Hybrid geostatistical and deep learning framework for geochemical characterization in historical mine tailings

Keyumars Anvari; Jörg Benndorf; Gabriel Gerber; Uta Alisch

doi:10.1038/s41598-025-19441-5

journal article Open Access Oct 07, 2025

Hybrid geostatistical and deep learning framework for geochemical characterization in historical mine tailings

Keyumars Anvari Jörg Benndorf Gabriel Gerber Uta Alisch

Scientific Reports Vol. 15 No. 1 · Springer Science and Business Media LLC

View at Publisher Save 10.1038/s41598-025-19441-5

Abstract

Abstract
Sustainable mine tailings management has become a worldwide priority given increasing critical raw materials (CRMs) demand and growing environmental concerns. While these anthropogenic deposits are often enriched with useful metals, they may also contain hazardous substances and thus provide both opportunities for resource recovery and environmental risk. In this work a hybrid geostatistical–deep learning framework was established to model geochemical distribution in old tailings. This study integrates ordinary kriging (OK) with a one-dimensional convolutional neural network and a bidirectional long short-term memory model (1D CNN and BiLSTM). The hybrid relies exclusively on features derived from the OK spatial covariance structure, computed from covariance matrices over the sampled locations, to inform the deep model and enhance prediction accuracy. The framework, applied to a historical tailings site, significantly outperformed traditional geostatistical methods as it can provide high-resolution predictions across all points of interest, while accounting for spatial heterogeneity. These results highlight the applicability of this strategy in sustainable resource recovery and environmental remediation, in accordance with circular economy concepts.

Topics

No keywords indexed for this article. Browse by subject →

References

79

[1]

ICA. Copper - The Pathway To Net Zero41 (International Copper Association, 2023).

[2]

Rock-to-Metal Ratio: A Foundational Metric for Understanding Mine Wastes

Nedal T. Nassar, Graham W. Lederer, Jamie L. Brainard et al.

Environmental Science & Technology 2022 10.1021/acs.est.1c07875

[3]

Muniruzzaman, M., Karlsson, T., Ahmadi, N. & Rolle, M. Multiphase and multicomponent simulation of acid mine drainage in unsaturated mine waste: modeling approach, benchmarks and application examples. Appl. Geochem. 120, 104677 (2020). 10.1016/j.apgeochem.2020.104677

[4]

Blowes, D., Ptacek, C., Jambor, J. & Weisener, C. The geochemistry of acid mine. Environ. Geochem. 9, 149 (2005).

[5]

Rezaie, B. & Anderson, A. Sustainable resolutions for environmental threat of the acid mine drainage. Sci. Total Environ. 717, 137211 (2020). 10.1016/j.scitotenv.2020.137211

[6]

Aryafar, A., Gholami, R., Rooki, R. & Doulati Ardejani, F. Heavy metal pollution assessment using support vector machine in the Shur river, Sarcheshmeh copper mine, Iran. Environ. Earth Sci. 67, 1191–1199 (2012). 10.1007/s12665-012-1565-7

[7]

Mumford, K. A., Pitt, B., Townsend, A. T., Snape, I. & Gore, D. B. Long-term acid-generating and metal leaching potential of a sub-Arctic oil shale. Minerals 4, 293–312 (2014). 10.3390/min4020293

[8]

Nordstrom, D. K. Mine waters: acidic to circmneutral. Elements 7, 393–398 (2011). 10.2113/gselements.7.6.393

[9]

Toubri, Y. et al. Merging 3D geological modeling and stochastic simulation to foster waste rock upstream management. J. Geochem. Explor. 224, 106739 (2021). 10.1016/j.gexplo.2021.106739

[10]

Zhang, C., Ma, L. & Liu, W. A. Machine learning approach for prediction of the quantity of mine waste rock drainage in areas with spring freshet. Minerals 13, 376 (2023). 10.3390/min13030376

[11]

Kinnunen, P., Karhu, M., Yli-Rantala, E., Kivikytö-Reponen, P. & Mäkinen, J. A review of circular economy strategies for mine tailings. Clean. Eng. Technol. 8, 100499 (2022). 10.1016/j.clet.2022.100499

[12]

Araya, N., Ramírez, Y., Kraslawski, A. & Cisternas, L. A. Feasibility of re-processing mine tailings to obtain critical Raw materials using real options analysis. J. Environ. Manage. 284, 112060 (2021). 10.1016/j.jenvman.2021.112060

[13]

Blannin, R., Frenzel, M., Tolosana-Delgado, R., Büttner, P. & Gutzmer J. 3D Geostatistical modelling of a tailings storage facility: resource potential and environmental implications. Ore. Geol. Reviews, 105337 (2023). 10.1016/j.oregeorev.2023.105337

[14]

Gómez-Arias, A. et al. Mine waste from carbonatite deposits as potential rare Earth resource: insight into the Phalaborwa (Palabora) complex. J. Geochem. Explor. 232, 106884 (2022). 10.1016/j.gexplo.2021.106884

[15]

Zhang, Y. et al. Extraction of lithium and aluminium from bauxite mine tailings by mixed acid treatment without roasting. J. Hazard. Mater. 404, 124044 (2021). 10.1016/j.jhazmat.2020.124044

[16]

Ghazi, A. B., Jamshidi-Zanjani, A. & Nejati, H. Utilization of copper mine tailings as a partial substitute for cement in concrete construction. Constr. Build. Mater. 317, 125921 (2022). 10.1016/j.conbuildmat.2021.125921

[17]

Simão, F. V. et al. Mine waste as a sustainable resource for facing bricks. J. Clean. Prod. 368, 133118 (2022). 10.1016/j.jclepro.2022.133118

[18]

Stubbs, A. R. et al. Direct measurement of CO2 drawdown in mine wastes and rock powders: implications for enhanced rock weathering. Int. J. Greenhouse Gas Control. 113, 103554 (2022). 10.1016/j.ijggc.2021.103554

[19]

Salom, A. T. Remining and Restructure of a Tailing Deposit-Technical Feasibility (Universidade do Porto, 2017).

[20]

Singo, N. & Kramers, J. Feasibility of tailings retreatment to unlock value and create environmental sustainability of the Louis Moore tailings dump near giyani, South Africa. J. South Afr. Inst. Min. Metall. 121, 361–367 (2021). 10.17159/2411-9717/1138/2021

[21]

Tripodi, E. E. M., Rueda, J. A. G., Céspedes, C. A., Vega, J. D. & Gómez, C. C. Characterization and Geostatistical modelling of contaminants and added value metals from an abandoned Cu–Au tailing dam in Taltal (Chile). J. S. Am. Earth Sci. 93, 183–202 (2019). 10.1016/j.jsames.2019.05.001

[22]

Wilson, R., Toro, N., Naranjo, O., Emery, X. & Navarra, A. Integration of Geostatistical modeling into discrete event simulation for development of tailings dam retreatment applications. Miner. Eng. 164, 106814 (2021). 10.1016/j.mineng.2021.106814

[23]

Soto, F. et al. Transitive kriging for modeling tailings deposits: A case study in Southwest Finland. J. Clean. Prod. 374, 133857 (2022). 10.1016/j.jclepro.2022.133857

[24]

Kasmaeeyazdi, S., Dinelli, E. & Braga, R. Mapping Co–Cr–Cu and Fe occurrence in a legacy mining waste using geochemistry and satellite imagery analyses. Appl. Sci. 12, 1928 (2022). 10.3390/app12041928

[25]

Karacan, C. Ö., Erten, O. & Martín-Fernández, J. A. Assessment of resource potential from mine tailings using Geostatistical modeling for compositions: A methodology and application to Katherine mine site, arizona, USA. J. Geochem. Explor. 245, 107142 (2023). 10.1016/j.gexplo.2022.107142

[26]

Lemos, M. et al. Geochemistry and mineralogy of auriferous tailings deposits and their potential for reuse in Nova Lima region, Brazil. Sci. Rep. 13, 4339 (2023). 10.1038/s41598-023-31133-6

[27]

Bai, T. & Tahmasebi, P. Accelerating Geostatistical modeling using geostatistics-informed machine learning. Comput. Geosci. 146, 104663 (2021). 10.1016/j.cageo.2020.104663

[28]

Jalloh, A. B., Kyuro, S., Jalloh, Y. & Barrie, A. K. Integrating artificial neural networks and geostatistics for optimum 3D geological block modeling in mineral reserve estimation: A case study. Int. J. Min. Sci. Technol. 26, 581–585 (2016). 10.1016/j.ijmst.2016.05.008

[29]

Furrer, R., Genton, M. G. & Nychka, D. Covariance tapering for interpolation of large Spatial datasets. J. Comput. Graphical Stat. 15, 502–523 (2006). 10.1198/106186006x132178

[30]

Katzfuss, M. & Guinness, J. A general framework for Vecchia approximations of Gaussian processes. Stat. Sci. 36, 124–141 (2021). 10.1214/19-sts755

[31]

Tadić, J. M., Qiu, X., Miller, S. & Michalak, A. M. Spatio-temporal approach to moving window block kriging of satellite data v1. 0. Geosci. Model Dev. 10, 709–720 (2017). 10.5194/gmd-10-709-2017

[32]

Van Stein, B., Wang, H., Kowalczyk, W., Emmerich, M. & Bäck, T. Cluster-based kriging approximation algorithms for complexity reduction. Appl. Intell. 50, 778–791 (2020). 10.1007/s10489-019-01549-7

[33]

Yamamoto, J. K. Correcting the smoothing effect of ordinary kriging estimates. Math. Geol. 37, 69–94 (2005). 10.1007/s11004-005-8748-7

[34]

Russell, S. J. & Norvig, P. Artificial intelligence: a modern approachPearson,. (2016).

[35]

SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation

Vijay Badrinarayanan, Alex Kendall, Roberto Cipolla

IEEE Transactions on Pattern Analysis and Machine... 2017 10.1109/tpami.2016.2644615

[36]

Karn, A. et al. Artificial intelligence in computer vision. Sustainable Development Through Machine Learning, AI and IOT: second, 102 (2024).

[37]

Li, X. & Shi, Y. in International Conference on Virtual Reality and Intelligent Systems (ICVRIS). 22–25 (IEEE, 2018).

[38]

Alivernini, S., Cañete, J. D., Bacardit, J. & Kurowska-Stolarska, M. Using explainable artificial intelligence to predict and forestall flare in rheumatoid arthritis. Nat. Med. 30, 925–926 (2024). 10.1038/s41591-024-02818-w

[39]

Theofilatos, K. et al. Predicting protein complexes from weighted protein–protein interaction graphs with a novel unsupervised methodology: evolutionary enhanced Markov clustering. Artif. Intell. Med. 63, 181–189 (2015). 10.1016/j.artmed.2014.12.012

[40]

Witten, J. et al. Artificial intelligence-guided design of lipid nanoparticles for pulmonary gene therapy. Nature Biotechnology, 1–10 (2024). 10.1038/s41587-024-02490-y

[41]

Anvari, K. et al. A hybrid recurrence analysis and wavelet transformation for rock boundary identification. Available SSRN 5049654 (2024). 10.2139/ssrn.5049654

[42]

Anvari, K., Mousavi, A., Sayadi, A. R., Sellers, E. & Salmi, E. F. Automatic detection of rock boundaries using a hybrid recurrence quantification analysis and machine learning techniques. Bull. Eng. Geol. Environ. 81, 398 (2022). 10.1007/s10064-022-02898-3

[43]

Bakay, A. & R. in SPE Annual Technical Conference and Exhibition? D021S025007 (SPE).

[44]

Fouedjio, F. & Arya, E. Locally varying Geostatistical machine learning for Spatial prediction. Artif. Intell. Geosci. 5, 100081 (2024).

[45]

Hill, E. J., Pearce, M. A. & Stromberg, J. M. Improving automated geological logging of drill holes by incorporating multiscale Spatial methods. Math. Geosci. 53, 21–53 (2021). 10.1007/s11004-020-09859-0

[46]

Maldonado-Cruz, E. & Pyrcz, M. J. Tuning machine learning dropout for subsurface uncertainty model accuracy. J. Petrol. Sci. Eng. 205, 108975 (2021). 10.1016/j.petrol.2021.108975

[47]

Kamrava, S., Sahimi, M. & Tahmasebi, P. Quantifying accuracy of stochastic methods of reconstructing complex materials by deep learning. Phys. Rev. E. 101, 043301 (2020). 10.1103/physreve.101.043301

[48]

Reconstruction of three-dimensional porous media using generative adversarial neural networks

Lukas Mosser, Olivier Dubrule, Martin J. Blunt

Physical Review E 2017 10.1103/physreve.96.043309

[49]

Niu, Y., Mostaghimi, P., Shabaninejad, M., Swietojanski, P. & Armstrong, R. T. Digital rock segmentation for petrophysical analysis with reduced user bias using convolutional neural networks.. Water Resources Research 56, e2019WR026597 (2020). 10.1029/2019wr026597

[50]

Anvari, K. & Benndorf, J. Real time Mining—A review of developments within the last decade. Mining 5, 38 (2025). 10.3390/mining5030038

Showing 50 of 79 references

Cited By

6

Explainable data-driven spatial predictions of soil organic carbon at the national level

Azamat Suleymanov, Evgeny Abakumov · 2026

Science of The Total Environment

Metrics

6

Citations

79

References

Details

Published: Oct 07, 2025
Vol/Issue: 15(1)
License: View

Authors

Funding

Technische Universität Bergakademie Freiberg

Cite This Article

Keyumars Anvari, Jörg Benndorf, Gabriel Gerber, et al. (2025). Hybrid geostatistical and deep learning framework for geochemical characterization in historical mine tailings. Scientific Reports, 15(1). https://doi.org/10.1038/s41598-025-19441-5

Hybrid geostatistical and deep learning framework for geochemical characterization in historical mine tailings

You May Also Like