Study of ore deposits by the dynamic systems investigation methods: 2. Clustering of ore deposits and interpretation of the results

详细信息    查看全文

• 作者：M. V. Rodkin ; A. R. Shatakhtsyan
• 关键词：ore deposits ; Tanimoto similarity measure ; correlation dimension ; clustering of ore deposits
• 刊名：Izvestiya, Physics of the Solid Earth
• 出版年：2015
• 出版时间：May 2015
• 年：2015
• 卷：51
• 期：3
• 页码：428-436
• 全文大小：240 KB
• 参考文献：Frank, S., Stiller, H., and Seifert, W., Effects of high temperature phase transitions on chemical differentiation, in High Pressure Investigations in Geosciences, Berlin: Akademie-Verlag, 1989, pp. 195鈥?03.
  Hentschel, H.G.E. and Procaccia, I., The infinite number of generalized dimensions of fractals and strange attractors, Phys. D (Amsterdam, Neth.), 1983, vol. 8, no. 3, pp. 435鈥?44.View Article
  Largest Mineral Deposits of the World, CD-ROM, Commission on Geological Map of the World, 2006.
  Mandel, I.D., Klasternyi analiz (Cluster Analysis), Moscow: Finansy i statistika, 1988.
  Rodkin, M.V., Zotov, I.A., Graeva, E.M., Labuntsova, L.M., and Shatakhtsyan, A.R., Power distribution in ore and oil genesis鈥攊nterpretation and generating mechanisms, Rus. J. Earth Sci., 2010, vol. 11, no. 3. doi 10.2205/2009ES000408
  Rodkin, M.V. and Shatakhtsyan, A.R., Statistical analysis of catalogs of large and superlarge ore deposits: empirical regularities and their interpretation, Geoinformatika, 2013, no. 4, pp. 25鈥?2.
  Rodkin, M.V. and Shatakhtsyan, A.R., Study of Ore Deposits by the Dynamic Systems Investigation Methods: 1. Calculation of the Correlation Dimension, Izv., Phys. Solid Earth, 2015, vol. 51, no. 3, pp. 419鈥?27.View Article
  Rundkvist, D.V., Tkachev, A.V., Cherkasov, S.V., et al., Krupnye i superkrupnye mestorozhdeniya rudnykh poleznykh iskopaemykh. V 3 tomakh (Large and Superlarge Ore Deposits, 3 vols.), Moscow: IGEM RAN, 2006.
  Sobolev, P.O. and Rundkvist, D.V., Statistical analysis of the spatial distribution of large and superlarge minearl deposits, in Krupnye i superkrupnye mestorozhdeniya rudnykh poleznykh iskopaemykh, T. 1 (Large and Superlarge Ore Deposits, vol. 1), Moscow: IGEM RAN, 2006, pp. 177鈥?06.
  Taylor, S.R. and McLennan, S.M., The Continental Crust: Its Composition and Evolution, Oxford: Blackwell, 1985.
  Tikhotskii, S.A., Planetary distribution of the large and superlarge mineral deposits in the anomalous gravity and magnetic fields of the Earth, in Krupnye i superkrupnye mestorozhdeniya rudnykh poleznykh iskopaemykh, T. 1 (Large and Superlarge Ore Deposits, vol. 1), Moscow: IGEM RAN, 2006, pp. 149鈥?76.
  Tkachev, A.V., Global analysis of spatiotemporal pattern of the formation and layout of the large and superlarge deposits, in Krupnye i superkrupnye mestorozhdeniya rudnykh poleznykh iskopaemykh. T. 1 (Large and Superlarge Ore Deposits, vol. 1), Moscow: IGEM RAN, 2006, pp. 49鈥?48.
  
• 作者单位：M. V. Rodkin (1) (2)
  A. R. Shatakhtsyan (3)
  
  1. Institute of Earthquake Prediction Theory and Mathematical Geophysics, Russian Academy of Sciences, ul. Profsoyuznaya 84/32, Moscow, 117997, Russia
  2. Institute of Marine Geology and Geophysics, Far East Branch, Russian Academy of Sciences, ul. Nauki 1b, Yuzhno-Sakhalinsk, 693022, Russia
  3. Geophysical Center, Russian Academy of Sciences, ul. Molodezhnaya 3, Moscow, 119296, Russia
  
• 刊物主题：Geophysics/Geodesy;
• 出版者：Springer US
• ISSN：1555-6506

文摘

Ore deposits are complex natural objects. Each deposit is unique in its own way. At the same time, efficient data processing for a large number of the deposits and better understanding of the ore formation processes require intensional clustering of the deposits. A series of schemes exists for clustering (classifying) the deposits depending on the set of the contained ore components and other parameters. We discuss the new formal ways for clustering the ore deposits based on the Tanimoto similarity measure (for the data on the set of the contained metal components) and on the mixed correlation dimension (for the data on the relative spatial locations of the deposits). We demonstrate the close correlation between the cross-similarity estimates calculated for different types of the deposits using the Tanimoto measure and mixed correlation dimension. We discuss the version of clustering the deposits based on the Tanimoto measure and suggest a tentative interpretation of a series of the formally identified statistical regularities. Some results of this clustering can be treated as an argument supporting the previous hypothesis which states that the formation of the large and extralarge ore deposits is a byproduct of the transformation of the material of the tectonosphere from one geochemical reservoir (upper continental crust, lower continental crust, oceanic crust, upper mantle, etc.) into another. The energy of these tectonic processes drives the massive negentgropy processes concentrating ore material into the deposits.