Search Results (64)

Understanding the factors controlling tropospheric nitrogen dioxide (NO₂) variability is essential for improving air-quality assessment and environmental management in tropical regions. This study analyzed the spatial and interannual variability of tropospheric NO₂ in ten municipalities of Mato Grosso do Sul, Brazil, located within the Cerrado–Pantanal transition zone, during the period 2020–2024. Tropospheric NO₂ column densities were obtained from Sentinel-5P/TROPOMI observations and integrated with environmental and anthropogenic indicators, including fire density derived from the Brazilian National Institute for Space Research (INPE), land-use and land-cover data from MapBiomas, road density, and meteorological variables obtained from CEMTEC-MS. Descriptive statistics, Pearson correlation analysis, and multiple linear regression were applied to evaluate the relationships between NO₂ concentrations and the explanatory variables. The results revealed moderate spatial variability of tropospheric NO₂, with annual mean column densities ranging from 1.42 × 10⁻⁵ to 1.74 × 10⁻⁵ mol·m⁻². Higher concentrations were observed in municipalities characterized by greater urbanization and transport infrastructure, particularly Três Lagoas, Corumbá, and Ladário. Pasture area exhibited the strongest negative association with NO₂ concentrations (r = −0.81, p = 0.004), followed by agricultural area (r = −0.67, p = 0.034), whereas fire density showed a moderate positive relationship with NO₂ variability (r = 0.62, p = 0.056), highlighting the contribution of biomass burning to regional atmospheric pollution. Meteorological variables, especially precipitation and wind speed, also influenced NO₂ distribution through atmospheric removal and dispersion processes. These findings demonstrate that tropospheric NO₂ variability in Mato Grosso do Sul is controlled by the combined effects of land use, biomass burning, urban infrastructure, and meteorological conditions. The study provides new insights into the environmental drivers of atmospheric pollution in the Cerrado–Pantanal transition region and contributes to the development of monitoring and air-quality management strategies in tropical environments. Full article

Land use and land cover (LULC) changes are major drivers of environmental transformation in sensitive regions, such as the Marajó Archipelago in the Brazilian Amazon. This study assessed accumulated anthropization of LULC in the Immediate Geographic Region of Soure–Salvaterra, Eastern Amazon, between the reference years 1985 and 2023, using MapBiomas data and spatial statistical techniques. Bivariate Local Moran’s I (LISA) was applied to evaluate intertemporal spatial associations between areas classified as natural in 1985 and anthropized in 2023. In this approach, High–Low indicates natural areas associated with low anthropization in 2023, whereas High–High indicates areas where natural cover in 1985 was spatially associated with higher anthropization in 2023. The results indicated a strong predominance of High–Low, with values above 94% in all municipalities and up to 99.86% in Santa Cruz do Arari. In contrast, High–High had localized concentrations in Salvaterra (3.53%), Cachoeira do Arari (1.28%), and Soure (1.16%), especially in coastal zones and inland sectors. Low–Low, associated with lower anthropogenic pressure or possible signs of natural regeneration, was extremely low (≤0.0004%). These findings indicate that LISA is useful for identifying local LULC patterns and supporting environmental assessment and territorial planning in tropical regions. Full article

Wildfires have intensified in several regions worldwide, and the Brazilian Pantanal has become increasingly vulnerable due to the combined effects of human activities and climate change. This study analyzed the spatiotemporal patterns of burned areas and burn severity in the Brazilian Pantanal over 39 years (1985–2023), integrating burned-area dynamics, land use and land cover information, and hydroclimatic variables. Burned areas were quantified using MapBiomas Fire Project data, including annual burned areas, affected land use and land cover classes, seasonal fire distribution, fire-scar size, and fire recurrence. Burn severity was assessed using the Differenced Normalized Burn Ratio (ΔNBR), and hydroclimatic trends were evaluated using the Mann–Kendall test. The largest burned areas occurred in 1999 (27,260.65 km²) and 2020 (25,602.65 km²), with grassland representing the most affected land use and land cover class throughout the historical series. Fires were concentrated during the late dry season, and recurrent burning was more evident in the southwestern Pantanal and in smaller northern areas. The 2020 fire season showed the greatest extent of high-, moderate–high-, and moderate–low-severity classes. Wildfire occurrence, recurrence, extent, and severity were associated with hydroclimatic variability, especially reduced precipitation and relative humidity and increased air and land surface temperatures. These findings provide a long-term basis for understanding changes in fire regimes in the Brazilian Pantanal and can support fire management, ecological restoration, biodiversity conservation, and climate adaptation strategies. Full article

Carbon has increasingly been incorporated into economic and financial architectures as a tradable commodity within contemporary climate governance. Yet, carbon is not produced, stored, or mobilized in abstract space; it emerges from territorially specific land-use systems, ecological processes, and socio-spatial trajectories. This study examines carbon as a territorial commodity by analyzing long-term land-use and land-cover (LULC) dynamics in the municipality of Alegrete, located in the Brazilian Pampa biome, between 1985 and 2024. Based on MapBiomas Collection 10, and using cloud-based processing in Google Earth Engine combined with reproducible statistical workflows in R, the analysis identifies structural land-use trajectories shaping carbon-relevant territorial conditions. Results reveal a strong contraction of native grasslands, corresponding to approximately 17.8% of the municipal territory and a 24.2% reduction relative to the 1985 grassland area, alongside the expansion of mechanized agriculture, particularly soybean cultivation (+10.8% of the territory; +1343% relative to 1985 soybean area), and the consolidation of flooded rice systems (+7.6% of the territory; +146% relative to 1985 rice area). Rather than estimating carbon stocks or fluxes, the study establishes a territorial baseline linking land-use trajectories to key carbon-relevant processes, including soil carbon stability, disturbance intensity, permanence constraints, and multi-gas trade-offs. From a historical–structural perspective, these trajectories contrast with prevailing policy narratives and market-based instruments that assume an expanding carbon sequestration capacity, revealing a governance gap between valuation mechanisms and land-use realities. By conceptualizing carbon as a territorially embedded economic asset linked to land-use trajectories, the article contributes to interdisciplinary debates on climate governance, MRV integrity, environmental valuation, and the structural limits of market-based environmental instruments. Full article

Mapping wildfire patterns in Wildland–Urban Interface (WUI) areas is a fundamental tool for fire management and prevention, particularly in regions where urban expansion occurs in close proximity to natural vegetation. This mapping approach makes it possible to identify critical zones and to support more effective interventions adapted to the specific conditions of each territory. This work analyzed wildfires in the state of Tocantins, Brazil, using detailed geospatial data and advanced analysis techniques and statistics to characterize the dynamics of burned areas. Data used for the project were retrieved from MapBiomas and the Geoprocessing Laboratory of the Public Ministry of Tocantins (LABGEO), applying logistic regression models to explore the relationship between the distance of WUIs and the frequency of wildfires. The methodology covered the spatial distribution of fires and the different dynamics observed by type and size of burned area, allowing for a more detailed analysis. The results indicated significant variations in the proportion of burned areas inside and outside the WUIs, suggesting that proximity to these interfaces plays a critical role in the occurrence pattern of fires. Notably, Palmas, the state capital, stood out as one of the municipalities with the highest concentration of impacts in WUI areas, highlighting the relevance of these zones in environmental risk management. The study emphasizes the importance of adopting regional approaches that consider local specificities in the management and prevention of wildfires. The integration of geospatial data with robust statistical methodologies can guide more effective management strategies, assisting in the planning of public policies adapted to the socio-environmental dynamics of Tocantins. Full article

Context: Accurate citrus crop mapping is essential for agricultural monitoring, production planning, and supply-chain management, particularly in Brazil, one of the world’s leading orange producers and the leading orange-juice exporter. Satellite image time series from Sentinel-2 provide rich spectral and temporal information for crop identification. However, citrus mapping remains challenging due to fragmented agricultural landscapes, cloud contamination, class imbalance, and spectral overlap with other vegetation classes. Problem: Conventional machine learning models often depend on handcrafted vegetation indices, while attention-based deep learning models may require larger datasets and can become unstable under geographically constrained conditions. Therefore, there is a need for a compact and robust deep learning architecture capable of extracting citrus phenological signatures directly from multispectral time-series data. Methods: This study evaluates a Spatio-Temporal Pixel-Set Encoder Convolutional Neural Network (PSE-CNN) for citrus crop classification in the immediate geographic regions of São João da Boa Vista and Mogi Guaçu, São Paulo, Brazil. MapBiomas Collection 10.1 data from 2019 to 2024 were used to derive reference polygons, and Sentinel-2 imagery was processed into cloud-masked, 15-day temporal composites using ten spectral bands. The proposed PSE-CNN was benchmarked against PSE-TAE, PSE-Transformer, Random Forest, and XGBoost using spatially grouped data partitioning and temporal test years. Results: The proposed PSE-CNN achieved the highest Unified F1-Score of $0.704$ and the lowest coefficient of variation of $3.03\%$, indicating stronger inter-annual stability across test years and random seeds among the evaluated models. It also outperformed classical models that relied on handcrafted vegetation indices and demonstrated greater overall stability than attention-based deep learning alternatives. Conclusions: The results indicate that combining pixel-set encoding with temporal convolution provides a resource-aware and stable framework for retrospective citrus crop mapping from Sentinel-2 satellite image time series. These findings suggest that PSE-CNN can support scalable agricultural monitoring, contributing to sustainable crop inventory systems in regions where labeled data and computational infrastructure are limited. Full article

Land use and land cover change (LULCC) is increasingly recognized as a dominant driver of hydrological alteration in tropical watersheds, often exceeding the influence of climatic variability. This study evaluates the spatiotemporal dynamics of LULCC and their implications for hydrological sustainability in the Uberabinha River Basin, southeastern Brazil, between 1990 and 2020. Utilizing MapBiomas data and statistical analysis, the results reveal a marked expansion of mechanized agriculture, particularly soybean cultivation, which grew from 3426 ha to 54,162 ha, and urban areas, which expanded by approximately 89.4%. Conversely, natural vegetation and pasturelands decreased continuously, with pastures showing the sharpest absolute reduction, from 72,248 ha to 34,535 ha. Despite a 10.76% increase in annual precipitation between 1990 and 2020, the hydrological response exhibited a severe decline in streamflow, characterized by a 76.35% drop in minimum flow. Furthermore, the runoff index decreased from 0.0574 in 1990 to 0.0211 in 2020, indicating a critical loss in the basin’s capacity to convert rainfall into streamflow. These findings demonstrate a clear decoupling between precipitation and streamflow driven by LULCC, posing a severe threat to regional water security and highlighting the urgent need for integrated land–water management. Full article

This study examines long-term mangrove vegetation dynamics in the Cananéia–Iguape Coastal System (CICS), southeastern Brazil, with emphasis on their relevance for coastal ecosystem monitoring and sustainability. Land-use and land-cover (LULC) data from MapBiomas were combined with MODIS-derived vegetation indices, namely the Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI), and Fractional Vegetation Cover (FVC) to assess spatial variability and temporal trends from 2003 to 2024. Spatial anomalies were calculated as deviations from long-term mean conditions, whereas temporal trajectories were evaluated using the non-parametric Mann–Kendall test and Sen’s slope estimator. The results indicate limited spatial variability, with 98.35% of the study area for the NDVI and 99.51% for the EVI showing no detectable deviations from long-term averages. Within mangrove areas, however, statistically significant positive trends were identified for the NDVI (Z_MK = 2.43; p = 0.02), EVI (Z_MK = 2.03; p = 0.04), and FVC (Z_MK = 2.43; p = 0.02), suggesting a gradual increase in spectral greenness and FVC-derived vegetation density. The moderate correlation between mangrove extent and the NDVI (r = 0.61; p < 0.05) indicates that the mapped mangrove area is partially associated with variations in spectral greenness, although this relationship should not be interpreted as direct evidence of ecological recovery or improved ecosystem conditions. Overall, the findings point to low-magnitude but consistent vegetation index changes in a predominantly stable mangrove system. The integration of LULC information, spectral indices, and FVC provides a consistent regional-scale basis for interpreting mangrove dynamics in heterogeneous coastal environments and for guiding long-term monitoring efforts. Full article

This study evaluated the spatial and temporal variability of rainfall erosivity (R factor) and its implications for soil loss in the Velhas River Basin, Minas Gerais, Brazil. Rainfall erosivity was estimated from 49 rain gauge stations and CHIRPS precipitation data using empirical equations-based on monthly and annual precipitation totals. Soil loss was estimated using the RUSLE model for the years of minimum and maximum erosivity. Between 2014 and 2024, annual R values ranged from approximately 3900 to more than 9000 MJ mm ha⁻¹ h⁻¹ yr⁻¹, with the lowest values recorded in 2014 and the highest in 2022. Although 2020 had the highest annual rainfall, 2022 showed the highest erosivity, indicating that rainfall intensity and temporal concentration were more important than total rainfall volume. Furthermore, the comparison of erosivity was estimated from ANA stations and derived from CHIRPS agreement for paired station-year observations (r = 0.7196), although CHIRPS slightly underestimated erosivity values (mean bias −5.74%). Estimated soil loss ranged from 0.60 to 274.17 Mg ha⁻¹ yr⁻¹, with the highest values occurring mainly in exposed soil and agricultural areas. These findings highlight the importance of rainfall temporal distribution in erosion risk and support the use of satellite-derived precipitation products for regional-scale erosion assessments in data-scarce tropical basins. Full article

Socio-environmental conflicts in mining regions are often examined through political, economic, or social lenses, while the role of land-system transformation remains less integrated into quantitative analysis. This study examines the co-evolution of socio-environmental conflict and territorial change in Las Bambas (Apurímac, Peru) as a socio-territorial process. Annual conflict records from the Peruvian Ombudsman’s Office (2007–2024) were combined with annual land-cover data from MapBiomas. Yearly conflict influence zones were reconstructed from reported affected communities and geographic features using buffered spatial entities and concave hull polygons. Clustering methods (K-medoids, DBSCAN, and agglomerative hierarchical clustering) and FP-Growth association rule mining were applied to 23 unique conflicts consolidated from the original records and encoded with 10 root causes. The most intense conflict phases were accompanied by measurable landscape transformations, including the emergence of mining-related land cover from 2012 onward, sustained loss of high-Andean natural vegetation, expansion of agricultural mosaics, urban growth along the Apurímac–Cusco corridor, and hydrological alterations in wetlands and headwaters. Three conflict typologies were identified, with unfulfilled company commitments emerging as the most recurrent co-occurring grievance. The dynamic polygon approach offers a replicable framework for linking conflict records with land-system change in extractive regions. Full article

High spatio-temporal resolution satellite imagery has become increasingly accessible thanks to advancements in the aerospace industry which, combined with a growing computational power, has enabled the spring of novel techniques regarding recognition in remote sensing (RS) images. However, there is still a lack of studies around 3D convolutions for spatio-temporal data applied to classification problems in RS. Hence, this study investigates the feasibility of 3D convolutional neural networks (3DCNNs) within a spatio-temporal perspective for scene classification and semantic segmentation in RS images, focusing on the identification of mining sites. We firstly developed a dataset covering several parts of Brazil based on MapBiomas products and Planet imagery, then we evaluated the effectiveness of 3DCNNs in capturing temporal information from a sequence of monthly captured images. Moreover, not only for scene classification but also for semantic segmentation, we compared 3D and 2D approaches. As for scene classification, a 3DCNN was better than the corresponding 2D model, while a 2D U-Net was better than a U-Net3D for semantic segmentation. The main explanation for this lies in the fact that a less costly annotation and training time strategy was adopted, but this may have harmed spatio-temporal approaches for semantic segmentation but not for scene classification. However, U-Net3D presented the highest Precision of all models, meaning that it is highly accurate when it predicts a positive. Moreover, 3DCNN (U-Net3D) presented significantly better performance with respect to semantic segmentation compared to other spatio-temporal approaches like ConvLSTM+U-Net and TempCNN. Sensitivity analysis revealed that the near-infrared (NIR) band played a decisive role in distinguishing mining areas, emphasizing its importance in highlighting subtle spectral variations associated with land-cover disturbances. Full article

Seasonal tropical wetlands such as the Brazilian Pantanal are increasingly threatened by climate variability and extreme hydrological events, creating a need for robust monitoring tools that capture flood dynamics at high spatial and temporal resolution. This study used Sentinel-1 Synthetic Aperture Radar (SAR) imagery to map and monitor flooding in the northern Pantanal, a Ramsar site renowned for its wildlife, between 2017 and 2020. Ground Range Detected (GRD) VV-polarized scenes were preprocessed using radiometric terrain normalization and speckle filtering (Lee filter, 5 × 5 window) to improve the separability of water and non-water surfaces. Flooded areas were initially extracted with Otsu’s histogram thresholding and validated using high-resolution optical imagery (PlanetScope and Landsat-8). A supervised Random Forest classifier then refined land-cover discrimination into three classes (open water/flood, open land/vegetation, and others), achieving an overall accuracy of 97.70% on the independent testing dataset (n = 6622), while temporal consistency was supported by Cuiabá River hydrological data. The results revealed strong interannual variability in flood extent, with inundation covering 34.7% of the reserve in March 2017 compared with 0.75% in March 2020 and reaching a peak of 79.9% in April 2017. Overall, Sentinel-1 SAR effectively delineated open water and flood-affected surfaces under persistent cloud cover, demonstrating its value for complementing existing products such as MapBiomas, strengthening wetland management, and supporting scalable flood monitoring in other tropical flood-prone Ramsar sites. Full article

Recent large-scale renewable energy projects, such as the Luzia Solar and Chafariz Wind energy plants in Santa Luzia, Paraíba, Brazil, raised environmental concerns due to their impact on vegetation cover and landscape structure. This study used geospatial technologies to evaluate changes in tree cover and landscape configuration resulting from the installation of these projects. Sentinel-2 imagery processed in Google Earth Engine generated NDVI, SAVI, NDWIveg, and LAI vegetation index data for the dry and rainy seasons of the six years between 2019 and 2024. With these vegetation index values and considering MapBiomas (version 8.0) and FRAGSTATS software (version 4.2), we analyzed the changes in land use and vegetation cover of Santa Luzia municipality during this six-year period. Land use and vegetation cover remained stable from 2019 to 2020 (before the installation of the energy plants), characterized by an NDVI value of 0.60, while tree cover decreased in the following four years, during or after the installation of the energy plants, as indicated by the consistent decreases in NDVI and NDWIveg values. Grassland class areas declined from 41.80% (18,434.59 ha) in 2019, to 34.36% (15,151.22 ha) in 2023, while non-vegetated areas increased by 148%. Landscape metrics showed increased fragmentation, with patch density rising from 3.31 to 3.88 patches/100 ha and core area decreasing from 3045.60 ha to 1395.01 ha. These data demonstrated measurable ecological impacts linked to the infra-structure built to run the two solar and wind energy plants in the semi-arid region of Santa Luzia, Paraíba, Brazil. Full article

Roads impact medium- and large-sized mammal populations through both collisions and barrier effects. This study examined how landscape characteristics influence roadkill occurrences along the Dom Pedro I highway (SP-065), located in the Cantareira-Mantiqueira Corridor, São Paulo State, Brazil. The SP-065 crosses important remnants of the Brazilian Atlantic Forest, a global hotspot for biodiversity. Roadkill records were obtained from the Environmental Company of the State, and land use data were extracted from the MapBiomas platform. We analyzed seven landscape variables (in percentage): native forest, pasture, agriculture, forestry, urban areas, mosaic of uses, and water bodies, considering multiple spatial scales. Mammal species were grouped functionally by home range size and tolerance to anthropogenic environments. In total, 1418 roadkills of 24 species were recorded, including eight threatened species. Capybaras (Hydrochoerus hydrochaeris) were the most frequently killed species. Generalized linear models showed that, for Group G1 (small home range, common in human-modified areas), roadkills were positively associated with native forest and pasture, and negatively with mosaic landscapes. For Group G3 (large home range, tolerant to anthropogenic areas), agriculture had a positive effect, especially within a 3000 m radius. For Group G5 (capybara), roadkills increased with pasture and agriculture, while mosaic uses had a negative effect. Since pasture and agriculture were frequently linked to higher roadkill rates, environmental impact assessments should consider these land-use types when planning mitigation actions. Ultimately, responsibility for roadkill extends beyond highway managers to rural landowners and local governments, as land-use patterns around roads strongly influence mammal movement and mortality. Full article

This study investigates the influence of land use and land cover (LULC) on the distribution of extreme rainfall in the tropical coastal city of Natal, Brazil. Hourly precipitation data from eight automatic rain gauges (2014–2023) were quality-controlled, with only days containing 24 h continuous records retained. Rainfall events were classified into light (<5 mm), normal (5–10 mm), intense (40–50 mm), and extreme (>50 mm) categories, and for each category daily accumulation, duration, intensity, and maximum hourly peaks were calculated. Seasonal and spatial differences across administrative zones were assessed using multivariate analysis of variance (MANOVA). The LULC changes were evaluated from the MapBiomas Collection 9 dataset. Results show that between 1985 and 2020, the proportion of urbanized (non-vegetated) surfaces increased from 27.7% (42.3 km²) to 64.3% (99.7 km²), mainly in the North and West zones, replacing agricultural and vegetated areas. The East and North zones, the most urbanized areas, recorded higher daily averages of extreme rainfall in the dry season (85–88 mm) than in the wet season (78–82 mm), with maximum peaks up to 26 mm/h and durations exceeding 17 h. These findings demonstrate that rapid urban expansion intensifies rainfall extremes, underscoring the importance of incorporating LULC monitoring (e.g., MapBiomas) and spatial planning into climate adaptation strategies for medium-sized cities. Full article