Chemie Der Erde-Geochemistry最新文献

{"title":"Unique achondritic impact debris in the CH3 chondrite Acfer 182","authors":"Kainen L. Utt ,&nbsp;Ryan C. Ogliore ,&nbsp;Michael J. Krawczynski ,&nbsp;Paul K. Carpenter ,&nbsp;Alian Wang","doi":"10.1016/j.chemer.2024.126202","DOIUrl":"10.1016/j.chemer.2024.126202","url":null,"abstract":"<div><div>The metal-rich CH carbonaceous chondrites contain abundant xenolithic clasts originating from different regions of the Solar System. In the CH3 chondrite Acfer 182, we identified two phosphide spherules (one 95-μm in diameter and the other <span><math><mn>50</mn><mspace></mspace><mi>μm</mi><mo>×</mo><mn>60</mn><mspace></mspace><mi>μm</mi></math></span>) of schreibersite ((Fe,Ni)₃P) and barringerite ((Fe,Ni)₂P) with kamacite eutectic structures. These objects are likely to have formed during an impact between planetesimals during the debris-disk phase of the protoplanetary disk before being incorporated into the CH chondrite parent body. In the same sample we identified a <span><math><mn>130</mn><mspace></mspace><mi>μm</mi><mo>×</mo><mn>60</mn><mspace></mspace><mi>μm</mi></math></span> heideite grain (iron‑titanium sulfide: (Fe,Cr)_1.15(Ti,Fe)₂S₄) with exsolution lamellae of calcium-rich titanium oxide. Thin veins of shock-induced kamacite cross-cut the oxide lamellae, suggesting that it was ejected into the protoplanetary debris disk during an impact event before eventually being accreted by the CH chondrite parent body. This assemblage is distinct from heideite grains found in enstatite chondrites, aubrites, and the Kaidun meteorite. We propose that this object originated from a highly-reduced planetesimal in the inner Solar System that may have been similar to proto-Mercury.</div></div>","PeriodicalId":55973,"journal":{"name":"Chemie Der Erde-Geochemistry","volume":"84 4","pages":"Article 126202"},"PeriodicalIF":2.6,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143140423","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"3D mineral prospectivity modeling at the Axi epithermal gold deposit, NW China by using a feature adaptive fusion strategy","authors":"Xiancheng Mao,&nbsp;Jiaxuan Song,&nbsp;Zhankun Liu,&nbsp;Hao Deng,&nbsp;Jin Chen,&nbsp;Shuyan Yu,&nbsp;Yanan Wang,&nbsp;Ruike Xu,&nbsp;Yuanqian Nie,&nbsp;Yang Zheng","doi":"10.1016/j.chemer.2024.126190","DOIUrl":"10.1016/j.chemer.2024.126190","url":null,"abstract":"<div><div>Mineralization distribution is always intricately affected by multiple ore-controlling geological units that play different roles in a mineral system (e.g., driver, trap, and throttle). How to effectively balance and integrate ore-controlling features from various 3D geological models during 3D mineral prospectivity modeling (MPM) is still a challenging task. In this paper, we introduce a novel approach, the feature adaptive fusion convolutional neural networks (CNN), which is designed to learn multiple 3D geological models with ore-controlling functions. The method is validated in the Axi epithermal gold deposit, northwestern China that mineralization distribution is jointly controlled by fault, volcanic phase, and phyllic alteration. The geology units are firstly constructed by explicit-implicit modeling and their ore-controlling features are subsequently described by high-frequency Laplace-Beltrami eigenfunctions and reassembled into multi-channel images as input to CNN. To learn the differences in ore-controlling effects among various geological units, we designed a fully connected layer to achieve adaptive quantification and weighted integration of the ore-controlling features by automatically optimizing weight allocation parameters and bias vectors using the neural network intelligence. Comparison results between the proposed method and other prospectivity methods suggest that the feature adaptive fusion CNN produces more reliable predictions, characterized by: (1) high consistency with known mineralization, (2) the highest AUC value and success rate, and (3) accurate prediction of deep voxels explored by drilling. Therefore, the proposed method effectively integrates the ore-controlling effects of multiple geological units and is suitable for complex scenarios of 3D MPM. Utilizing the prospectivity results generated by our method, we identified five potential mineralization in the Axi gold deposit, laying a robust foundation for future gold exploration.</div></div>","PeriodicalId":55973,"journal":{"name":"Chemie Der Erde-Geochemistry","volume":"84 4","pages":"Article 126190"},"PeriodicalIF":2.6,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143140762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The influence of supervised and unsupervised methods in the analysis of low density stream sediments in the regional prospecting of pegmatites in the state of Espírito Santo, Araçuaí Orogen - Eastern Pegmatite Province of Brazil","authors":"Flávia Compassi da Costa ,&nbsp;Brener Otávio Luiz Ribeiro ,&nbsp;Lucas Pereira Leão ,&nbsp;Paulo Augusto de Paiva-Silva ,&nbsp;Ricardo Scholz","doi":"10.1016/j.chemer.2024.126181","DOIUrl":"10.1016/j.chemer.2024.126181","url":null,"abstract":"<div><div>The state of Espírito Santo is located in Brazil southeastern and it is geologically inserted in the Eastern Brazilian Pegmatite Province in the Araçuaí Orogen. This province is one of the most important pegmatite provinces in the world, known mainly for the occurrence of gem-quality minerals and high grade lithium deposits. Although Espírito Santo has a geological framework with many granite intrusions similar to those hosting mineralized pegmatites in Minas Gerais, it hasn't been acknowledged as a potential pegmatitic area. There are reports of mineralization of beryl (aquamarine), topaz, amethyst, and chrysoberyl in the state's mineral history. I this research we use regional stream sediment geochemical data of the Espirito Santo State, to create prospective geochemical maps to map gemological pegmatites. To do so, primary descriptive statistical analyses as well as Pearson and Spearman correlations and Principal Component Analysis (PCA) were used to select elements associated with pegmatites. The geochemical prospectivity mapping was carried out by supervised and unsupervised approaches. The supervised method was based on integration of the elements by C-A fractal models, used to define the geochemical thresholds, and fuzzy logic methodology to integrate weighted geochemical maps. The unsupervised approach, however, was based on interpolation of the representative principal component scores. All models were validated using the Prediction Area plot (P-A plot). The results of our research showed that unsupervised methodology overperformed the other weighted supervised models, with a prediction rate of 78.14 % and prospectivity of 0.87. The final results also show a pegmatite prospectivity map for the Espirito Santo state, that indicates a great exploration potential for gemological minerals.)</div></div>","PeriodicalId":55973,"journal":{"name":"Chemie Der Erde-Geochemistry","volume":"84 4","pages":"Article 126181"},"PeriodicalIF":2.6,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142255597","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Lithium exploration targeting through robust variable selection and deep anomaly detection: An integrated application of sparse principal component analysis and stacked autoencoders","authors":"Saeid Esmaeiloghli ,&nbsp;Alexandre Lima ,&nbsp;Behnam Sadeghi","doi":"10.1016/j.chemer.2024.126111","DOIUrl":"10.1016/j.chemer.2024.126111","url":null,"abstract":"<div><div>Lithium is a strategic metal for high-technology industries that plays a vital role in realizing electromobility and effective energy storage for smartphones and electric/hybrid vehicles, in addition to being important in accumulating energy from renewable sources. Motivated by this, the development of state-of-the-art workflows is imperative to make Li exploration more efficient and guarantee a sustainable supply to the demanding markets in the future. In this paper, a computational protocol was described to address two critical challenges raised in the Li exploration: (i) selection of mineralization-related pathfinder variables and (ii) recognition of multi-element geochemical anomalies related to ore-forming processes. Robust sparse principal component analysis (SPCA) was employed to reduce active (non-zero) loading coefficients and straightforward selection of Li pathfinder variables. The selected variables were then fed into the training network of stacked autoencoders (SAE) to learn deep representations of multivariate input signals and quantify reconstruction errors linked to Li-vectoring geochemical anomalies. As a representative example, the proposed workflow, i.e., SPCA + SAE, was implemented in the R programming language and applied to a real-case experiment with stream sediment geochemical data pertaining to the Moalleman district, NE Iran. Moreover, compositional robust principal component analysis (RPCA) and Mahalanobis distance (MD) technique were adopted to constitute two comparative models, RPCA + SAE and SPCA + MD, as benchmarks to judge the competence of the SPCA + SAE model in variable selection and anomaly detection, respectively. A performance appraisal by success-rate curves and relevant area under the curves (AUCs) indicated that the SPCA + SAE model brings, by calculating the highest AUC, spatial patterns that are more promising for vectoring towards Li-mineralized grounds. Moreover, <span><math><mi>AUC</mi><mfenced><mrow><mtext>SPCA</mtext><mo>+</mo><mi>MD</mi></mrow></mfenced></math></span> was measured to be greater than <span><math><mi>AUC</mi><mfenced><mrow><mtext>RPCA</mtext><mo>+</mo><mi>SAE</mi></mrow></mfenced></math></span>, suggesting robust variable selection is a more important paradigm than deep anomaly detection for Li exploration targeting. Student's <span><math><mi>t</mi></math></span>–statistic method was eventually applied to the anomaly map from the SPCA + SAE model to define relevant thresholds for narrowing down prospective areas for the next round of Li exploration within the study area.</div></div>","PeriodicalId":55973,"journal":{"name":"Chemie Der Erde-Geochemistry","volume":"84 4","pages":"Article 126111"},"PeriodicalIF":2.6,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140205247","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"REE and trace element mobility during the transformation of basalt to laterite and bauxite, Payas Province, Türkiye","authors":"Hüseyin Öztürk ,&nbsp;Zeynep Cansu ,&nbsp;Cem Kasapçı ,&nbsp;Nurullah Hanilçi ,&nbsp;David Banks","doi":"10.1016/j.chemer.2024.126083","DOIUrl":"10.1016/j.chemer.2024.126083","url":null,"abstract":"<div><div><span><span><span>The bauxites of the Payas area in the eastern </span>Mediterranean region<span> of Turkey occur as a stratigraphically continuous layer between Early and Late Cretaceous shallow marine carbonates. The laterites represent an in situ formation and mark a key stratigraphic layer in the region. Bauxite pockets are also present in the laterites, formed from the reworking of the laterites and accumulated in depressions in the karst terrain. Therefore, bauxite occurrences are found locally in this stratigraphic layer. The parental rock was a Ti-rich </span></span>basalt that is compositionally similar to Hawaiian basalts, as indicated by petrographic features and distinctive trace element composition (Zr/TiO</span>₂<span> and Nb/Y). The behavior of elements during the lateritization<span> and bauxitization processes was interpreted using the composition of the Hawaiian basalts with the iron laterite and Ti-rich bauxites of the Payas Region. During lateritization, Fe, Al, Ti, Cr, Nb, and Ta, were largely immobilie while REE<span> and other trace elements, except for Rb, Ni, Co, and Pb were significantly removed. Approximately 75 % of the REE were removed at the end of the bauxitization and/or transportation of the lateritic soil into the karstic depressions. REE were not equally mobile with La to Ce and Lu to Yb having a relatively lower mobility than highly mobile middle REE. During the transformation of laterite to bauxite, low mobility elements such as Ti, Nb, Th, Cr, Hf, and Sn were enriched in the bauxite phase. In comparison to other elements, Rb, As, Pb, Mo, and Ni were strongly removed from the lateritic material during the bauxitization processes. The transfer of MREE during lateritization and bauxitization would have resulted in the enrichment of both light and heavy REE (concave pattern) in contemporaneous seawater. In other words, silicate weathering periods on land should be marked by a concave REE pattern with positive Eu anomaly in marine sediments throughout the geological period.</span></span></span></div></div>","PeriodicalId":55973,"journal":{"name":"Chemie Der Erde-Geochemistry","volume":"84 4","pages":"Article 126083"},"PeriodicalIF":2.6,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139495158","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing training performance of convolutional neural network algorithm through an autoencoder-based unsupervised labeling framework for mineral exploration targeting","authors":"Seyyed Ataollah Agha Seyyed Mirzabozorg ,&nbsp;Maysam Abedi ,&nbsp;Mahyar Yousefi","doi":"10.1016/j.chemer.2024.126197","DOIUrl":"10.1016/j.chemer.2024.126197","url":null,"abstract":"<div><div>Convolutional neural networks (CNNs), a prominent deep learning approach, have garnered significant interest in the field of mineral potential mapping (MPM) due to their capability to capture and learn spatial features that traditional algorithms tend to overlook. The effectiveness of CNNs is closely tied to the quantity of training data available, thereby impacting the outcomes of MPM. Moreover, uncertainties arising from delineation of negative samples can compromise the reliability of MPM assessments. To deal with these challenges, we propose the utilization of an autoencoder-based anomaly detection technique for the purpose of annotating locations in an unsupervised manner. Subsequently, a CNN is trained using the unsupervised annotated data to generate a Fe prospectivity model within a specific region in Iran. To validate the effectiveness of the proposed method, we first perform an MPM with insufficient positive training samples that extract from the location of known occurrences. We then execute another MPM on a set of samples labeled based on the reconstruction error of an autoencoder network. When comparing the two prospectivity models, namely using augmented data or inadequate samples, it is evident that modeling with augmented data outperforms the MPM model trained with insufficient samples. This confirms the effectiveness of the adopted approach and shows that the unsupervised labeling technique proposed in this work can significantly improve the performance of the CNN in MPM.</div></div>","PeriodicalId":55973,"journal":{"name":"Chemie Der Erde-Geochemistry","volume":"84 4","pages":"Article 126197"},"PeriodicalIF":2.6,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143140118","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Variations in mineral composition and the weathering crust zoning of the REE-Nb chuktukon deposit (Chadobetsky uplift, Krasnoyarsk region)","authors":"A.V. Malyutina ,&nbsp;A.G. Doroshkevich ,&nbsp;I.A. Zhukova ,&nbsp;I.R. Prokopyev","doi":"10.1016/j.chemer.2024.126210","DOIUrl":"10.1016/j.chemer.2024.126210","url":null,"abstract":"<div><div>The Chuktukon REE-Nb deposit occurs within the carbonatite laterite supergene zone Chadobets upland, South of the Siberian craton. A unique laterite profile up to 600 m thick was identified by deep drilling. The laterite developed upon ultramafic lamprophyres and Ti-Th-REE-rich carbonatites. Here we report new mineralogical and geochemical data on carbonatites and laterites that reveal the evolution of supergene processes in the Chuktukon deposit within vertical zonation: (from top to bottom): 1) laterite horizon or ferruginous cap rock (ferricrete); 2) loose ochreous weathering products (ochre zone or regolith zone); 3) secondary phosphate enrichment horizon (apatite-rich zone); 4) unevenly altered disintegrated carbonatites. The bulk contents of CaO and P₂O₅ increases from primary carbonatite upwards due to the crystallization of carbonate-fluorapatite, and decrease in near-surface horizon due to the high leaching of rocks. The total REE₂O₃ content is highest in the rocks of the regolith zone and the enriched zone due to the redistribution and concentration of these elements both in the form of impurities and as a result of the redeposition of their own mineral phases. The variations in compositions of apatite, pyrochlore and monazite from different horizons of laterite profile were chosen as a proxy for the REE distribution in the supergene zone. Fluorapatite is gradually being replaced by hydroxyl-fluorapatite and carbonate-hydroxyl-fluorapatite, as the degree of weathering of primary carbonatites increases. Pyrochlore-Ca-Na-F is being replaced by the (Ba<img>Sr)-variety (with Ce and Pb) during the same processes. Monazite changes from (Ce)-dominated to (La)-dominated. This is influenced by the oxidation factor, the consequence of which is also the formation of cerianite in the upper horizon. The study of the mineral and geochemical features of such an object as Chuktukon provides an improved understanding of the patterns of redistribution of the main and ore components of carbonatites in the supergene zone and can be useful in the study of other supergene deposits and delineation of dispersion halos of ore components.</div></div>","PeriodicalId":55973,"journal":{"name":"Chemie Der Erde-Geochemistry","volume":"84 4","pages":"Article 126210"},"PeriodicalIF":2.6,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143140520","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Precipitation of CaCO3 in natural and man-made aquatic environments - Mechanisms, analogues, and proxies","authors":"Martin Dietzel,&nbsp;Ronny Boch","doi":"10.1016/j.chemer.2024.126206","DOIUrl":"10.1016/j.chemer.2024.126206","url":null,"abstract":"<div><div>Calcium carbonate precipitates occur in diverse and widespread aquatic environments, both from marine to terrestrial natural surroundings and from anthropogenic to engineered settings. In aquatic media, calcium and carbonate ions react to form hydrous and anhydrous CaCO₃-containig solids in types of amorphous and crystalline phases, with calcite, aragonite and dolomite being the most frequent carbonate minerals in the Earth's crust. Understanding the CaCO₃ formation kinetics and mechanisms is key to exploring and evaluating individual aspects of biomineralization, speleothem and travertine growth, lime−/dolostone genesis, diagenetic pathways, climate reconstruction, CO₂ sequestration, tailored synthesis, unwanted scaling, mediated cementation, (geo)chemical forensics, etc. Although the literature dealing with CaCO₃ abundance, formation conditions, reaction mechanisms, nano- and macrostructures, elemental and isotope proxies is extensive, a comprehensive and up-to-date review of the highly diverse environments of CaCO₃ precipitation from aqueous media is valuable due to the tremendous increase in our knowledge about (i) CaCO₃ formation settings, analogues, reaction dynamics, and (potential) applications, (ii) novel high-precision, temporally and spatially highly-resolved analytical techniques, and (iii) monitoring, tracing and modeling tools. The present review on the formation of CaCO₃ in natural and man-made aquatic environments focuses on a systematic compilation and assessment of (i) reaction kinetics, mechanisms, and pathways of carbonate precipitation, (ii) analytical tools and proxies for tracking and reconstructing the solid-fluid-gas interactions and physico(<em>bio</em>)chemical environmental conditions during CaCO₃ precipitation, and (iii) natural and man-made analogs for the precipitation settings. The present review approach is not encyclopedic, but aims at a diverse array of environmental settings, the complex and coupled reaction pathways, state-of-the-art elemental and isotopic environmental proxies, and novel monitoring and modeling concepts for CaCO₃ precipitation.</div></div>","PeriodicalId":55973,"journal":{"name":"Chemie Der Erde-Geochemistry","volume":"84 4","pages":"Article 126206"},"PeriodicalIF":2.6,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143140413","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Detrital zircon and rutile U–Pb dating of garnet-mica schist in the Istranca (Strandja) Massif (NW Türkiye): Mineral chemistry and metamorphic conditions","authors":"Ersin Kaygısız ,&nbsp;Namık Aysal ,&nbsp;Kıymet Deniz Yağcıoğlu","doi":"10.1016/j.chemer.2024.126172","DOIUrl":"10.1016/j.chemer.2024.126172","url":null,"abstract":"<div><div>The Strandja Massif is one of the important metamorphic Massifs of Türkiye. The Strandja Massif consists of metasedimentary, and metamagmatic rocks such as migmatite, ortho- and paragneiss, metagranite, garnet-mica-schist, calcschist and marbles. For the first time, mineral chemistry analyzes of garnet-mica schists were performed to better understand the degree of metamorphism, and possible P-T conditions were defined in the north-northwest of the Strandja Massif. Based on almandine garnet and Fe-biotite mineral chemistry, garnet bearing mica-schist are affected by lower amphibolite facies metamorphism under 6.1 kbar pressure and approximately 547 °C. Moreover, detrital zircon U<img>Pb dating was performed on 107 points to describe the maximum depositional age of this garnet-mica schist. The youngest zircon ²⁰⁶Pb/²³⁸U age obtained as 198 ± 17 Ma (98 % concordant, Th/U = 0.74) and the oldest ²⁰⁶Pb/²³⁸U age yielded as 1896.8 ± 33.4 Ma (100.12 % compatible, Th/U = 0.63). Additionally, a zircon U<img>Pb concordia protolith age of 302.33 ± 1.79 Ma was obtained from orthogneisses. In-situ rutile-in garnet U<img>Pb dating shows that garnet-mica schists underwent metamorphism at 119 ± 49 Ma. P-T conditions and age of metamorphism show that these rocks effected by the lower amphibolite facies metamorphism during late Jurassic – early Cretaceous period.</div></div>","PeriodicalId":55973,"journal":{"name":"Chemie Der Erde-Geochemistry","volume":"84 4","pages":"Article 126172"},"PeriodicalIF":2.6,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142255628","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"3D mineral prospectivity modeling using deep adaptation network transfer learning: A case study of the Xiadian gold deposit, Eastern China","authors":"Jin Chen ,&nbsp;Xin Zuo ,&nbsp;Zhankun Liu ,&nbsp;Liqun Jiang ,&nbsp;Yuezhi Li ,&nbsp;Zhengkai Fu ,&nbsp;Hao Deng ,&nbsp;Xiancheng Mao","doi":"10.1016/j.chemer.2024.126189","DOIUrl":"10.1016/j.chemer.2024.126189","url":null,"abstract":"<div><div>Mineralization distribution is commonly heterogeneous in space due to the various geology and structure conditions or fluid flow and evoluation, causing the possible difference in mineralization distribution regularity between shallow and deep at a deposit. This presents important challenges for three-dimensional mineral prospectivity modeling (3D MPM) for deep zones with the scarcity data conditions. Transfer learning has shown promising generalization performance in tasks involving shifts in data distribution, reducing reliance on labeled samples and enhancing learning capability with limited data. In this study, we propose an approach of 3D MPM, namely DAN-CBAM, based on the deep adaptation network (DAN) augmented with the convolutional block attention module (CBAM). It theoretically can harmonize the distribution of ore-controlling features between shallow and deep zones of deposits while effectively extract critical high-dimensional features and spatial patterns. The Xiadian orogenic gold deposit was selected as a case study to validate the approach. The DAN integrates multi-layer, multi-kernel adaptation at the top layer of the CNN, resulting in improved alignment of marginal distributions across domains. Metrics for distribution similarity such as Wasserstein distance, were reduced by 0.250 and KL divergence decreased by 0.032. Additionally, the inclusion of the CBAM module led to a lower MK-MMD loss and a faster convergence rate. And the DAN-CBAM model achieves superior prediction accuracy (0.85) compared to traditional deep neural network (DNN) models (0.81). These highlight CBAM's effectiveness in enhancing the model's ability to capture spatial similarities in mineralization. Furthermore, the area under the curve (AUC) value of the DAN-CBAM model (0.869) significantly outperforms traditional machine learning methods, including CNN (0.786) and Random Forest (0.703) models, underscoring its superior predictive efficiency in 3D MPM for deep mineralization. Therefore, the proposed DAN-CBAM model is promising to be applied in the 3D MPM, in particular for the deposits with different mineralization distribution regularities in space.</div></div>","PeriodicalId":55973,"journal":{"name":"Chemie Der Erde-Geochemistry","volume":"84 4","pages":"Article 126189"},"PeriodicalIF":2.6,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143140137","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}