Groundwater flow and transport of metals under deposits of mine tailings: A case study in Brumadinho, Minas Gerais, Brazil

Case Studies in Chemical and Environmental Engineering, Journal Year: 2024, Volume and Issue: 9, P. 100690 - 100690

Published: March 13, 2024

The deposits of mine tailings can be a source groundwater contamination by metals. In this study, we simulated the concentrations iron, manganese and aluminum in potentially affected drainage from located Brumadinho (Brazil). aim was to verify whether observed region attributed these drainages. simulation used FREEWAT graphical interface, which incorporates MODFLOW model, hydraulic properties existing unconfined confined aquifers, spatial distribution tailings' deposits, dissolved iron measured drilled wells. period 20 years, starting 2019 after collapse B1 dam Córrego do Feijão Vale, S.A. modeling results revealed plumes metal progressively less dispersed over time, aquifer, increased aquifer. both aquifers were generally lower than legal limits imposed for human consumption, although some areas vicinity had higher those limits, especially widened time. most relevant result revelation that contribution wells might have not exceeded 1%. This is important management standpoint, because monitoring anthropogenic cases (where rock weathering dominates chemistry) becomes more challenging.

Language: Английский

---

Siderite's green revolution: From tailings to an eco-friendly material for the green economy

The Science of The Total Environment, Journal Year: 2024, Volume and Issue: 914, P. 169922 - 169922

Published: Jan. 8, 2024

Language: Английский

---

Downstream Flood Inundation Assessment Due to Dam Breach of Dudhkoshi Storage Hydroelectric Project Using HEC‐RAS 2D

Advances in Civil Engineering, Journal Year: 2025, Volume and Issue: 2025(1)

Published: Jan. 1, 2025

Dam breach is rare but catastrophic event that occurs when a dam fails and releases impounded water downstream causing significant damage. has low probability of occurrence carries high risk destruction. Dudhkoshi Storage Hydroelectric Project (DKSHEP) concrete‐faced rock‐fill (CFRD) was studied for under overtopping piping failure modes. analysis, flood routing, inundation hazard mapping, sensitivity analysis (SA) parameters are essential in identifying minimizing risks associated with floods. Shallow equations (SWEs) were used to perform two‐dimensional (2D) using Hydrologic Engineering Center’s River Analysis System (HEC‐RAS). The model based on two scenarios: the base‐case scenario, which average value parameters, worst‐case values resulted maximum output. estimated according guidelines set by United States Army Corps Engineers (USACE), 2007. Local global analyses performed four namely width (DBW), formation time (BFT), weir coefficient (WC), trigger elevation (TFE) failure, while (PC) evaluated instead TFE failure. routing SA river profiles (R1 R2) separated Dudhkoshi–Sunkoshi confluence. peak discharge, velocity, arrival time, surface analyzed. DBW, BFT, WC, TFE, PC scenario 206 m, 0.5 h, 2.8, 640.3 0.55, 309 0.25 3.0, 640.5, 0.5, respectively. Aeronautical Reconnaissance Coverage Geographic Information (ArcGIS), HEC‐RAS, OriginPro 2022 analysis. Overtopping more critical than mode DKSHEP both scenarios.

Language: Английский

---

Remote sensing of hazards: The spatio-temporal evolution of land surface temperature over tailings flows and related drivers

Remote Sensing Applications Society and Environment, Journal Year: 2024, Volume and Issue: 35, P. 101237 - 101237

Published: May 9, 2024

Since mining activities began in Brazil some 300 years ago, disasters related to tailings dams have been recorded. In 2019, the B1 dam located at Vale, S.A.'s Córrego do Feijão mine municipality of Brumadinho (state Minas Gerais) collapsed, releasing a vast amount iron ore into Ferro-Carvão stream, which spread Paraopeba River 10 km downstream. This tragedy devastated around 294 hectares Ribeirão watershed, including infrastructure, farmland and housing, is among biggest environmental ever recorded Brazil. The torrent mud debris substances released by collapse, namely metals (e.g. iron, manganese), addition completely flooding bed banks remaining on them for several as thick blanket, caused significant changes land surface temperature (LST). aim this study was spatially analyze LST period before after burst (2018 - 2021), find its relationship with granulometric chemical parameters deposited materials, well use cover that occurred burst. Landsat 8 satellite images processed Google Earth Engine platform were used estimate LST. results showed between parameter characteristics tailings, lower temperatures being associated sectors impacted area covered fine fractions (presumably greater heat dissipation capacity). A reduction also detected over analyzed, removal revegetation works area. integrated indicate an effective characterizing involving dams, monitoring natural recovery or result mitigation measures implemented areas. We therefore recommend remote sensing general particular.

Language: Английский

---

Carbon footprints of tailings dams' disasters: A study in the Brumadinho region (Brazil)

The Science of The Total Environment, Journal Year: 2024, Volume and Issue: 949, P. 175026 - 175026

Published: Aug. 2, 2024

Tailings dams' breaks are environmental disasters with direct and intense degradation of soil. This study analyzed the impacts B1 tailings dam rupture occurred in Ribeirão Ferro-Carvão watershed (Brumadinho, Brazil) January 25, 2019. Soil organic carbon (SOC) approached degradation. The analysis encompassed wetlands (high-SOC pools) located so-called Zones Decreasing Destructive Capacity (DCZ5 to DCZ1) defined along Ferro-Carvão's stream bed banks after disaster. Remote sensed water indices were extracted from Landsat 8 Sentinel-2 satellite images spanning 2017-2021 period used distinguish other land covers. annual SOC was MapBiomas repository inside outside DCZs same period, assessed field 2023. Before collapse, maintained stable levels SOC, while afterwards they decreased substantially reaching minimum values reductions abrupt: for example, DCZ3 decrease 51.28 ton/ha 2017 4.19 Besides, increased near farther site, a result attributed differences percentages clay silt tailings, which also direction. as whole experienced slight reduction average nearly 43 38 2021. use changes related management namely opening accesses remove them valley, creation spaces temporary deposits, among others. Overall, highlighted footprints accidents on affect not only areas impacted mudflow but systemically surrounding watersheds. is noteworthy.

Language: Английский

---

The economic valuation of environmental damages in scenarios of tailings dams’ ruptures: The case of Brumadinho’s catastrophe, Minas Gerais, Brazil

Case Studies in Chemical and Environmental Engineering, Journal Year: 2024, Volume and Issue: unknown, P. 101037 - 101037

Published: Nov. 1, 2024

Language: Английский

---

Assessment of Forest Cover Dynamics in Dedo District, Oromia Region, Ethiopia, Using Machine Learning Algorithm

Journal of Sensors, Journal Year: 2024, Volume and Issue: 2024(1)

Published: Jan. 1, 2024

Forests, which are vital for ecological equilibrium, experiencing widespread dynamics due to multiple anthropogenic pressures. To grasp the of this process, accurate and spatially explicit information is urgently required. However, quantifying temporal forest cover Dedo Ethiopia has been lacking despite its significance surrounding ecology. This phenomenon places seriously threatened natural resources, particularly forests. Thus, study was intended quantify over three decades (1990–2022). Satellite images Landsat 5 8 were used years 1990, 2000, 2022. In addition, explores application machine learning (ML) algorithms land use (LULC) classification employs a random algorithm supervised technique. Accordingly, aims assess in district, Oromia Region, Ethiopia, using combination techniques remote sensing comprehensively analyze dynamics. The results depicted that significant reductions losses forested areas have observed by study’s analysis results. There 67.21% decline overall amount or 859.5 ha. These highlight how community‐driven conservation reforestation programs needed stop additional ecosystem degradation protect areas. Likewise, approach can offer benefits resource management, environmental monitoring, estimation services. Furthermore, findings aid decision‐makers, planners, making timely data‐driven decisions interventions afforestation protection activities mitigate rapid resources area.

Language: Английский

---

Assessing and Enhancing Predictive Efficacy of Machine Learning Models in Urban Land Dynamics: A Comparative Study Using Multi-Resolution Satellite Data

Remote Sensing, Journal Year: 2023, Volume and Issue: 15(18), P. 4495 - 4495

Published: Sept. 12, 2023

Reliable and accurate land-use/land cover maps are vital for monitoring mitigating urbanization impacts. This necessitates evaluating machine learning simulations incorporating valuable insights. We used four primary models, logistic regression (LR), support vector machine, random decision forests, artificial neural network (ANN), to simulate land Tsukuba City, Japan. incorporated an auxiliary input that multinomial enhance the ANN obtained a fifth model (ANN was run twice, with without new input). Additionally, we developed sixth simulation by integrating predictions of LR using fuzzy overlay, wherein had additional alongside driving forces. study employed six classified three different resolutions: first involved 15 m (ASTER) covering area 114.8 km2, second third, 5 0.5 (derived from WorldView-2 GeoEye-1) 14.8 models were then evaluated. Due synergistic effect, demonstrated highest kappa in all data, 86.39%, 72.65%, 70.65%, respectively. The results indicate stand-alone learning-based achieved satisfactory accuracy, minimalistic approaches can be improve their performance.

Language: Английский

---

Extracting Urban Built-up Areas from Optical and Radar Data Fusion using Machine Learning Algorithms

International Journal of Image and Data Fusion, Journal Year: 2024, Volume and Issue: 15(2), P. 154 - 173

Published: April 2, 2024

Accurate and up-to-date information on urban built-up areas is significant for managing growth development. Earth Observation (EO) data are valuable sources meeting this demand. However, the extraction of from EO challenging due to limitations sources. To overcome challenge, study follows an approach that assesses performance optical (Sentinel-2), radar (Sentinel-1) fused (Sentinel-1 Sentinel-2) extract using machine learning algorithms including Random Forest (RF), K-Nearest Neighbors (KNN) KDTree KNN. The results were statistically analyzed by considering Overall Accuracy (OA) kappa coefficient. In addition, 15 cm GSD (Ground Sample Distance) aerial photography area was used validate results. According results, Sentinel-2 produced better representation accuracy than Sentinel-1 even image. Regarding classification performance, RF performed in both OA Kappa coefficient along all datasets. research findings can have implications various domains, such as planning, land use management open avenues further comparisons different extraction.

Language: Английский

---

Evaluation of OLI Landsat-8 images based on spectral indices in detecting areas affected by mining tailings mud: a case study of the Brumadinho dam rupture, Brazil

Boletim de Ciências Geodésicas, Journal Year: 2024, Volume and Issue: 30

Published: Jan. 1, 2024

The socio-environmental impacts caused by the collapse of a mining dam can be irreversible. In Brazil, at Córrego do Feijão Mine was considered one worst disasters in country. Remote sensing-based approaches have been used to detect and monitor areas affected tailings from rupture. Therefore, it proposed identify area tailing mud, Brumadinho, Minas Gerais State, through analysis three different spectral indices: Normalized Difference Vegetation Index (NDVI), Ferrous Minerals Ratio (FMR) Clay (CMR). These indices were computed Operational Land Imager (OLI) images Landsat-8. Different thresholds tested define best range for delineating area. For validation, limits area, obtained higher resolution sensor, GeoEye-1, as reference. methodology demonstrated great potential detecting failure. NDVI FMR delimited interest with high performance, precision varying between 95% 92%; recall 88% 87%; F-score 91% 89%; global accuracy 84% 80%, showing suitable mapping such disasters.

Language: Английский

---