Gilberto Ribeiro de Queiroz

Selective logging often marks the beginning of forest degradation, but remote sensing alert systems typically detect it long after it starts. This type of disturbance produces a subtle canopy signal not easily recognised in the early stages, causing alerts in some areas of the Brazilian Legal Amazon to lag behind the first visible signs by several months. This study explores whether a geospatial foundation model can help reduce this time gap. For each documented disturbance site, we compile Harmonized Landsat and Sentinel-2 (HLS) image time series spanning October 2024 to May 2026 into a datacube and use it to pretrain, without labels, a compact spatiotemporal masked autoencoder (ST-MAE). This model learns to embed the dynamics of forest degradation into latent-space representations, the so-called "embeddings". Following this, a lightweight downstream model — trained with few labelled samples — interprets ST-MAE's embeddings to evaluate every new HLS observation for recent logging activity. To effectively translate the evaluation scores into production-ready alerts, we apply a simple change-detection approach: an edge filter for sudden logging-level changes, paired with a persistence filter for sustained detections. We conducted a leave-one-out cross-validation assessment across 20 well-documented logging sites: yielding no false alarms prior to the first disturbance evidence, the change detector confirmed the onset of the 20 logging activities with a median delay of 15 days after the first post-disturbance image. The downstream model's scores remain low during the pre-disturbance period but increase once logging begins, confirming that it responds to logging-related events rather than to other landscape dynamics or calendar artefacts. This label-free pretraining enables effective performance with minimal annotations, providing a practical solution to reduce the temporal gap in monitoring forest degradation.

The Brazilian Cerrado, a global biodiversity hotspot, is under increasing pressure from agricultural expansion and native vegetation conversion, underscoring the need for efficient monitoring to support conservation and environmental policies. In heterogeneous landscapes, land use and land cover (LULC) mapping using supervised classification methods faces a major bottleneck: the need for extensive and high-quality training datasets. To address this challenge, we propose a semi-automated, clustering-based methodology for mapping secondary vegetation within previously deforested areas, reducing training-sample requirements and enabling scalable mapping through the clustering of satellite image time series. In the first stage, an unsupervised process integrates graphics processing unit (GPU)-accelerated Self-Organizing Maps and hierarchical clustering with Dynamic Time Warping to produce spectro-temporal clusters. In the second stage, specialists label and refine these clusters by visual interpretation, transferring expert knowledge from individual pixels to grouped spectro-temporal patterns. Applied to 692,000 km² of previously deforested land in the Cerrado biome, the methodology produced a mapped secondary vegetation area of 81,209 km² (11.74%). The design-based estimated area was 98,683 ± 10,071 km², with an overall accuracy of 96.45 ± 1.52%, a user’s accuracy of 96.27 ± 2.40%, a producer’s accuracy of 79.22 ± 7.94%, and an F1-score of 86.90%. The initial cluster labeling accounted for 86.3% of the final secondary vegetation area and limited the interpretation task to approximately 3000 cluster-level decisions. Implemented in the TerraClass Cerrado 2024 cycle, the workflow reduced the secondary vegetation mapping phase from approximately two years to six months while maintaining the thematic accuracy required for large-scale operational monitoring.

Just over a decade ago, the availability of automated solutions for accessing Earth observation (EO) data from imaging sensors, even in minimally preprocessed form, was not yet a reality. At that time, users typically had to consult image catalogs manually and apply a sequence of preprocessing techniques themselves. These efforts were further intensified when the applications required time series of images or the integration of data from multiple sensors. In addition to these labor-intensive tasks, users also needed to rely on local computational infrastructure capable of storing and processing large volumes of data. As a result, a significant amount of time was spent on dataset preparation, often delaying the actual analytical tasks of interest.

Reliable remote sensing-based Land Use and Land Cover (LULC) information is crucial for assessing Earth’s surface activities. Brazil’s agricultural dynamics, including year-round cropping, multiple cropping, and regional climate variability, make LULC monitoring a highly challenging task. The country has thirteen remote sensing-based LULC products specifically tailored for this purpose. However, the differences and the results of these products have not yet been synthesized to provide coherent guidance in assessing their spatio-temporal agricultural dynamics and identifying promising approaches and issues that affect LULC analysis. This review represents the first comprehensive assessment of the advantages, challenges, and limitations, highlighting the main issues when dealing with contrasting LULC maps. These challenges include incompatibility, a lack of updates, non-systematic classification ontologies, and insufficient data to monitor Brazilian LULC information. The consequences include impacts on intercropping estimation, diminished representation or misrepresentation of croplands; temporal discontinuity; an insufficient number of classes for subannual cropping evaluation; and reduced compatibility, comparability, and spectral separability. The study provides insights into the use of these products as primary input data for remote sensing-based applications. Moreover, it provides prospects for enhancing existing mapping efforts or developing new national-level initiatives to represent the spatio-temporal variation of Brazilian agriculture.

Water management is a key field to support life and economic activity nowadays. The greatly increased mechanization of agriculture, mainly through center pivot irrigation systems, represents a big challenge to control this resource. Irrigated agriculture makes up the large majority of consumptive water use, therefore it is important to identify and quantify these systems. Currently, with 6.95x10⁶ ha, Brazil is among the 10 largest countries in irrigation areas in the world. In this study, a combined Computer Vision and Machine Learning approach is proposed for the identification of center pivots in remote sensing images. The methodology is based on Circular Hough Transform (CHT) and Balanced Random Forest (BRF) classifier using vegetation indices NDVI/SAVI generated from Landsat 8 images and Land Use and Land Cover (LULC) data provided by project MapBiomas. The candidate's circles of pivots identified on images are filtered based on vegetation behavior and shape characteristics of these areas. Our approach was able to detect 7358 pivots, reaching 83.86% of Recall for 52 scenes analyzed overall Brazil compared with mapping done by the Brazilian National Water and Sanitation Agency (ANA). In some scenes, the Recall reaches up to 100%.

The potential to perform spatiotemporal analysis of the Earth's surface, fostered by a large amount of Earth Observation (EO) open data provided by space agencies, brings new perspectives to create innovative applications. Nevertheless, these big datasets pose some challenges regarding storage and analytical processing capabilities. The organization of these datasets as multidimensional data cubes represents the state-of-the-art in analysis-ready data regarding information extraction. EO data cubes can be defined as a set of time-series images associated with spatially aligned pixels along the temporal dimension. Some key technologies have been developed to take advantage of the data cube power. The Open Data Cube (ODC) framework and the Brazil Data Cube (BDC) platform provide capabilities to access and analyze EO data cubes. This paper introduces two new tools to facilitate the creation of land use and land over (LULC) maps using EO data cubes and Machine Learning techniques, and both built on top of ODC and BDC technologies. The first tool is a module that extends the ODC framework capabilities to lower the barriers to use Machine Learning (ML) algorithms with EO data. The second tool relies on integrating the R package named Satellite Image Time Series (sits) with ODC to enable the use of the data managed by the framework. Finally, water mask classification and LULC mapping applications are presented to demonstrate the processing capabilities of the tools.

Medindo o brasil a partir do espaço: tecnologias de big data e inteligência artificial

The development of analytical software for big Earth observation data faces several challenges. Designers need to balance between conflicting factors. Solutions that are efficient for specific hardware architectures can not be used in other environments. Packages that work on generic hardware and open standards will not have the same performance as dedicated solutions. Software that assumes that its users are computer programmers are flexible but may be difficult to learn for a wide audience. This paper describes sits, an open-source R package for satellite image time series analysis using machine learning. To allow experts to use satellite imagery to the fullest extent, sits adopts a time-first, space-later approach. It supports the complete cycle of data analysis for land classification. Its API provides a simple but powerful set of functions. The software works in different cloud computing environments. Satellite image time series are input to machine learning classifiers, and the results are post-processed using spatial smoothing. Since machine learning methods need accurate training data, sits includes methods for quality assessment of training samples. The software also provides methods for validation and accuracy measurement. The package thus comprises a production environment for big EO data analysis. We show that this approach produces high accuracy for land use and land cover maps through a case study in the Cerrado biome, one of the world’s fast moving agricultural frontiers for the year 2018.

In their comments about our paper, the authors remark on two issues regarding our results relating to the MACCS-ATCOR Joint Algorithm (MAJA). The first relates to the sub-optimal performance of this algorithm under the conditions of our tests, while the second corresponds to an error in our interpretation of MAJA’s bit mask. To answer the first issue, we acknowledge MAJA’s capacity to improve its performance as the number of images increases with time. However, in our paper, we used the images we had available at the time we wrote our paper. Regarding the second issue, we misread the MAJA’s bit mask and mistakenly labelled shadows as clouds. We regret our error and here we present the updated tables and images. We corrected our estimation and, consequently, there is an increment in MAJA’s accuracy in the detection of clouds and cloud shadows. However, these increments are not enough to change the conclusion of our original paper.