Evolução espaço-temporal do uso e cobertura da terra no município de Barão de Cotegipe (RS)

Abstract

Human-induced transformations on the Earth's surface, particularly land use and land cover changes, directly influence environmental dynamics. This study analyzed the temporal evolution of land cover in the municipality of Barão de Cotegipe (RS), Brazil, between 2003 and 2023, focusing on implications for environmental planning. The methodology was based on digital processing of Landsat 7 and 8 imagery using the Semi-Automatic Classification Plugin (SCP) in QGIS. Machine learning algorithms Random Forest (RF) and Support Vector Machine (SVM) were applied for supervised classification, with thematic classes defined as Forest, Agriculture, Bare Soil, and Water. Accuracy assessment indicated high reliability of the RF classifier, with overall accuracy above 98% and Kappa coefficient around 0.97. The results reveal a significant landscape transformation over two decades, marked by the expansion of agricultural areas and the continuous reduction of forest cover, emphasizing the need for management and planning strategies aimed at sustainable land use.