مدلسازی پتانسیل معدنی با به ‌کارگیری شبکه خودرمزنگار عمیق در پهنه اکتشافی دهسلم، شرق ایران

نوع مقاله: مقاله پژوهشی

نویسندگان

1 دانشکده مهندسی معدن و متالورژی، دانشگاه یزد

2 مرکز مطالعات مهندسی منابع، موسسه تکنولوژی بمبئی هند، بمبئی، هندوستان

10.29252/anm.2020.12606.1412

چکیده

شناسایی نواحی مستعد مرتبط با کانی­سازی و تلفیق مجموعه داده­های چندمنبعی اکتشافی در مدلسازی پتانسیل معدنی ضروری است. در این پژوهش، از روش تحلیل داده­های حجیم و یک الگوریتم یادگیری عمیق بدون ناظر، جهت شناسایی اهداف اکتشافی مرتبط با کانی­سازی مس- طلای پورفیری در پهنه اکتشافی دهسلم، شرق ایران استفاده شده است. بر اساس شرایط زمین­شناسی و تشکیل این تیپ کانی­سازی، در این پژوهش 32 متغیر ورودی، شامل داده­های زمین­شناسی (لیتولوژی و ساختاری)، سنجش از دور (دگرسانی­های آرژیلیک و اکسید آهن)، آنالیز 27 عنصر ژئوشیمی رسوبات آبراهه­ای و نقشه برگردان به قطب مغناطیس هوابرد جهت مدلسازی اکتشافی مس و طلای پورفیری با به کارگیری الگوریتم شبکه خودرمزنگار عمیق، استفاده و نتایج حاصل با خروجی مدل میانگین هندسی مقایسه گردید. ارزیابی عملکرد مدل­های پتانسیل معدنی تولیدشده با استفاده از نمودار آهنگ پیش­بینی- مساحت بهبودیافته بررسی شد. نتایج حاصل از این پژوهش نشان دهنده عملکرد مناسب مدل تولیدشده به روش خودرمزنگار عمیق، در شناسایی اهداف اکتشاف جهت برنامه­ریزی فعالیت‌های اکتشافی تفصیلی است. خروجی مدل تولید شده منجر به شناسایی اهداف اکتشافی جدیدی در قسمت­های شرق، شمال، غرب و جنوب غرب منطقه مورد مطالعه شده است. نتیجه این پژوهش، نشان دهنده پتانسیل روش­های مبتنی برتحلیل داده‌های حجیم و یادگیری عمیق در مدلسازی پتانسیل معدنی است.

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

Mineral Potential Modeling Using Deep Learning Auto-Encoder Network in the Dehsalm District, Eastern Iran

نویسندگان [English]

  • majid keykhay hosseinpoor 1
  • Amir hossein Kohsary 1
  • Amin Hossein Morshedy 1
  • Alok Porwal 2
1 Dept. of Mining and Metallurgy, Yazd University, Yazd, Iran
2 Center of studies in resource engineering, Indian institute of technology Bombay, Mumbai, India
چکیده [English]

Summary
Identification of promising areas associated with mineralization and integration of exploratory multi-resource data-sets are essential in mineral potential modeling. In this research, big data analysis method and an unsupervised deep auto-encoder network algorithm were used to identify the exploratory targets areas associated with porphyry copper-gold mineralization in the Dehsalm strict of Iran. The results show that the identified exploratory target areas have strong spatial relationships with known mineral indices in the study area. The Prediction-Area (P_A) plot analysis shows that the generated model performs well. The result of this study demonstrates that big data analytics supported by deep learning methods is a potential technique to be considered for use in mineral prospectivity mapping.
 
Introduction
New theories and analytical methods are required for mapping, interpreting and integrating diverse geo-information to increase the success rate and reduce the costs of mineral exploration. Particularly, as the amounts of high-quality data from multiple sources covering a broad range of scales have recently become readily available. The massive collection of earth observation data presents an unprecedented opportunity to apply big data approaches to solving problems in the geosciences.  The significance of applying big data approaches to mineral exploration is not only to generate a variety of anomaly maps using different kinds of big data, but also to identify the statistical and spatial characteristics of distribution, enrichment and depletion of metallogenic elements. The core function of big data analytics is a prediction, making it an ideal approach in mineral potential modeling.
 
Methodology and Approaches
In this study, big data analytics and a deep auto-encoder network were used to learn and mine meaningful patterns from massive amounts of input data for mapping mineral prospectivity in the Dehsalm strict of eastern Iran. This study aims to demonstrate the effectiveness of big data analytics and deep learning methods for mapping mineral prospectivity in this area.
 
Results and Conclusions
The case study of mapping porphyry Cu-Au mineralization in the Dehsalm strict of Iran demonstrates the effectiveness of big data analytics and deep learning algorithms for mineral prospectivity mapping. The output of the generated model predicted new exploratory target areas in the north, west and southwest parts of the study area.

کلیدواژه‌ها [English]

  • Mineral potential modeling
  • Big data
  • Deep auto-encoder network
  • Porphyry Cu-Au deposits
  • Dehsalm

شناسایی اهداف اکتشافی تیپ خاص کانی‌سازی مورد جستجو در مقیاس ناحیه­ای، با به ‌کارگیری و تحلیل داده‌های اکتشافی تحت عنوان مدلسازی پتانسیل معدنی (MPM[i]) توسعه یافته است[1]. وجود حجم بالای داده‌های اکتشافی با کیفیت بالا که طیف وسیعی از مقیاس‌ها را پوشش می­دهند (شامل داده‌های زمین‌شناسی، ژئوفیزیکی، ژئوشیمیایی و سنجش از دور)، سبب ناتوانی روش‌های آماری مرسوم در کشف روابط حاکم بر فضای ورودی داده­ها شده است[2]. بنابراین توسعه و به ‌کارگیری نظریه‌ها و روش­های تحلیلی جدید، برای پتانسیل‌یابی، تفسیر و تلفیق اطلاعات اکتشافی متنوع برای افزایش میزان موفقیت و کاهش هزینه­های اکتشاف مواد معدنی ضروری است[3، 4]. بر اساس تعریف، اصطلاح " داده‌های حجیم[ii]" نه تنها به معنای مجموعه داده‌های با حجم بالا، بلکه به خصوصیات ذاتی داده­ها با انواع مختلف فرمت‌ها و منابع داده اشاره دارد[5]. این تعریف در ارتباط با داده­های مورد استفاده در حوزه علوم زمین، ذاتی و شناخته شده است[6، 7].



[i] Mineral potential Mapping

[ii] Big data

[1]           G. F. Bonham-Carter, 1994, “Geographic information systems for geoscientists-modeling with GIS,” Comput. methods Geosci., vol. 13, p. 398.
[2]           V. F. Rodriguez-Galiano, M. Chica-Olmo, and M. Chica-Rivas, 2014, “Predictive modelling of gold potential with the integration of multisource information based on random forest: a case study on the Rodalquilar area, Southern Spain,” Int. J. Geogr. Inf. Sci., vol. 28, no. 7, pp. 1336–1354.
[3]           E. J. M. Carranza and A. G. Laborte, 2015, “Random forest predictive modeling of mineral prospectivity with small number of prospects and data with missing values in Abra (Philippines),” Comput. Geosci., vol. 74, pp. 60–70.
[4]           Y. Gao, Z. Zhang, Y. Xiong, and R. Zuo, 2016, “Mapping mineral prospectivity for Cu polymetallic mineralization in southwest Fujian Province, China,” Ore Geol. Rev., vol. 75, pp. 16–28.
[5]           H. Hu, Y. Wen, T. S. Chua, and X. Li, 2014, “Toward scalable systems for big data analytics: A technology tutorial,” IEEE Access, doi: 10.1109/ACCESS.2014.2332453.
[6]           H. Moeini and F. M. Torab, 2017, “Comparing compositional multivariate outliers with autoencoder networks in anomaly detection at Hamich exploration area, east of Iran,” J. Geochemical Explor., doi: 10.1016/j.gexplo.2017.05.008.
[7]           Y. Xiong, R. Zuo, and E. J. M. Carranza, 2018, “Mapping mineral prospectivity through big data analytics and a deep learning algorithm,” Ore Geol. Rev., doi: 10.1016/j.oregeorev.2018.10.006.
[8]           M. F. Goodchild, 2008, “The use cases of digital earth,” Int. J. Digit. Earth, doi: 10.1080/17538940701782528.
 
 
[9]           M. Deng and L. Di, 2009, “Building an online learning and research environment to enhance use of geospatial data,” …Journal Spat. Data Infrastructures Res., doi: 10.2902/1725-0463.2009.04.art4.
[10]         A. Gore, 1998, “The Digital Earth: Understanding our planet in the 21st Century,” Open GIS Consortium.
[11]         R. Zuo and Y. Xiong, 2018, “Big Data Analytics of Identifying Geochemical Anomalies Supported by Machine Learning Methods,” Nat. Resour. Res., doi: 10.1007/s11053-017-9357-0.
[12]         C. Wang, X. Ma, J. Chen, and J. Chen, 2018, “Information extraction and knowledge graph construction from geoscience literature,” Comput. Geosci., doi: 10.1016/j.cageo.2017.12.007.
[13]         V. Mayer-Schönberger and K. Cukier, 2013, Big Data: A Revolution That Will Transform How We Live, Work, and Think.
[14]         X.-X. Liu, Y.-C. Chen, D.-H. Wang, F. Huang, and Z. Zhao, 2016, “The metallogenic geomorphic condition analysis of the ion-absorbing type rare earths ore in the Eastern Nanling region based on DEM data,” Acta Geosci. Sin., doi: 10.3975/cagsb.2016.02.05.
[15]         J. M. Luo et al., 2017, “Application of integrated geophysical and geochemical data processing to metallogenic target zone quantitative prediction and optimization,” Bull. Mineral. Petrol. Geochemistry, vol. 36, no. 6, pp. 886–891.
[16]         M. Chen, S. Mao, and Y. Liu, 2014, “Big data: A survey,” in Mobile Networks and Applications, doi: 10.1007/s11036-013-0489-0.
[17]         A. Porwal, E. J. M. Carranza, and M. Hale, 2003, “Artificial neural networks for mineral-potential mapping: a case study from Aravalli Province, Western India,” Nat. Resour. Res., vol. 12, no. 3, pp. 155–171.
[18]         D. Harris and G. Pan, 1999, “Mineral favorability mapping: a comparison of artificial neural networks, logistic regression, and discriminant analysis,” Nat. Resour. Res., vol. 8, no. 2, pp. 93–109,.
[19]         H.-J. Oh and S. Lee, 2010, “Application of artificial neural network for gold--silver deposits potential mapping: a case study of Korea,” Nat. Resour. Res., vol. 19, no. 2, pp. 103–124,.
[20]         M. Abedi, G.-H. Norouzi, and A. Bahroudi, 2012, “Support vector machine for multi-classification of mineral prospectivity areas,” Comput. Geosci., vol. 46, pp. 272–283,.
 
[21]         M. Shabankareh and A. Hezarkhani, 2017, “Application of support vector machines for copper potential mapping in Kerman region, Iran,” J. African Earth Sci., vol. 128, pp. 116–126,.
[22]         S. Hariharan, S. Tirodkar, A. Porwal, A. Bhattacharya, and A. Joly, 2017, “Random forest-based prospectivity modelling of greenfield terrains using sparse deposit data: An example from the Tanami Region, Western Australia,” Nat. Resour. Res., vol. 26, no. 4, pp. 489–507,.
[23]         G. LeCun, Y., Bengio, Y., Hinton, 2015, “Deep learning. nature 521 (7553): 436,” Nat. 521 (7553), 436–444., doi: 10.1038/nature14539.
[24]         Y. Chen, Z. Lin, X. Zhao, G. Wang, and Y. Gu, 2014, “Deep learning-based classification of hyperspectral data,” IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., doi: 10.1109/JSTARS.2014.2329330.
[25]         L. Zhang, L. Zhang, and B. Du, 2016, “Deep learning for remote sensing data: A technical tutorial on the state of the art,” IEEE Geosci. Remote Sens. Mag., doi: 10.1109/MGRS.2016.2540798.
[26]         N. Kussul, M. Lavreniuk, S. Skakun, and A. Shelestov, 2017, “Deep Learning Classification of Land Cover and Crop Types Using Remote Sensing Data,” IEEE Geosci. Remote Sens. Lett., doi: 10.1109/LGRS.2017.2681128.
[27]         A. P. Valentine and J. Trampert, 2012, “Data space reduction, quality assessment and searching of seismograms: Autoencoder networks for waveform data,” Geophys. J. Int., doi: 10.1111/j.1365-246X.2012.05429.x.
[28]         Z. E. Ross, M. A. Meier, and E. Hauksson, 2018, “P Wave Arrival Picking and First-Motion Polarity Determination With Deep Learning,” J. Geophys. Res. Solid Earth, doi: 10.1029/2017JB015251.
[29]         Y. Chen, 2015, “Mineral potential mapping with a restricted Boltzmann machine,” Ore Geol. Rev., vol. 71, pp. 749–760,.
[30]         Y. Chen and W. Wu, 2018, “Isolation Forest as an Alternative Data-Driven Mineral Prospectivity Mapping Method with a Higher Data-Processing Efficiency,” Nat. Resour. Res., pp. 1–16,.
[31]         M. Karimpour and others, 2011, “Review of age, Rb-Sr geochemistry and petrogenesis of Jurassic to Quaternary igneous rocks in Lut Block, Eastern Iran,” Geopersia, vol. 1, no. 1, pp. 19–54,.
[32]         A. M. Shafaroudi and M. H. Karimpour, 2013, “Hydrothermal alteration mapping in northern Khur, Iran, using ASTER image processing: a new insight to the type of copper mineralization,” Acta Geol. Sin. Ed., vol. 87, no. 3, pp. 830–842,.
[33]         R. M. Beydokhti, M. H. Karimpour, and S. A. Mazaheri, 2014, “Studies of remote sensing, geology, alteration, mineralization and geochemistry of Balazard copper-gold prospecting area, west of Nehbandan,” Iran. J. Crystallogr. Mineral., vol. 23, no. 3, pp. 459–470,.
[34]         R. M. Beydokhti, M. H. Karimpour, S. A. Mazaheri, J. F. Santos, and U. Klötzli, 2015, “U--Pb zircon geochronology, Sr--Nd geochemistry, petrogenesis and tectonic setting of Mahoor granitoid rocks (Lut Block, Eastern Iran),” J. Asian Earth Sci., vol. 111, pp. 192–205,.
[35]         A. M. Shafaroudi, M. H. Karimpour, and C. R. Stern, 2015, “The Khopik porphyry copper prospect, Lut Block, Eastern Iran: geology, alteration and mineralization, fluid inclusion, and oxygen isotope studies,” Ore Geol. Rev., vol. 65, pp. 522–544,.
[36]         M. A. Akrami and N. Naderi Mighan, 2005, Geological map of Dehsalm(1:100,000). Geological Survey of Iran,.
[37]         G. E. Hinton, S. Osindero, and Y.-W. Teh, 2006, “A fast learning algorithm for deep belief nets,” Neural Comput., vol. 18, no. 7, pp. 1527–1554,.
[38]         Y. Xiong and R. Zuo, 2018, “GIS-based rare events logistic regression for mineral prospectivity mapping,” Comput. Geosci., doi: 10.1016/j.cageo.2017.10.005.
[39]         G. E. Hinton, 2002, “Training products of experts by minimizing contrastive divergence,” Neural Comput., doi: 10.1162/089976602760128018.
[40]         G. Hinton, 2010, “A Practical Guide to Training Restricted Boltzmann Machines A Practical Guide to Training Restricted Boltzmann Machines,” Computer (Long. Beach. Calif)., doi: 10.1007/978-3-642-35289-8_32.
[41]         E. J. M. Carranza and M. Hale, 2002, “Where are porphyry copper deposits spatially localized? A case study in Benguet province, Philippines,” Nat. Resour. Res., vol. 11, no. 1, pp. 45–59,.
[42]         A. C. Philip, 2005, “Magmatic processes in the development of porphyry-type ore systems,” Econ. Geol., vol. 100, pp. 25–38,.
[43]         R. H. Sillitoe, 2010, “Porphyry copper systems,” Econ. Geol., vol. 105, no. 1, pp. 3–41,.
[44]         B. R. Berger, R. A. Ayuso, J. C. Wynn, and R. R. Seal, 2008, “Preliminary model of porphyry copper deposits,” US Geol. Surv. open-file Rep., vol. 1321, p. 55,.
 
[45]         D. R. Cooke, P. Hollings, and J. L. Walshe, 2005, “Giant porphyry deposits: characteristics, distribution, and tectonic controls,” Econ. Geol., vol. 100, no. 5, pp. 801–818,.
[46]         R. H. Sillitoe, 1972, “A plate tectonic model for the origin of porphyry copper deposits,” Econ. Geol., vol. 67, no. 2, pp. 184–197,.
[47]         R. H. Sillitoe, 2000, “Gold-rich porphyry deposits: descriptive and genetic models and their role in exploration and discovery,” Rev. Econ. Geol., vol. 13, pp. 315–345,.
[48]         E.-J. Holden, S. C. Fu, P. Kovesi, M. Dentith, B. Bourne, and M. Hope, 2011, “Automatic identification of responses from porphyry intrusive systems within magnetic data using image analysis,” J. Appl. Geophys., vol. 74, no. 4, pp. 255–262,.
[49]         Z. Hou, H. Zhang, X. Pan, and Z. Yang, 2011, “Porphyry Cu (--Mo--Au) deposits related to melting of thickened mafic lower crust: examples from the eastern Tethyan metallogenic domain,” Ore Geol. Rev., vol. 39, no. 1–2, pp. 21–45,.
[50]         M. Parsa, A. Maghsoudi, M. Yousefi, and M. Sadeghi, 2016, “Prospectivity modeling of porphyry-Cu deposits by identification and integration of efficient mono-elemental geochemical signatures,” J. African Earth Sci., vol. 114, pp. 228–241,.
[51]         B. Roshanravan, H. Aghajani, M. Yousefi, and O. Kreuzer, 2019, “Particle Swarm Optimization Algorithm for Neuro-Fuzzy Prospectivity Analysis Using Continuously Weighted Spatial Exploration Data,” Nat. Resour. Res., pp. 1–17,.
[52]         M. Yousefi and E. J. M. Carranza, 2016, “Data-driven index overlay and Boolean logic mineral prospectivity modeling in greenfields exploration,” Nat. Resour. Res., vol. 25, no. 1, pp. 3–18,.
[53]         V. Pawlowsky-Glahn and A. Buccianti, 2011, Compositional data analysis: Theory and applications. John Wiley & Sons,.
[54]         E. J. M. Carranza, 2008, Geochemical anomaly and mineral prospectivity mapping in GIS, vol. 11. Elsevier,.
[55]         M. Yousefi, A. Kamkar-Rouhani, and E. J. M. Carranza, 2012, “Geochemical mineralization probability index (GMPI): a new approach to generate enhanced stream sediment geochemical evidential map for increasing probability of success in mineral potential mapping,” J. Geochemical Explor., vol. 115, pp. 24–35,.
[56]         D. A. Clark, 2014, “Magnetic effects of hydrothermal alteration in porphyry copper and iron-oxide copper--gold systems: a review,” Tectonophysics, vol. 624, pp. 46–65,.
[57]         H. H. Asadi, A. Porwal, M. Fatehi, S. Kianpouryan, and Y.-J. Lu, 2015, “Exploration feature selection applied to hybrid data integration modeling: Targeting copper-gold potential in central Iran,” Ore Geol. Rev., vol. 71, pp. 819–838,.
[58]         A. H. Ansari and K. Alamdar, 2009, “Reduction to the pole of magnetic anomalies using analytic signal,” World Appl. Sci. J., vol. 7, no. 4, pp. 405–409,.
[59]         A. R. Bansal, G. Gabriel, V. P. Dimri, and C. M. Krawczyk, 2011, “Estimation of depth to the bottom of magnetic sources by a modified centroid method for fractal distribution of sources: An application to aeromagnetic data in Germany,” Geophysics, vol. 76, no. 3, pp. L11--L22,.
[60]         M. H. Zadeh, M. H. Tangestani, F. V. Roldan, and I. Yusta, 2014, “Sub-pixel mineral mapping of a porphyry copper belt using EO-1 Hyperion data,” Adv. Sp. Res., vol. 53, no. 3, pp. 440–451,.
[61]         M. Yousefi and E. J. M. Carranza, 2017, “Union score and fuzzy logic mineral prospectivity mapping using discretized and continuous spatial evidence values,” J. African Earth Sci., vol. 128, pp. 47–60,.
[62]         H. Ranjbar, F. Masoumi, and E. J. M. Carranza, 2011, “Evaluation of geophysics and spaceborne multispectral data for alteration mapping in the Sar Cheshmeh mining area, Iran,” Int. J. Remote Sens., doi: 10.1080/01431161003745665.
[63]         J. R. Gozzard, 2006, Image processing of ASTER multispectral data. Geological Survey of WA,.
[64]         T. Hengl, 2006, “Finding the right pixel size,” Comput. Geosci., vol. 32, no. 9, pp. 1283–1298,.
[65]         M. Yousefi and E. J. M. Carranza, 2015, “Fuzzification of continuous-value spatial evidence for mineral prospectivity mapping,” Comput. Geosci., vol. 74, pp. 97–109,.
[66]         M. Yousefi and E. J. M. Carranza, 2015, “Geometric average of spatial evidence data layers: A GIS-based multi-criteria decision-making approach to mineral prospectivity mapping,” Comput. Geosci., doi: 10.1016/j.cageo.2015.07.006.
[67]         A. Porwal, E. J. M. Carranza, and M. Hale, 2006, “Bayesian network classifiers for mineral potential mapping,” Comput. Geosci., vol. 32, no. 1, pp. 1–16,.
[68]         B. Roshanravan, H. Aghajani, M. Yousefi, and O. Kreuzer, 2019, “An Improved Prediction-Area Plot for Prospectivity Analysis of Mineral Deposits,” Nat. Resour. Res., doi: 10.1007/s11053-018-9439-7.
[69]         M. Yousefi and E. J. M. Carranza, 2015, “Prediction--area (P--A) plot and C--A fractal analysis to classify and evaluate evidential maps for mineral prospectivity modeling,” Comput. Geosci., vol. 79, pp. 69–81,.
[70]         X. Glorot and Y. Bengio, 2010, “Understanding the difficulty of training deep feedforward neural networks,” in Proceedings of the 13th International Conference On Artificial Intelligence and Statistics,.
[71]         H. Larochelle, Y. Bengio, J. Louradour, and P. Lamblin, 2009, “Exploring Strategies for Training Deep Neural Networks,” J. Mach. Learn. Res.,.
[72]         D. E. Rumelhart, G. E. Hinton, and R. J. Williams, 1986, “Learning representations by back-propagating errors,” Nature, doi: 10.1038/323533a0.
[73]         L. Li and Y. Wang, 2014, “What drives the aerosol distribution in Guangdong - The most developed province in Southern China?,” Sci. Rep., doi: 10.1038/srep05972.
[74]         J. Vesanto, 1999, “SOM-based data visualization methods,” Intell. Data Anal., doi: 10.3233/IDA-1999-3203.
[75]         A. Joly, A. Porwal, and T. C. McCuaig, 2012, “Exploration targeting for orogenic gold deposits in the Granites-Tanami Orogen: Mineral system analysis, targeting model and prospectivity analysis,” Ore Geol. Rev., doi: 10.1016/j.oregeorev.2012.05.004.
[76]         M. M. Najafabadi, F. Villanustre, T. M. Khoshgoftaar, N. Seliya, R. Wald, and E. Muharemagic, 2015, “Deep learning applications and challenges in big data analytics,” J. Big Data, doi: 10.1186/s40537-014-0007-7.