This study analyzed spatiotemporal changes in land cover and land use (LULC) in the Araranguá River Watershed (ARW), southern Santa Catarina, Brazil, between 2016 and 2023. Sentinel-2A satellite data were used, with RGB, NIR, and SWIR 1 bands selected, and EVI2, MNDWI, and NDBI indices applied to classify eight LULC classes. The data were processed using Google Earth Engine (GEE) and validated with platform-generated data. The overall accuracy was 93% for both years. The Native Forest class was the most representative, increasing by 1.62% over seven years. The Built Area class showed the highest growth, while the Pasture/Herbaceous Vegetation class decreased by 5.6%. The results revealed subtle landscape changes, maintaining native forest areas and urban expansion. These data can assist policymakers and decision-makers in managing the basin with a focus on conservation and natural resource preservation. Keywords: Environmental degradation; Machine learning; Decision trees.This study analyzed spatiotemporal changes in land cover and land use (LULC) in the Araranguá River Watershed (ARW), southern Santa Catarina, Brazil, between 2016 and 2023. Sentinel-2A satellite data were used, with RGB, NIR, and SWIR 1 bands selected, and EVI2, MNDWI, and NDBI indices applied to classify eight LULC classes. The data were processed using Google Earth Engine (GEE) and validated with platform-generated data. The overall accuracy was 93% for both years. The Native Forest class was the most representative, increasing by 1.62% over seven years. The Built Area class showed the highest growth, while the Pasture/Herbaceous Vegetation class decreased by 5.6%. The results revealed subtle landscape changes, maintaining native forest areas and urban expansion. These data can assist policymakers and decision-makers in managing the basin with a focus on conservation and natural resource preservation. Keywords: Environmental degradation; Machine learning; Decision trees.