The tropical savanna in Brazil known as the Cerrado covers circa 23% of the Brazilian territory, but only 3% of this area is protected. High rates of deforestation and degradation in the woodland and forest areas have made the Cerrado the second-largest source of carbon emissions in Brazil. However, data on these emissions are highly uncertain because of the spatial and temporal variability of the aboveground biomass (AGB) in this biome. Remote-sensing data combined with local vegetation inventories provide the means to quantify the AGB at large scales. Here, we quantify the spatial distribution of woody AGB in the Rio Vermelho watershed, located in the centre of the Cerrado, at a high spatial resolution of 30 metres, with a random forest (RF) machine-learning approach. We produced the first high-resolution map of the AGB for a region in the Brazilian Cerrado using a combination of vegetation inventory plots, airborne light detection and ranging (LiDAR) data, and multispectral and radar satellite images (Landsat 8 and ALOS-2/PALSAR-2). A combination of random forest (RF) models and jackknife analyses enabled us to select the best remote-sensing variables to quantify the AGB on a large scale. Overall, the relationship between the ground data from vegetation inventories and remote-sensing variables was strong (R2 = 0.89), with a root-mean-square error (RMSE) of 7.58 Mg ha−1 and a bias of 0.43 Mg ha−1.
Anthropogenic actions influence landscapes, and the resulting mosaic is a mix of natural and anthropogenic elements that vary in size, shape, and pattern. Considering this, our study aimed to analyse the land use and land cover changes in the Tietê–Jacaré watershed (São Paulo state, Brazil), using the random forest (RF) algorithm and Sentinel-2 satellite data from 2016 to 2018 to detect landscape changes. By overlapping the environmental data and the proposed model evaluation, it was possible to observe the landscape structure, produce information about the state of this region, and assess the environmental responses to anthropic impacts. The land use and land cover analysis identified eight classes: exposed soil, citriculture, pasture, silviculture, sugar cane, urban area, vegetation, and water. The RF classification for the three years reached high accuracy with a kappa index of 0.87 in 2016, 0.85 in 2017, and 0.85 in 2018. The model developed was essential for the temporal analysis since it allowed us to comprehend the driving forces that act in this landscape and contribute to the discussions about their impacts over time. The results showed a predominance of agricultural activities over the three years, with approximately 900.000 ha (76% of the area), mainly covered by sugarcane cultivation.