Modeling land-use change in the Volta Basin of Ghana (Volume 14) (English shop)

Preview

Table of Contents, Datei (190 KB)
Extract, Datei (200 KB)

The Volta Basin in West Africa drains about 400,000km2 area of land, including 70% of
mainland Ghana. Low rainfall reliability, high population growth rate and macroeconomic
transformations in the last few decades have had a profound influence on the livelihood
strategies of the largely rural populace. Therefore, many parts of the basin are hotspots of land-
use / land-cover change (LUCC). The determinants of LUCC in a 5,400km2 area within the
Volta Basin of Ghana were identified using multiscale, spatial statistical analyses and household
surveys. Land-cover change trajectories were defined using multitemporal Landsat TM images
acquired in 1984, 1992 and 1999. Training signatures for land-cover classification using
maximum likelihood algorithm were developed based on PCA, tasseled cap and NDVI
transforms, whereas ground truth data were obtained from aerial photo interpretation and field
surveys. Change detection was based on synergy between image-differencing and post
classification comparison techniques. Statistical relationships between land-cover and selected
proxy driving factors of change were determined using two modeling techniques. The first
technique answered the question: what are the factors that determine the conversion of a
location to cropland? It involved the use of logistic regression to determine the relationships
between cropland change and selected variables at different cell resolutions ranging from 30m
(household level) to 5100m (village level). The second technique using linear multiple
regression was used to answer the question: given the socioeconomic characteristics of
households within a locality, how much land will be cleared for agriculture for domestic and
commercial purposes?
The results of the study show that the dominant land-cover change process was conversion of
natural vegetation to cropland, which occurred at an annual rate of 5% between 1984 and 1999.
A higher increase in woody biomass in 10% and a simultaneous decrease in 9% of the landscape
indicates a certain level of rainfall-induced resilience in the savannah ecosystem. A study of the
association between land-cover change and soil properties indicated that correlation between
organic C, an index of agricultural sustainability and selected soil fertility indices – ECEC, N
and P – declined as cultivation persisted. There is evidence of nutrient mining in about 12% of
the land area continuously cultivated between 1984 and 1999.
Logistic regression results show the time and scale-dependency of LUCC patterns in the study
area. Between 1984 and 1992, the main drivers of change were altitude, distance from roads,
distance from the main market and localities, initial population density and change in population
density. In the second period (1992-1999), land suitability index, distance from villages and
localities, change in population density, and rainfall zone were the main drivers of cropland
change. Linear multiple regression identified increase in household size, frequency of tractor
use, proportion of rice marketed, child dependency ratio, labor availability, and distance from
localities to the main market as the major factors determining the amount of land a household
cleared for agriculture. This suggests the coexistent processes of agricultural intensification and
extensification in the land-use strategies of the populace.
The choice of an appropriate scale for LUCC models to support land-use planning requires a
trade-off between spatial detail and extent. At the scale of individuals, households and
commercial farmers (30m-1050m), land-use change processes are highly heterogeneous,
requiring a large amount of data for characterization. The variograms of standardized logit
residuals suggest that the size of land change processes at the village ranged from 3km to 7km.
Future land change models should therefore be based on these spatial scales.
Policy suggestions for sustainable land management include coercive environmental protection,
integration of livestock and crops in the production systems, research to improve the quality of
soil organic matter, and institutional arrangement to promote agricultural commercialization.