Land-cover change monitoring in Obuasi, Ghana: an integration of earth observation, geoinformation systems and stochastic modelling

For over twenty years, Obuasi Municipality, Ghana, has experienced land-cover change arising from gold mining and urbanisation. This project quantified the land-cover changes that have taken place and projected likely future land-cover. An integration of Earth Observation (or EO), Geographical Information Science (or GIS) and Stochastic Modelling was examined. Post-Classification Change Detection employed Landsat TM or ETM+ images from 1986, 2002 and 2008. Subsequently, Markov Chain Analysis projected the land-cover distribution for 2020. Seven broad land-use and land-cover classes were identified and mapped, namely: built-up areas; mine sites; tailing ponds; barren land; forestland; farmland; and, rangeland. The results obtained for the 2008 to 2020 projection revealed a continuous expansion of built-up areas (1.63%), mine sites (0.89%) and farmland (3.4%), and a reduction of forestland (4.17%) and rangeland (2.59%). Despite the advent of very high resolution satellite imagery, this use of EO and GIS technology focussed on low-cost and lower resolution satellite imagery, coupled with Markov Modelling and was found to be beneficial in describing and analysing land-cover change processes in the study area, and was hence potentially useful for strategic planning purposes