Geophysics and Geophysical Exploration (지구물리와물리탐사)

Korean Society of Earth and Exploration Geophysicists (한국지구물리·물리탐사학회)
권호 표지
DOI QR코드

DOI QR Code

Analysis of Regional Potential Mapping Factors of Metal Deposits using Machine Learning

머신러닝을 이용한 광역 금속 광상 배태 잠재성 평가 인자 분석

  • Park, Gyesoon (Convergence Research Center for Development of Mineral Resources, Korea Institute of Geoscience and Mineral Resources)
  • 박계순 (한국지질자원연구원, DMR융합연구단)
  • Received : 2020.06.11
  • Accepted : 2020.08.24
  • Published : 2020.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

The genesis of ore bodies is a very diverse and complex process, and the target depth of mineral exploration increases. These create a need for predictive mineral exploration, which may be facilitated by the advancement of machine learning and geological database. In this study, we confirm that the faults and igneous rocks distributions and magnetic data can be used as input data for potential mapping using deep neural networks. When the input data are constructed with faults, igneous rocks, and magnetic data, we can build a potential mapping model of the metal deposit that has a predictive accuracy greater than 0.9. If detailed geological and geophysical data are obtained, this approach can be applied to the potential mapping on a mine scale. In addition, we confirm that the magnetic data, which provide the distribution of the underground igneous rock, can supplement the limited information from the surface igneous rock distribution. Therefore, rather than simply integrating various data sets, it will be more important to integrate information considering the geological correlation to genesis of minerals.

복잡하고 다양한 광상 생성 과정과 부존 위치의 심부화로 인하여 광물자원 탐사의 효율을 높일 수 있는 예측탐사의 필요성이 점차 높아지고 있다. 최근의 머신러닝 기법의 활용성 증가와 광역 지질 데이터베이스의 구축 상황을 고려하면, 예측탐사의 기반인 광상 배태 가능성 평가 기술의 신뢰도는 점차 높아질 것으로 예상된다. 이번 연구에서는 심층신경망을 이용하여 화성암과 단층 및 자력탐사 정보의 광화 인자로의 활용 가능성을 확인하였다. 지질 정보의 수치화 기법으로 단층, 화성암, 자력 정보를 입력 자료로 구성하여 0.9 이상의 정확도를 가지며 예측 값이 안정적으로 수렴하는 금속 광상 예측 모델을 구축할 수 있었다. 이 기술은 추후 정밀한 지질 조사 결과와 물리탐사 정보가 확보된다면, 광화대 규모에서의 예측 탐사에도 활용할 수 있을 것으로 기대된다. 또한, 이 연구를 통해 지하의 화성암 정보를 제공하는 자력자료를 활용할 경우 지표의 화성암 정보를 보완하여 보다 높은 성능의 모델을 구축할 수 있는 것으로 확인되었다. 즉, 단순히 많은 자료를 융합하는 것 보다는 광체 성인과의 지질학적 상관관계를 고려하여 입력 자료를 구성하는 것이 보다 중요하다.

Keywords

References

  1. Choi, S.-G., Park, S. J., Lee, P.-K., and Kim, C. S., 2004, An Overview of Geoenvironmental Implications of Mineral Deposits in Korea, Econ. Environ. Geol., 37(1), 1-19 (in Korean with English abstract).
  2. Choi, S.-G., and Park, S. J., Kim, S. W., Kim C. S., and Oh, C.-W., 2006, Mesozoic Gold-Silver Mineralization in South Korea: Metallogenic Provinces Reestimated to the Geodynamic Setting, Econ. Environ. Geol., 39(5), 567-581 (in Korean with English abstract).
  3. Choi, S.-G. and Park, S. J., 2007, The Origin and Evolution of the Mesozoic Ore-forming Fluids in South Korea: Their Genetic Implications, Econ. Environ. Geol., 40(5), 517-535 (in Korean with English abstract).
  4. Choi, S.-G., Choi, B. K., Ahn, Y. H., and Kim, T. H., 2009, Reevaluation of Genetic Environments of Zinc-lead Deposits to Predict Hidden Skarn Orebody, Econ. Environ. Geol., 42(4), 301-314 (in Korean with English abstract).
  5. Choi, S.-G., Koo, M.-H., Kang, H.-S., and Ahn, Y. H., 2011, Major Molybdenum Mineralization and Igneous Activity, South Korea, Econ. Environ. Geol., 44(2), 109-122 (in Korean with English abstract). https://doi.org/10.9719/EEG.2011.44.2.109
  6. KIGAM, 1995, Geological Map of Korea(1:1,000,000), SUNGJI MUNWHA Co. (in Korean).
  7. KIGAM, 2012, Evaluation of development possibility for the security of the industrial mineral resources(Cu, Pb, Zn, Au, etc) on the domestic mines, KIGAM, 1-309 (in Korean with English abstract).
  8. KIGAM, 2016, Development of airborne geophysical techniques for regional mapping and exploration of mineral deposits, KIGAM, 12-39 (in Korean with English abstract).
  9. Kim C. S., Choi, S.-G., Choi, S.-H., and Lee, I. W., 2002, Hydrothermal Alteration and Its Genetic Implication in the Gasado Volcanic-hosted Epithermal Gold-Silver Deposit: Use in Exploration, J. Miner. Soc. Korea, 15, 205-220 (in Korean with English abstract).
  10. Kim, O. J, 1971, Metallogenic epochs and provinces of South Korea, J. Geol. Soc. Korea, 7, 37-59.
  11. Lee, C., Oh, H.-J., Cho, S.-J., Kihm, Y., Park, G., and Choi, S.-G., 2019, Three-dimensional Prospectivity mapping of skarntype mineralization in the southern Taebaek area, Korea, Geosci. J., 23(2), 327-339, doi: 10.1007/s12303-018-0035-y.
  12. Mira Geoscience, 2018, VPmg 9.2 - User guide, Mira Geoscience. 1-69.
  13. Park, G., Cho, S.-J., Oh, H.-J., and Lee, C.-W., 2014a, Mineral Potential Mapping of Gagok Mine Using 3D Geological Modeling, J. Korean Earth Sci. Soc., 35(6), 412-421 (in Korean with English abstract), doi: 10.7582/GGE.2014.17.4.209.
  14. Park, G., Park, S., Son, J.-S., Kim, C., and Cho, S.-J., 2014b, Potential Mapping of Moisan area Using SIP and 3D Geological Modeling, Geophy. and Geophys. Explor., 17(4), 209-215 (in Korean with English abstract), doi: 10.7582/GGE.2014.17.4.209.
  15. Park, G., 2018, Method of Treating Geologic Data, Korean Patents 10-1830318 (in Korean).
  16. Park, Y.-S., Rim, H., Lim, M., and Shin, Y. H., 2019, The Magnetic Anomaly Map of Korea, Geophy. and Geophys. Explor., 22(1), 29-36 (in Korean with English abstract), doi: 10.7582/GGE.2019.22.1.029.
  17. So, C.-S., Shelton, K. L., Chi, S.-J.. and Yun, S.-T., 1991, Geochemical studies of the Gyeongchang W-Mo mine, Republic of Korea: Progressive meteoric water inundation of magmatic hydrothermal system, Econ. Geol., 86, 750-767, doi: 10.2113/gsecongeo.86.4.750.
  18. Yoon, D., Kim, S., Kim, J., Park, G., Park, H., Byun, J., Suh, J., Lee, C., Jang, I., Cho, S., and Choi, Y., 2018, Introduction to Machine Learning in Resource Development, CIR press, 265-288, 327-347 (in Korean).