health, nutrition and consumption patterns
Changes in educational status
Changes in income-generating activities
Changes in women's time use
Women's time use and seasonality

Mariama Awumbila


The severity of economic conditions in Sub-Saharan Africa during the 1980s resulted in many African countries adopting structural adjustment programmes (SAPs) sponsored by the International Monetary Fund and the World Bank, in an effort to restructure their economies. In the short term, however, some of the policies adopted have worsened conditions for the most vulnerable groups in the population. Widespread concern now exists about the deteriorating standards of living and the severe erosion of the human resource base of the economy following the economic crisis and resulting adjustment (Cornia et al. 1987). A more recent debate centres around how crisis is experienced by different members of the household, it often having a different impact on women and men (Commonwealth Secretariat 1989).

The short-term objectives of the SAPs are not only contributing to an increase in economic and gender inequalities, but also exacerbating the trends towards environmental degradation as women and men struggle to survive on an increasingly deteriorating resource base. As environmental degradation accelerates, women - who have primary responsibility for the management of the land and household reproduction - are often seen as the source of unsustainable exploitation of the resource base. The debate on the relationship between gender, environment and development has evolved from women being seen as victims of environmental degradation, often bearing the heaviest burden of environmental change, to an emphasis on their roles as efficient environmental managers within the development process and. more recently, to their important role as solvers of environmental problems, playing an important role in both conserving and improving the environment. Thus, women are seen as having a multiplicity of roles as victims, agents and saviours in relation to environmental change.

Figure 15.1 The upper East Region of Ghana

However, little attention has so far been focused on the spatial dynamics of crisis and adjustment. Regions which are ecologically marginal and politically peripheral often develop coping strategies in response to climatic and economic changes which often have a gendered impact.

Using recent research in rural north-eastern Ghana, a stressed savanna environment (fig. 15.1), the purpose of this paper is to contribute to the ongoing debate concerning the extent to which structural adjustment programmes, even if unintentionally, have disadvantaged members of low income households on the basis of gender. It explores the link between adjustment policies, deteriorating environmental conditions and the feminisation of poverty, through a comparative analysis of women's time use as a measure of changes in gender roles under environmental and economic stress.

Fieldwork was undertaken in Zorse, a village in north-eastern Ghana (fig. 15.1). A longitudinal case-study of rural low income women in Zorse was undertaken in 1984, a year of drought and the beginning of an SAP programme, and in 1990, a year of more abundant, though poorly distributed, rainfall and eight years into the SAP programme.

Background to the research: Economic crisis and structural adjustment

After period of relative progress, the late 1970s and early 1980s were a period of economic crisis for many developing countries Ghana's economic decline began in the late 1960s. A positive growth rate of 2.1 per cent per annum between 1960 and 1970 declined to -6.1 per cent per annum between 1980 and 1983, and the inflation rate increased from 9 per cent per annum in 1970 to 122 per cent in 1983. Stagnation or decline characterised almost all sectors of the economy, while the population increased at a faster rate. The social impact was just as severe, with a deterioration in health and educational standards and a reduction in real incomes per head. The result of the economic decline was a general impoverishment of the nation as a whole, with about 50 per cent of urban households and 65-75 per cent of rural households living below the poverty line by the early 1980s (UNICEF 1988). Prolonged drought, bushfires and the expulsion of nearly 1 million Ghanaians from Nigeria in 1983 put severe strain on an already critical food and unemployment situation and exacerbated the crisis.

Faced with the desperate economic situation, the government of the PNDC adopted in 1983 an IMF/World Bank sponsored structural adjustment programme aimed at halting the economic decline and restructuring the economy to foster growth and development. The programme had as its guiding principle the "realignment of the price and incentive system in the economy in favour of the productive, particularly the export sectors" (Government of Ghana 1987).

The SAP adopted in Ghana was a fairly standard IMF and World Bank package and included the following policies: demand restraint through cuts in government expenditure, public sector employment and real wages; price decontrol involving the removal of subsidies on food, consumer items and agricultural inputs; introduction or increases in user charges for social services; trade liberalisation and currency for social devaluation; long-term supply policies aimed at raising the long-term efficiency of the economy through privatisation of state owned institutions; and credit reform.

The short-term impact of these policies, in comparison with 1982/1983 levels, was the recovery of the economy. Between 1984 and 1989, GDP average annual growth rates were restored to the levels of the early 1970s, registering growth rates of 5 per cent per annum, although it has been pointed out that much of this recovery was due to improved environmental factors and improved terms of trade rather than to specific policy (Toye 1991). Some sectors registered positive growth rates, with inflation falling by over 90 per cent between 1983 and 1985.

However, the wider social impact has been less positive. Despite the impressive macrolevel growth statistics, both absolute and relative poverty increased among both urban and rural populations during the period of adjustment. It has been pointed out that the costs of adjustment are borne most heavily by those with a relatively poor ability to withstand such losses, thereby deepening poverty. l

Environmental degradation in North-Eastern Ghana

An important feature of the Ghanaian economy is the dichotomy between the southern forest zone, with its more abundant rainfall and resources, fertile soils and relatively developed infrastructure, and the northern savanna, covering about 60 per cent of the country, with low and erratic rainfall, few resources and a poor infrastructure. Since colonial times, the north has remained an area producing largely subsistence crops which are undervalued and underpriced in the national market. It is generally thought that northern underdevelopment is not due to the harsh physical environment or lack of resources, but rather it is a by-product of the need for northern labour by the export-oriented mining and cocoa interests of the south (Bening 1975). Today, the Upper East region (fig. 15.1) remains one of the poorest, the most rural and the least industrialised region in Ghana. It has the nation's highest average population density, outside the capital region of Greater Accra: 87 persons per square kilometre. Densities in the cultivated areas of Bawku District, in which the study village of Zorse is located, reach 270 persons per square kilometre.

In terms of environmental degradation, much of the West African savanna, including north-east Ghana, has been affected by drought conditions since the late 1960s. In Bawku District, the evidence does not indicate a long-term trend of declining rainfall, but rather a cyclical occurrence of drought. In the 1980s the major problem for farmers, as evidenced by rainfall records and supported by interviews, was increasing irregularity of rainfall. Sheet and gully erosion are widespread in the intensively farmed areas around Bawku, and many of the dams and reservoirs were silted up by the end of the 1960s.

Under the relatively low population densities prevailing at the beginning of the century, the main system of farming in the region was shifting cultivation. This has, however, been modified over the years to bush fallow and compound farming systems in response to growing population pressure. Field observation in 1991 revealed land shortage, devegetation, soil deterioration and general degradation of the land (Benneh and Gyasi 1993). The village of Zorse lies in one of the driest parts of Ghana.

Land shortage in the village is extreme because land was taken from the village in the 1940s for a reserve, thus artificially increasing population density in the village and indirectly exacerbating environmental degradation.

Compared to the rest of the country, north-eastern Ghana experiences a shortage of food of greater intensity because of the single rainfall season and the frequency of drought. The situation is further compounded by the fact that basic cash crops in the region are also food crops and many rural households sell off food crops to satisfy non-food needs, often leading to a long pre-harvest "hungry season." The long season of hunger from about February/March to July is characterised by low food stores, a high demand for agricultural labour and increased incidence of water-borne diseases such as malaria. In Zorse, it was common for meals to be missed in the hungry season.

The impact of structural adjustment, with its emphasis on the promotion of exports and the expenditure switching policies it necessitates, has been a structural reinforcement of existing contradictions between northern and southern Ghana, with a further marginalisation of the north.

Contents - Previous - Next