Learn more about the scientific articles prepared by the MapBiomas team related to land cover and land use mapping in Brazil.
Alencar et al. Long-Term Landsat-Based Monthly Burned Area Dataset for the Brazilian Biomes Using Deep Learning
The paper presents a new strategy using machine learning to map monthly burned areas from 1985 to 2020 using Landsat image mosaics and minimum NBR values. This new dataset contributes to the understanding of the long-term spatial and temporal dynamics of fire regimes that are fundamental to design appropriate public policies to reduce and control fires in Brazil.
Cayo et al. Mapping Three Decades of Changes in the Tropical Andean Glaciers Using Landsat Data Processed in the Earth Engine.
This paper presents the mapping and retreat dynamics of tropical Andean glaciers (TAGs) from the use of Landsat time series images from 1985 to 2020, with digital processing and classification of the satellite images on the Google Earth Engine platform.
Coelho-Junior et al - Unmasking the impunity of illegal deforestation in the Brazilian Amazon: a call for enforcement and accountability
This article provides a perspective on the dynamics of deforestation alerts, validated and refined by MapBiomas Alert (http://alerta.mapbiomas.org/), in the Brazilian Amazon and the actions of federal and state public enforcement agencies highlighting the urgency to reduce and combat deforestation.
Santos et al - Assessing the Wall-to-Wall Spatial and Qualitative Dynamics of the Brazilian Pasturelands 2010–2018, Based on the Analysis of the Landsat Data Archive
In this study, the spatio-temporal dynamics of pasture quality in Brazil between 2010 and 2018 were mapped and evaluated, considering three degradation classes: Absent (D0), Intermediate (D1) and Severe (D2). There was no variation in the total area occupied by pastures in the evaluated period, despite the accentuated spatial dynamics.
Cesar et al. - A Large-Scale Deep-Learning Approach for Multi-Temporal Aqua and Salt-Culture Mapping
Aquaculture and saliculture are relevant economic activities in the Brazilian Coastal Zone (BCZ). However, automatic discrimination of such activities from other water-related coverages / uses is not an easy task. In this sense, convolutional neural networks (CNN) have the advantage of predicting the class label of a given pixel, providing as input a local region (patches or named chips) around that pixel. Both the convolutional nature and the semantic segmentation capability provide the U-Net classifier with the ability to access the "context domain" instead of just isolated pixel values. Supported in the context domain, we present the results of the analyzes.