Cover Image
close this bookEnvironment, Biodiversity and Agricultural Change in West Africa (UNU, 1997, 141 pages)
close this folderPilot study of production pressure and environmental change in the forest-savanna zone of southern Ghana
close this folder7: Land use and cover patterns
View the document(introductory text...)
View the documentMethodology
View the documentResults
View the documentSummary
View the documentReferences

(introductory text...)

Methodology
Results
Summary
References

Gotfried T. Agyepong, with the assistance of Sosthenes K Kufogbe

The Yensiso-Sekesua-Amanase study areas lie in the dry, semi-deciduous forest zone of Ghana in the southern marginal and the inner zone subtypes (Hall and Swaine 1976), which are developed on soils of the ochrosol type (Brammer 1962). The zone forms the transition between the moist semideciduous forest in the north and the coastal savanna in the south (see fig. 5.1, p. 39). These forests have been subjected to extensive agricultural exploitation for more than 100 years. The land use pattern has changed from the dominance of a single tree cash crop, cocoa, to a mixed pattern with food crop dominance over the period. The resulting changes in land cover and the productive potential of the land are significant. The socio-economic and environmental implications are important for the sustained development of the zone. This paper examines the spatial patterns of land use and cover as a basis for the analysis of the socio-economic causes of the change in the environment and of the environmental consequences of the land use and cover change. It forms part of the multidisciplinary study of changes in the biophysical environment and agriculture in the southern sector of the forest-savanna transition zone.

Methodology

Two methods have been used in this study: the interpretation of aerial photographs and ground traversing. It became necessary to use aerial photographs for a number of reasons. The time frame and the project budget would not permit the use of satellite images. The use of satellite image analysis was also not adopted because of the spatial scale of study and the known complexity of the land use and cover, which required a better spatial resolution than that of the easily available current satellite programmes. The photographs used were black and white prints flown in 1974 at a scale of 1: 40,000, and borrowed from the Survey Department of Ghana. A print lay-down was prepared to obtain a general view of 25 km2 centred on each of the three study villages, Yensiso, Sekesua and Amanase. Land use and cover was visually delineated on the basis of geometrical shapes, lineations, height of vegetation, presence or absence and the density of tree tone and texture of the land surface. These together with the accounts of the land use and cover history of the areas, as indicated by the people, as well as common knowledge, formed the basis of the final interpretation of the photographs.

The elements and distribution of the current (1993) land use and cover were examined using walking line transect observations over a total distance of 5.295 km at the three sites. Surface cover and use categories were measured along the traverses using chain and line tape. The categories identified are described below.

Results

Photo Interpretation

The photo interpretation identified the following broad land use and cover categories:

  1. Small patches of relatively closed canopy climax forest or mature secondary regrowth forest. These are irregularly distributed and could not always be identified on account of small size and scale. At Sekesua these were associated with terrain that is difficult of access on account of slope.
  2. Areas with broken canopy high forest where the canopy cover, though present, shows varying extents of openness as the result of tree removal. These, as identified in the field, are often associated with cocoa farms which are often diseased and characterised by varying levels of care. These degraded forests also contain food farms, scattered oil-palms and fruit-trees. They are depleted of timber but provide sources of minor forest produce, for example, fuelwood, mushrooms, snails and small mammals.
  3. Currently cropped and fallow land. These were identified by the regular edges of farm plots, low plant cover, and the tone and smooth texture of the surface conditions. Planted rows and field boundaries were identified in the Kwamoso Oil Palm Plantation, which existed at the time of the photography. It was difficult to identify cropped land and fallow lands separately. Fallow and cropped land at various stages contain common elements of crops and non-crop plants in varied mixtures.

Table 7.1 contains the aerial photographic estimates of these land use and cover categories for Yensiso, which is characterised by the Akuapem-Ayigbe-Ewe mosaic landholding and usage pattern (fig. 6.1, p. 44), and for Amanase, which is characterised by both the Akuapem Ayigbe Ewe mosaic land use pattern and by the Krobo-Shai/Siade linear huza pattern (fig. 6.2, p. 45). Broken canopy forests and mature fallows, commonly containing oil-palms, fruit-trees (particularly mango and orange) and staple crops (cassava, plantain and cocoyam) with occasional cocoa farms, covered some 39 per cent of the interpreted Yensiso area, compared to about 57 per cent for Amanase. Amanase's larger proportion apparently reflects the presence of more cocoa farms, which are associated with broken canopy forests. This conjecture is supported by our ground traverse observations, and by the fact that the Amanase area was considered important enough as a cocoa producing area for inclusion in an extensive cocoa rehabilitation project in the 1970s. Current arable cultivation and short fallows occupied 61 per cent of the Yensiso area, compared to Amanase's approximately 43 per cent, which are indicative of greater pressure on land in the Yensiso area. A significant source of the pressure in Yensiso was the 406 ha cleared for the state-owned oil-palm plantation located at Kwamoso. The proportion of land under remnant forest with closed canopy was insignificant in both Yensiso and Amanase.

Table 7.1 1974 Aerial Photographic Estimates of Land Use and Cover Areas for Yensiso and Amanasea

Use and cover Yensiso Amanase
ha % ha %
Broken canopy forests and old mature fallows 1,683 38.6 2,490 56.6
Current arable cultivation and short fallows 2,653b 60.9 1,899 43.2
Remnant forest with closed canopy 20 0.5 8 0.2
Total 4,356 100.0 4,397 100.0

Source: 1974 aerial photographs supplied by the Survey Department, Accra.
a. Sekesua is omitted because the land use and cover types identified in the aerial photographs of the two other areas (Yensiso and Amanase) could not be clearly distinguished in the Sekesua aerial photographs.
b. Includes 406.1 ha (9.3%) for the state-owned oil-palm plantation of Kwamoso.

The Sekesua area, settled by migrant Krobo people on the basis of the linear huza agricultural land use system, showed near complete removal of the forest cover for cropping. Huza strip orientations, unlike the cultivated plots, were identified on the aerial photographs (fig. 6.3, p. 46). The Sekesua aerial photographs indicated a restriction of the limited dense bush or forest to narrow valleys, steep slopes and other localities of difficult access. The topography here is rather broken.

Ground Traverses

The ground traverse surveys show the disappearance of forest cover and the extensive conversion to agricultural land use. Two major land use and cover types, namely cropped lands and fallows, were observed. Others of a lesser coverage comprised settlements and roads. The traverse survey results are shown in table 7.2. An average of 54.4 per cent of the land in the three study sites was in active crop cultivation. The percentage distributions of the actively farmed lands ranged from 41.7 per cent in Sekesua to as high as 63.2 per cent in Amanase. The farms were generally intercropped with cassava and maize as the principal crops. Relict cocoa farms occurred under broken canopy tree cover. Several fields in Kokormu, Gyeabor and Yensiso and other localities in the Yensiso area were reported to have been cropped almost continuously since about 1950, when Ayigbe and Ewe migrants reportedly started renting land from the Akuapem owners for food crops. The continuous cropping had led to noticeable soil impoverishment and decreased crop yields in spite of the use of artificial chemical fertilizer and improved crop varieties. Home gardens fertilized with household refuse constituted a fairly significant proportion of the actively cultivated land, particularly within 500 m of the villages. The proportion of the total area covered by fallows was 34.3 per cent in Yensiso, 34.8 per cent in Amanase, and 41.7 per cent in Sekesua, with the average fallow period ranging from one to two years. The fallows were generally invaded by the weed Chromolaena odorata, popularly called akyeampong. We estimate the area in fallow, farms and other uses to cover over 90 per cent of the total area, which shows a remarkable correspondence with a 1990 estimate for the Mampong Valley Social Laboratory area, including Yensiso and the adjoining study sites (Gyasi et al. 1990).

Table 7.2 Distribution of Major Land Use/Cover along Traverses in Yensiso, Amanase and Sekesua in Metres and Percentages

Land use/cover Yensiso Amanase Sekesua Average of the percentages
  Length % Length % Length %  
Farm 1,056 58.3 1,122.5 63.2 712.5 41.7 54.4
Fallow 621 34.3 617.5 34.8 712.5 41.7 36.9
Other uses 133 7.4 35.0 2.0 285.0 16.6 8.7
Total 1,810 100.0 1,775.0 100.0 1,710.0 100.0 100.0

Source: October 1993 ground traverse survey along selected paths/roads from the study sites.

Table 7.3 Percentage Distribution of Land Use/Cover in Relation to Walking Distance along Traverse Paths from the Settlements

Land use/
cover
0 500 1,000 1,500
Yensi- so Ama- nase Seke- sue Yensi- so Ama- nase Seke- sue Yensi- so Ama- nase Seke- sue Yensi- so Ama- nase Seke-
sue
Farm 12 3 20.8 13.3 20.7 21.9 11.3 13.4 10.5 12.3 12.9 10.4 6.0
Fallow 12.2 10.4 2.4 10.0 3.4 15.2 9.8 17.6 15.7 2.3 3.4 5.1
Other 3.5 0.8 15.7 0 0 0 1.7 0.8 0 1.2 0 0
Total 28.0 32.0 31.4 30.7 25.3 29.5 24.9 28.9 28.0 16.4 13.8 11.1

Source: PLEC 1993 ground traverse survey.

In line with findings of previous studies (Chisholm 1962; Gyasi 1979; Gyasi et al. 1990), there appeared to be some relationship between land use/cover and distance from the settlements (table 7.3). The active farms tended to concentrate within the first 1,000 m (1 km) from the settlements, especially in Amanase, and the proportions of fallow and other uses showed rather irregular variation along the traverses.

Evidence of the original forest environment were small random pockets of remnant forest and more extensive areas of broken canopy forest. Wide areas at Yensiso and Amanase showed complete change from forest through perennial cocoa crop to seasonal arable crops. The change appeared even more dramatic in Sekesua, where there were only small pockets of tree cover limited to the areas of difficult accessibility. It is evident then that farming and other land uses have displaced very nearly all the original natural forest, and threaten to eliminate the rest.

Summary

The Yensiso, Sekesua and Amanase study areas were covered largely by dry, semideciduous forest until about the beginning of the present century. All three areas have experienced profound land cover changes as a result of land use. The forest cover is nearly completely eliminated. Traverse estimates show that about 50 per cent of the land in these areas is in active cultivation for arable and tree crops, while between 30 per cent and 40 per cent is in fallow. Land cover conversion is complete in the areas of bush fallow, where grasses and other weed invasion is complete. The trend of cover change and conversion seems to have proceeded from the original closed canopy high forest through the progressive opening of the canopy, elimination of forest, food cropping and the invasion of grasses and fortes. These processes have operated over the past 100 years. The data obtained from archive, aerial photographs and ground traverses for land use and cover are not precisely comparable. Yet the general indication is that the land usage and the natural vegetative cover loss have accelerated owing to increased socio-economic pressures, especially since the 1970s, with critical implications for the productive capacity of the biophysical environment.

References

Brammer, H. 1962. Soils. In: J.B. Wills, ea., Agriculture and Land Use in Ghana. London: Oxford University Press.

Chisholm, M. 1962. Rural Settlement and Land Use: An Essay in Location. London: Hutchinson University Library.

Gyasi, E.A. 1979. Distance as a factor in agricultural land use: case study of farming in Sekesua-Agbelitsom, Ghana. Land Administration Journal 1(2): 20-36.

Gyasi, E.A., Amaning-Kwarteng, K. and Oware-Gyekye, L. 1990. Mampong valley agroforestry baseline and evaluation survey report for the Ghana Rural Reconstruction Movement, University of Ghana, Legon. Unpublished.

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