Land use/land cover dynamics using support vector machine in the area of Lambaréné, Gabon, from 1988 to 2022

Abstract

Land use dynamics depict territorial changes, which include conversions and adjustments to landscape units caused by natural and human-induced processes. Human pressures impact landscape changes and various degradation in Lambaréné, which was established as a fully-fledged commune in 1963. The present study aimed to assess several Land Satellite (LANDSAT) pictures from 1988, 2000, 2013, and 2022 was utilized to map land use. The study generated data on land use changes in Lambaréné by cross-referencing these multiple maps. The supervised technique was utilized, and the Support Vector Machine algorithm (SVM) was used to study land-use changes over the last three decades. The resulting transition matrices were used to study the spatial and temporal dynamics of built-up areas, vegetation, and water bodies. The findings showed that urban areas had grown significantly, from 204.73 hectares in 1988 to 736.54 hectares in 2022, a 14.22 % rise, while vegetation dropped from 4057.40 hectares to 3488.86 hectares, a 67.36 % loss during the same time. This trend emphasized the disturbing influence of development on the region's already fragile ecosystems. Nonetheless, there was a slight recovery in the area's vegetation cover between 2013 and 2022, which was most likely due to the area's vulnerability to flooding disasters and thus low investment in infrastructural development during this time, particularly on the left bank of River Lambaréné compared to the right bank. Although these findings looked noteworthy, the use of higher-resolution images might be better for clearly understanding the complexity of land use change in Lambaréné.