The Changing Landscape of Dormaa Central Municipal Through Vegetation Losses: Evidence from Machine Learning and Remote Sensing Analysis
摘要
Vegetation assessment is paramount to the sustainability due to their local community’s over-reliance on these natural resources. AI-based tools such as Google Earth Engine (GEE) have proven to efficiently assess the sustainability of these resources through the analysis and modelling of changes in LULC systems. This study explored the potential of AI-based GEE’s Random Forest (RF) classifier to analyse the trends in LULC change for the past 36 years. Results indicate a rapid of forest vegetation loss at an annual rate of −0.79%, with a total decrease of 34,327.3 ha between 1987 and 2023. Cropland and settlement expanded by 24,022.2 and 1794.96 ha respectively during that period. The projections indicate forest vegetation loss of 3721.12 ha by 2032 and 4845.61 ha by 2047. Cropland and settlement are expected to increase by 524.9 and 1886.62 ha by 2032, and 3940.9 and 2742.43 ha by 2047 respectively. Cropland expansion remains a major threat, as most of the forest cover has been converted to agricultural lands. Projected LULC patterns suggest continued loss of forest to cropland and settlement. The study recommends strict enforcement of spatial plans promoting the conservation and sustainable management of the remaining forest vegetation within Dormaa Central Municipality.