Rapid and reliable identification of multivariate geochemical anomalies is critical for delineating prospective mineralized zones and reducing uncertainty in mineral exploration targeting. Extended isolation forest (EIF) is a powerful unsupervised ensemble learning algorithm that efficiently isolates anomalies from high-dimensional geochemical datasets using randomly oriented hyperplane partitions. Previous studies have demonstrated the effectiveness of EIF in multivariate geochemical anomaly detection and mineral potential modeling. However, its performance can be significantly affected by stochastic variability arising from random partitioning and random subsampling during isolation tree construction, which may result in unstable anomaly patterns and inconsistent exploration targets in complex geological environments. To mitigate this limitation, we developed a robust unsupervised framework for the identification of multivariate geochemical anomalies associated with gold mineralization in the Southwestern Yilgarn Craton, Australia. The proposed framework integrates robust factor analysis (RFA), a Jaccard-based stability index and EIF to enhance the reliability and reproducibility of anomaly detection. RFA was first applied to compositional soil geochemical data to identify the most significant pathfinder elements associated with gold mineralization, which were subsequently used as input variables for the EIF model. The model was then optimized using a Jaccard-based stability criterion to ensure consistent anomaly detection across repeated independent runs. Model performance was assessed using area under the receiver operating characteristic curve (AUC). The obtained AUC value of 0.82 indicates strong predictive capacity, confirming that the generated anomaly map effectively delineates mineralization-related geochemical patterns and provides a reliable proxy for mineral prospectivity mapping. Overall, the proposed framework offers a robust and reproducible unsupervised approach for multivariate geochemical anomaly detection with strong applicability in both greenfield and brownfield mineral exploration settings.
Water conservancy and hydropower projects enhance regional climate resilience and watershed water security yet inevitably trigger large-scale involuntary reservoir resettlement. As representative involuntarily displaced populations, reservoir resettlees' social identity directly impacts local social stability and regional sustainable socioeconomic development. Based on a ten-year longitudinal qualitative investigation including in-depth interviews and participant observation in Village Y of Wuxikou Reservoir, Jiangxi Province, this paper divides the entire resettlement process into three stages: relocation, stabilization and development. From the dual perspectives of host community identity and out-groups identity, this study explores the dynamic evolutionary rules of resettlees' social identity. The results indicate that resettlees sequentially develop alienated identity, superficial adaptive identity and segregated identity across different phases. On this basis, the core concept of differential identity is proposed, which features a dual structure of vertical temporal differentiation and horizontal spatial differentiation. This paper expands the applicable scope and explanatory power of the “differential mode of association” and social identity theory in involuntary resettlement contexts. Grounding on the differential identity framework, this paper puts forward targeted integrated governance solutions to break intergroup segregation, facilitate cross-group integration and build resilient resettlement communities consistent with Sustainable Development Goals (SDGs) 11 and 13.
The discrimination of ore deposit types is primarily based on geological, geochemical, and isotopic characteristics. Conventionally, these types are identified using specific element diagrams. However, traditional geochemical methods often fail to determine scheelite deposit types of the complex Xuefengshan Sb-Au-W metallogenic belt in China, where mineralization resulted from the superposition of multiphase geological events. Machine learning (ML) methods, have been increasingly applied to identify deposit genesis by establishing relationships between deposit characteristics and genetic types using extensive datasets. However, inaccurate data labels, the limitations of single models, and poor model interpretability lead to decreased accuracy. This study proposes a ML framework based on interpretable ensemble learning. We collects geochemical element data from typical orogenic and magmatic-hydrothermal scheelite deposits globally. Deep clustering is used to filter data and overcome the subjectivity of original data labels. An ensemble learning model is used to construct a classifier to improve the model's robustness and generalization ability. An interpretable model is introduced to analyze the contribution of individual feature elements, revealing the metallogenic genesis. This method demonstrates high accuracy on the test set. According to this method, the scheelite deposit type of the Xuefengshan metallogenic belt is primarily magmatic-hydrothermal in origin, with orogenic superposition. This helps resolve a long-standing controversy in the region and establishes a repeatable and interpretable new paradigm for ML-based discrimination of ore deposit genetic types.
Though volcanogenic massive sulfide (VMS) deposits are major global sources of indium (In), the physicochemical mechanisms and key factors controlling its significant enrichment remain poorly understood. To address the issue, this study investigates the Tiemurt VMS Pb-Zn-Cu deposit, utilizing detailed petrography, in-situ LA-ICP-MS analysis, and thermodynamic modeling to reveal the In enrichment mechanisms in VMS deposits. Petrographic observations identified two distinct generations of sphalerite corresponding to different mineralization stages. The early-stage sphalerite (Sp1) is euhedral-subhedral, associated with pyrite, and displays darker colors (red to brown), whereas the late-stage sphalerite (Sp2) is anhedral, intimately intergrown with chalcopyrite, and shows lighter colors (mainly yellow). The trace element results demonstrate that Sp1 has a significantly higher In content (average 317 ppm) than Sp2 (average 220 ppm). Additionally, In concentrations positively correlate with Fe contents. Because Fe is the primary chromophore that darkens sphalerite, this strong coupled enrichment mechanism allows macroscopic sphalerite color (red > brown > yellow) to serve as a reliable indicator for In concentration. Crystallization temperatures calculated using the GGIMFis thermometer range from 344 to 382 °C for Sp1 and 312 to 355 °C for Sp2, indicating a cooling trend during fluid evolution. Thermodynamic modeling data showed that in the early-stage hydrothermal fluids (≥360 °C), Zn2+ preferentially complexes with Cl−, leaving InCl2+ or In3+ as unstable species, and In efficiently precipitates into Sp1 under the environment of log fO2 = −32 to −26 and pH = 6–8. As the fluids cool at ~340 °C, weakened Zn2+ competition allows In3+ to form stable InCl3, and In precipitates into Sp2 under the conditions of log fO2 = −42 to −32 and pH = 5.5–11. We therefore conclude that the key factor controlling the difference in In content between Sp1 and Sp2 is the precipitation mechanism rather than migration capacity. This may be different from the In enrichment mechanism associated with magmatic hydrothermal systems, where In is mainly present as InCl3 complexes with strong migration capacity. These new findings enable us to understand how the physicochemical conditions of fluids control the enrichment of In in VMS deposits, and also highlight that the color of sphalerite can be used to target potential In resources in PbZn deposits.
The Qukulekedong Au–Sb deposit, located in the western East Kunlun Orogenic Belt, contains resources of 21.5 t Au and 73 kt Sb and has been interpreted as an intrusion-related Au–Sb system. However, the timing and exploration significance of deep, proximal skarn mineralization remain poorly constrained. Here, we present new petrographic, mineralogical, mineral geochemical, and UPb geochronological data for the intrusive rocks and newly identified deep skarn. Zircon UPb dating yielded crystallization ages of 211.5 ± 1.9 Ma for the diorite porphyry and 211.7 ± 1.1 Ma for the gabbro diorite, indicating broadly coeval Late Triassic magmatism. Garnet UPb dating of the deep skarn yielded ages of 208.4 ± 5.8 Ma and 208.2 ± 4.8 Ma, directly constraining a Late Triassic skarn mineralization event. The deep skarn is characterized by grossular-rich garnet, diopside, actinolite, vesuvianite, epidote, chlorite, pyrite, arsenopyrite, and pyrrhotite, and was subsequently overprinted by quartz–sulfide and quartz–stibnite–carbonate veins. Zircon and amphibole mineral chemistry indicate that the Late Triassic intrusions crystallized from relatively hot, hydrous magmas under near-FMQ to slightly reduced conditions. Trace-element variations in garnet record progressive fluid evolution from Grt-a, characterized by high HFSE, HREE, V, and Cr contents and high Th/U ratios, to Grt-b, characterized by lower REE and HFSE contents and a closer association with hydrous alteration minerals. These results indicate that the deep skarn represents a proximal, high-temperature component of the Qukulekedong intrusion-related Au–Sb system, thereby expanding exploration targets from shallow Au–Sb veins to deep contacts between intrusions and carbonate-bearing strata, fault-intersection zones, and skarn zones containing Grt-a-type garnet.
Reservoir resettlement is a complex, and long-term recovery process. It involves more than physical relocation, extend to livelihood reconstruction, improvement of living conditions, and psychological recovery. To design post-resettlement policies, it is crucial to understand how these areas interact. This study develops an integrated system dynamics and agent-based simulation (SD-ABS) framework to examine the multidimensional recovery trajectories of resettlers affected by the Yangtze-to-Huaihe Water Diversion Project in China. The framework captures the co-evolution of production recovery, living recovery, and psychological recovery, as well as the feedback mechanisms linking individual behaviors and system-level changes. The model is parameterized and validated using from official monitoring reports, statistical yearbooks, and household surveys. Results show production recovery strongly depends on employment opportunities, skills training, and policy timing. Growth type policies yield better long-term employment outcomes than decrease type approaches. Living recovery can be accelerated by increased investment, however the effectiveness of investment depends on its alignment with resettlers' actual needs. Psychological recovery lag behind other trends and is strongly shaped by kinship and rebuilding social network. These findings show that post-resettlement recovery is dynamic and shaped by nonlinear economic, living, and psychological factors. The proposed SD-ABS framework provides a novel way to analyze long-term recovery. The study offers practical policy insights for employment support, targeted investments, and community integration. The findings help improve resettlement management for large-scale water infrastructure projects.
Manganese carbonates can be formed via either diagenetic reduction of Mn oxides from hydrothermal and marine environments by organic matter, or direct growth on calcite or dolomite in anoxic environments. The Triassic Heqing manganese carbonate ores in the Heqing basin, Yunnan Province, South China provides an excellent case to show they form by redox between Mn oxides and methane during diagenesis in a low sulfate freshwater environment. The manganese ores are hosted by limestones and predominantly composed of micro-to fine-crystalline rhodochrosite and massive hausmannite. An oxic depositional environment is indicated by the bulk rocks showing negative Ce anomalies (Ce/Ce* = 0.36–1.59, average = 0.89) from which the hausmannite was deposited, and subsequently replaced or encrusted by rhodochrosite. The manganese ores exhibit no significant Eu anomaly (Eu/Eu* = 0.72–1.18, average = 1.01), and are plot on hydrogenous area of SiO2 versus Al2O3 diagram, reflecting that the manganese was supplied dominantly from chemical weathering. A low sulfate freshwater sedimentary and early diagenetic environment is indicated by low Y/Ho and La/La* ratios for the bulk rocks or Mn carbonates, the absence of pyrite and sulfate minerals. Under such an environment, hausmannite was reduced by biogenic methane and organic matter to generate rhodochrosite as shown by two mixing trends with endmember δ13C values lighter than −70‰ and − 25‰, respectively. We report a rare case showing that strongly 13C-depleted Mn carbonate ore was formed in a freshwater lacustrine environment and can serve as a methane sink.
The trace element composition of galena provides important constraints on the genetic classification of Zn-Pb deposits. However, conventional binary discrimination diagrams are limited in complex multi-element geochemical signatures and often result in ambiguous interpretations. In this study, we compiled a global dataset comprising 1649 LA-ICP-MS analyses of galena from five major genetic types of Zn-Pb deposits, including skarn, epithermal, Mississippi Valley-type (MVT), volcanogenic massive sulfide (VMS), and sedimentary exhalative (SEDEX). Eight trace elements, including Zn, Cd, Cu, Ag, Sn, Sb, Tl, and Bi, were selected to construct a machine learning classification framework. The Random Forest machine learning model achieves a macro-averaged F1-score of 0.82, demonstrating good predictive performance. The SHapley Additive exPlanations (SHAP) analysis reveals that Tl, Bi, Sn, and Ag exert the greatest influence on model predictions, highlighting their diagnostic significance. The trained model was applied to three deposits with debated genetic affinities. The Huayangchuan and Xiyi deposits are classified as SEDEX type, whereas the Shuangjianzishan deposit is epithermal type, consistent with the geological characteristics of the deposits. The results of this study demonstrate that the galena trace element compositions are effective in Zn-Pb deposit classification.
The water-level-fluctuation zone (WLFZ) of reservoirs is a critical hotspot for cadmium (Cd) accumulation, yet the stabilization kinetics of exogenous labile Cd in these periodically inundated soils remain poorly understood, limiting accurate ecological risk assessment. This study integrated long-term field monitoring, laboratory incubation, and multi-model simulations in the Xiangxi River WLFZ of the Three Gorges Reservoir to investigate soil Cd dynamics and develop a predictive model for Cd stabilization. Results revealed pronounced spatial heterogeneity in soil properties and Cd distribution, with total Cd concentrations ranging from 0.31 to 2.06 mg/kg (mean: 0.64 mg/kg), nearly five times the local background value. The mean geo-accumulation index (1.61) indicated light-to-moderate pollution, primarily from historical agricultural non-point sources, though their relative contribution has declined. Spiked Cd bioavailability decreased in three stages, accurately described by a pseudo-second-order kinetic model (R2 > 0.90 for 86.4% of samples). By coupling this kinetic equation with a Gradient Boosting algorithm, a hybrid predictive model was developed that robustly simulated exogenous Cd stabilization (R2 = 0.862, RMSE = 0.024). This study clarifies the key drivers of Cd fate in WLFZ soils and provides a quantifiable tool to support long-term risk assessment and targeted management of Cd contamination.
Constraining the physicochemical conditions that govern lithium (Li) enrichment and the formation of Li-bearing clay minerals is critical for understanding the genesis of clay-type Li deposits. However, these conditions remain difficult to define because Li-bearing clay minerals commonly form through multistage fluid–rock interaction, and direct evidence for Li-bearing fluids is rarely preserved in clay-rich systems. To address this issue, this study investigates the hydrothermal fluid–rock interaction responsible for Li enrichment and Li-bearing clay formation in the Mesoproterozoic Wumishan Formation, through thermodynamic modeling using GEM-Selektor. Based on recalculated clay-mineral structural formulas and custom thermodynamic data, titration and leaching models were performed at 200 °C, 220 °C, 240 °C, and 260 °C and 1000 bar to evaluate the effects of temperature, fluid/rock ratio, and host-rock buffering. The results show that Li enrichment was controlled by staged hydrothermal alteration in a carbonate-dominated host system, governed by temperature, fluid/rock ratio, and carbonate buffering. In both models, mineral assemblages evolved systematically with reaction progress, and pH was strongly buffered by the carbonate host rocks. Li-rich smectite-rich illite/smectite mixed-layer (Li-I/smectite) formation was closely linked to pH evolution and occurred within a specific pH window of approximately 7.6–8.8 during fluid–rock interaction. Li-I/smectite was the only Li-bearing clay phase consistently resolved in both models, suggesting preferential stabilization within a restricted hydrothermal-geochemical window. These results provide a thermodynamic perspective on the favorable conditions for Li-bearing clay formation and offer a new approach for evaluating the ore-forming environment of clay-type lithium deposits.
Lead‑zinc mineralization in South China is commonly associated with Mid-Late Jurassic (∼170–150 Ma) magmatic activity. However, the genesis of many deposits lacking direct magmatic connections remains poorly constrained owing to limited geochronological data. The Hengyang Basin, located within the Qin-Hang Metallogenic Belt (QHMB, South China), hosts numerous PbZn deposits whose origins remain debated, with proposed models including magmatic-hydrothermal and Mississippi Valley-type (MVT) styles. The Liushutang deposit (1.36 Mt at 0.85% Pb and 4.49% Zn) is a representative example. Its mineralization comprises three stages: (I) quartz-pyrite, (II) quartz-pyrite-chalcopyrite-galena-sphalerite (main ore stage), and (III) quartz-barite-sphalerite. Integrated geological, mineralogical, and isotopic evidence, including fault-hosted vein geometry, pervasive silicification, a quartz-barite-sulfide assemblage with chalcopyrite and tetrahedrite, magmatic-like HO isotopic signatures, and extremely high Rb/Sr ratios (mean 839) of hydrothermal roscoelite, collectively points to a magmatic-hydrothermal origin for Liushutang. High-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) and laser ablation-inductively coupled plasma mass spectrometry (LA-ICP-MS) data show that Ga enrichment in sphalerite (up to 857 ppm) results primarily from lattice incorporation via Cu-coupled substitution under high-salinity conditions associated with phase separation triggered by rapid fluid depressurization. The Stage II roscoelite RbSr isochron age of 105.8 ± 5.4 Ma (MSWD = 1.3) constrains the timing of PbZn mineralization and clearly distinguishes it from the ∼158 Ma Mid-Late Jurassic magmatic-hydrothermal PbZn deposits (e.g., Kangjiawan and Shuikoushan) in the Hengyang Basin. Although no coeval intrusions are exposed at Liushutang, geophysical data indicate the presence of concealed intrusions beneath the deposit. This ∼106 Ma age coincides with a regional extensional tectonic regime (ca. 135–80 Ma) that caused lithospheric thinning, asthenospheric upwelling, and the formation of NE-trending pull-apart basins, collectively providing the thermal driver and structural permeability for ore formation. Consequently, the Hengyang Basin records two distinct episodes of magmatic-hydrothermal PbZn mineralization: Mid-Late Jurassic and Early Cretaceous, with the Liushutang deposit exemplifying the latter. This finding reveals a previously underrecognized Early Cretaceous magmatic-hydrothermal metallogenic episode in the QHMB and provides a valuable framework for regional exploration by highlighting the roles of concealed intrusions and extensional tectonic settings in PbZn ore formation and critical metal (Ga) enrichment.
The Carajás Mineral Province in the Amazonian Craton hosts the world's oldest known world-class IOCG deposits (ca. 2.72–2.68 and 2.55 Ga). At the northern part of the province, iron-rich rocks containing magnetite and a distinctive CaFe- and KFe-rich silicate assemblage (garnet, amphibole, and biotite) envelop the copper‑gold ore within the IOCG deposits that are hosted by the Cinzento Shear Zone (e.g., Salobo, GT-46, QT-02, Furnas). Although these Fe-rich assemblages clearly predate IOCG mineralization, their origin remains debated, with interpretations attributing its formation to either ore-forming hydrothermal processes or metamorphism. This study integrates petrography, electron probe microanalysis, δ18O and δD isotopes, and LuHf and SmNd garnet geochronology to unravel the processes responsible for iron-rich rocks within the IOCG deposits along the Cinzento Shear Zone. The results indicate that early metasomatic minerals (spessartine, Ca-amphibole, Ti-rich biotite, and sillimanite) formed prior to the Fe-rich assemblage observed today. Garnet LuHf apparent ages of ca. 2.64 Ga in Salobo and GT-46 indicate crystallization prior to the main IOCG mineralizing event, whereas ca. 2.55 Ga SmNd data record complete isotopic resetting by IOCG fluids. At QT-02, similar LuHf (2546 ± 19 Ma) and SmNd dates (2527 ± 38 Ma) suggest syn-shear garnet growth at limited fluid circulation. The presence of spessartine, representative of primary garnet compositions, meta-BIFs, tourmaline-rich zones and garnet-sillimanite-rich rocks, together with the record of fluids derived from mixed magmatic and seawater sources, suggest the development of ancient exhalative systems within the Carajás Province. The early metasomatic paragenesis formed through the modification of subseafloor Fe-Mn-rich sediments and phyllic/argillic hydrothermal halos. This modification resulted from a combination of thermal input from the emplacement of sublithospheric melts, widespread magmatism and early IOCG-related fluids (ca. 2.72 Ga). Subsequently, hot (640 °C to 350 °C) K- and Fe-rich IOCG hydrothermal fluids at ca. 2.55 Ga, lead to the formation of substantial volumes of almandine-grunerite and Fe-biotite zones, which are recognized in these IOCG deposits as characteristic ore envelopes. The protracted thermal evolution of the province culminated in lithospheric weakening and strain localization, which facilitated the development of large-scale structures such as the Cinzento Shear Zone. These structures, in turn, channeled hydrothermal fluid and exerted primary control over the distribution of IOCG mineralization.
Excessive amounts of ammonium (NH4+) can trigger the occurrence of eutrophication, algal blooms and diseases. Hence, identifying the causes of NH4+ pollution is crucial for protecting water resources and the ecological environment. Although the causes of high NH4+ concentrations have been extensively studied, whether the dissimilatory nitrate reduction to ammonium (DNRA)-mediated conversion of nitrate (NO3−) to NH4+ results in ammonium pollution remains uncertain. In this work, the source of NH4+ in groundwater of the Yangtze Estuary was identified through the integration of nitrogen isotopic abundance, microbial communities and nitrogen (15N) isotope tracer analysis techniques. The contributions of introduced NO3− to NH4+ via DNRA were simulated through labelling experiments. Our results revealed that the DNRA rates in groundwater ranged from 0.261 to 0.362 μmol/L/h and exceeded those of nitrification, denitrification and anammox. Both the isotopic abundance and microbial communities indicated that DNRA is a critical factor contributing to the NH4+ concentration in groundwater of the Yangtze Estuary. The results of a simulation experiment revealed that the proportions of the DNRA rate to the total NO3− reduction and accumulation rates of NH4+ at different concentrations of K15NO3 (0.5, 1, 5, 10, 30 and 60 μmol/L) ranged from 67.04% ~ 80.78% and 0.163–0.506 μmol/L/h, respectively. Therefore, introduced NO3− could significantly contribute to NH4+ via DNRA in groundwater of the Yangtze Estuary. Notably, the contribution increased (y = 2.318ln(x) + 70.424, R2 = 0.782) with increasing NO3− input concentration. The input of NO3− in global DNRA-active areas should be considered, especially inputs from anthropogenic activities. Our results could advance our understanding of anthropogenic influences on groundwater contamination.
Intelligent mineral exploration is still constrained by fragmented data sources, disconnected analytical workflows, weak knowledge organization, and limited system-level traceability and reproducibility. To overcome these limitations, we present GoldMiner-AI, an end-to-end intelligent system engineered for mineral exploration tasks. The system leverages a RuoYi-Cloud-Plus microservice architecture and a collaborative multi-database framework (PostGIS, Neo4j, Milvus, MySQL) to unify the management of multi-source, heterogeneous geoscientific data spanning geological maps, geochemical and geophysical surveys, drilling logs, field observations, and textual reports. Its intelligent core integrates two key components: the K-means-Association Rule Graph (KAR-Graph) anomaly identification framework and the deep learning model for multi-source feature fusion. Coupled with a mineral-systems knowledge graph and retrieval-augmented generation (RAG), this architecture supports a domain-tailored large language model (LLM). This design establishes a closed-loop intelligent workflow that integrates anomaly detection, target delineation, knowledge-aided reasoning, and interactive Q&A. Validation experiments in representative ore districts (Youjiang Basin and southern Qin-Hang metallogenic belt) demonstrate that: (1) It effectively identifies Carlin-type gold-related Au-As-Sb-Hg anomalous assemblages and extracts diagnostic geochemical fingerprints of ore-forming processes; (2) Through multi-source data fusion, it accurately predicts the spatial distribution of PbZn mineralization zones; and (3) The domain-specific LLM mitigates the “hallucination” common in general-purpose models, significantly improving the accuracy of geoscientific knowledge retrieval and Q&A. Collectively, GoldMiner-AI provides a reproducible, scalable, and production-ready platform that advances mineral exploration toward fully intelligent and systematic workflows.
The growing global demand and supply risk of rare earth elements and yttrium (REY) have prompted increasing attention to coal-hosted REY deposits. The Luxi Region of Shandong Province, an important coal-bearing area in North China, remains poorly studied with respect to REY enrichment mechanisms. This study integrates geochemical analysis, sequential chemical extraction procedure, and provenance constraints to elucidate the occurrence and enrichment processes of REY in the No. 10 coal seam of the Chazhuang Mine, Luxi Region. The coal contains an average of 231 μg/g REY, reaching 1054 μg/g in the lower section. The REY host phases exhibit vertical differentiation characteristics: REY in the upper section is predominantly hosted in phosphate minerals, whereas those in the lower section occur in a mixed form of organic matter-bound and phosphate mineral-bound states. Geochemical evidence indicates that terrigenous clastic input, acidic groundwater leaching, and marine influence jointly constrained the REY enrichment processes in the coal seam. It is inferred that REY enrichment in the coal seam was based on provenance supply. The Sr/Ba and δCe values indicate that the coal was formed in a reducing marine environment, while the Zr/TiO2 and Nb/Y ratios as well as REY distribution patterns imply acidic groundwater activity; these two factors synergistically participated in the leaching, migration, and enrichment of REY. One sample from the lower section shows anomalously high LREY content, a phenomenon that may be closely related to the vertical migration of rare earth elements under acidic conditions and the secondary enrichment of rare earth elements driven by marine influence. Provenance analysis indicates dominant material contributions from Neoarchean granites of the Yinshan Oldland, with additional contributions from the Qinling Orogenic Belt. The average total rare earth oxide (REO) content of 2337 μg/g and a Coutl value of 1.46 suggest moderate economic potential. These results refine the understanding of REY enrichment processes in coal-bearing basins and guide the evaluation of the critical metal potential of coals in North China.
To effectively manage the extensive risk of manganese (Mn) contamination in soils, the Mn content must be monitored through low-cost and efficient methods. Visible and near-infrared reflectance spectroscopy presents a promising alternative to traditional soil assessment methods. In this study, we constructed and evaluated a deep convolutional neural network (CNN) model for precisely predicting soil Mn contents on a continental scale. The correlation between the actual soil Mn content and the spectral under different preprocessing methods was analyzed. The results indicate that combining Savitzky-Golay smoothing and the second derivative could effectively improve the CNN predictive performance. After hyperparameter tuning, the R2, root mean square error, mean absolute error, and ratio of performance to deviation were 0.66, 132.02 mg/kg, 89.24 mg/kg, and 1.71 for the optimal CNN model on the test set, respectively. The distribution of soil Mn content predicted by the model was highly similar to the actual distribution. The uncertainty analysis shows that the CNN prediction was stable and reliable. Our model has shown a practical and effective predictive capability for soil Mn content on a large scale, providing an important tool and reliable method for agricultural management and environmental monitoring.
Arsenic release from mine waste rock can occur under circumneutral pH conditions, leading to contaminated neutral drainage (CND), even in the absence of acid generation. In cold and subarctic environments, low temperatures and freeze-thaw cycles further influence the hydrogeological and geochemical processes controlling contaminant mobilization. This study investigates arsenic release from non-acid generating waste rock containing gersdorffite (NiAsS) using a combination of laboratory column experiments and reactive transport modeling. Laboratory column experiments conducted under controlled temperature conditions (21 degrees C, 5 degrees C, and-2 degrees C) and for different grain size fractions were reproduced using the MIN3P reactive transport code. Due to the absence of gersdorffite from standard thermodynamic databases, this mineral was explicitly implemented in MIN3P and parameterized as a kinetically controlled primary phase. Results show that grain size exerts a first-order control on arsenic release through mineral liberation and reactive surface area, while decreasing temperature reduces but does not suppress arsenic mobilization as long as liquid water is present. Long-term simulations extending to 100 years reveal strongly nonlinear arsenic depletion behaviour, with substantially longer persistence than suggested by linear extrapolation of laboratory kinetic testing results, particularly under cold conditions.
Short-wave infrared (SWIR) combined with the in-situ chemistry analyses of alteration minerals could provide critical constraints on ore-forming processes and represents a valuable exploration targeting tool for hydrothermal mineral deposits. Here we present SWIR spectral data and trace element compositions of white mica from two world-class gold deposits in Western Australia: Kanowna Belle (283 t Au) and Sunrise Dam (314 t Au). White mica (Al-OH absorption wavelengths = 2195 to 2219 nm) in the Kanowna Belle deposit occurs in the hydrothermal veins, with enrichments in Li, Fe, Cu, Cs, V, Zn and depleted in As, Pb, and Rb. White mica (Al-OH absorption wavelengths = 2197 to 2212 nm) in the Sunrise Dam deposit occurs in bent quartz-carbonate-mica veins, which are enriched in As, Pb, Cr and depleted in Cu, Li, Zn, and V. Elevated As and Pb concentrations in Sunrise Dam white mica suggest metamorphic-sourced ore-forming fluids, whereas MgFe enrichment in Kanowna Belle white mica implies magmatic-hydrothermal contributions. Significant linear correlations between white mica Cu/Zn ratios, Rb/Sr ratios, and Al-OH wavelengths reflect variations in hydrothermal alteration intensity. Thermodynamic modeling combined with V/Sc ratios reveals that ore-forming fluids in Kanowna Belle and Sunrise Dam are oxidized, silica-rich (Log alpha SiO2 = similar to -1.6, LogfO(2) = similar to -30 to -32, pH < 4) and reduced, sodic (Log alpha SiO2 = similar to -1.7, LogfO(2) = similar to -35 to -38, pH > 4), respectively. We propose diagnostic criteria, in which mineralized white mica at the Kanowna Belle deposit exhibits phengitic composition with elevated As concentrations, high V/Sc ratios (>2.5), and long Al-OH wavelengths. In contrast, mineralized white mica from the Sunrise Dam deposit shows lower Cu/Zn ratios, shorter wavelengths, and PbAs enrichment.
The presence of sulfide minerals in Carlin-type gold deposits can host significant concentrations of critical raw materials (CRMs) in addition to gold mineralization. This study employs a range of analytical techniques, including bulk geochemical analysis, optical and scanning electron microscopy, electron probe microanalysis (EPMA), and laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS), to investigate the trace element composition of sulfide minerals in the Zarshuran deposit, northwestern Iran. The results reveal notable enrichments of CRMs such as As, Hg, Te, and Sb within the gold-bearing and arsenical ores. The findings demonstrate that the late-stage arsenical ore not only ensures economic efficiency in As extraction but also represents a potentially valuable economic source of Sb (1260-207,453 ppm, average 42,100.82 ppm), Te (8.77-1789 ppm, average 396.81 ppm), Tl (159.7-4483 ppm, average 1613.5 ppm), Hg (0.008-0.152 wt%, average 0.047 wt% by EPMA), and Se (2.56-2313 ppm, average 473.73 ppm), as well as Au (32.69-2534 ppm, average 655.91 ppm). These elements may therefore constitute recoverable resources. Micron-scale inclusions of stibnite and getchellite, together with Sb-rich bright bands in orpiment, are the primary sources of Sb enrichment, whereas micro-inclusions of lorandite, simonite, galkhaite, and coloradoite in orpiment account for the enrichment of Tl, Te, and Hg. In contrast, pyrite and arsenopyrite primarily act as the principal hosts for gold mineralization. This study contributes significantly to the understanding of Carlin-type gold deposits by highlighting the importance of monitoring CRMs, thereby supporting the alignment of Iran's mining sector with global resource and sustainability standards.
