数据驱动斑岩型矿床时空预测模型

doi:10.13745/j.esf.sf.2025.2.5

摘要/Abstract

摘要：

矿产资源预测与评价一直是大数据和人工智能在地球科学研究中的重要应用领域。自20世纪60年代,人工智能的研究浪潮和技术革命深刻影响和推动着矿产资源预测领域的发展,催生了重大理论突破和方法技术创新,有效支撑了找矿勘查实践。在当前地球系统科学研究的主旋律下,矿产资源定量预测亟需跳出“静态控矿要素空间相关分析”的思维惯性,考虑成矿系统“源-运-储-变-保”深时动态演化历史,向“全要素跨尺度动态综合预测”方向延伸,进而发展时空数据耦合的矿产资源智能预测评价理论和方法。斑岩型矿床作为全球铜、钼、金等矿产的重要来源,记录了板块构造运动驱动的地球层圈相互作用和物质循环的关键信息;无论是斑岩型矿床勘查还是板块构造重建,都积累了大量相关的全球地学时空数据。本文主要介绍了时空耦合的数据驱动斑岩型矿床成矿预测研究思路,包括深时数据集构建、机器学习模型开发及其应用实践案例;提出了融合深时岩浆岩地球化学特征和俯冲板块动力学参数的斑岩型矿床时空预测机器学习模型;通过大数据分析揭示了俯冲碳酸盐岩通量是决定岩浆成矿禀赋的关键动力学参数,为沉积物俯冲在大规模岩浆成矿中的关键作用提供了地球动力学证据;定量评价了安第斯成矿带斑岩矿床时空分布规律及其资源潜力。因此,发展时空耦合的成矿预测理论和方法可为理解深时物质循环和资源效应、揭示矿产资源时空分布规律以及指导矿产勘查提供重要思路和独特视角。

关键词: 数据驱动, 人工智能, 斑岩型矿床, 成矿预测, 地球系统科学

Abstract:

Mineral resource prediction and evaluation have long been important application areas for big data and artificial intelligence (AI) in geosciences research. Since the 1960s, the waves of AI research and technological revolutions have profoundly impacted the development of mineral resource prediction, leading to significant breakthroughs in both theory and technical methods in this field, thereby supporting mineral exploration efforts. In the context of current Earth system science research, quantitative mineral resource prediction needs to move beyond the traditional thinking of “static spatial correlation analysis of ore-controlling factors” and consider the deep-time dynamic evolution history of metallogenic systems, including the “source-transport-storage-transformation-preservation” processes. This requires extending to “multi-factor, cross-scale dynamic integrated prediction” within Earth system science, and developing theories and methods for intelligent mineral resource prediction and evaluation that couple spatiotemporal data. Porphyry deposits are major sources of copper, molybdenum, gold, and other minerals globally, recording key information about the Earth’s lithospheric interactions and material cycles driven by plate tectonics. Both porphyry deposit exploration and plate tectonic reconstruction have accumulated vast amounts of relevant global geospatial data. This paper primarily introduces a data-driven spatiotemporal machine learning model for predicting porphyry deposits, including the construction of deep-time geoscience datasets, machine learning models, and their application cases. We propose a spatiotemporal machine learning model for porphyry deposit prediction that integrates deep-time igneous geochemical data with dynamics parameters of plate subduction. The big data mining revealed that the carbonate subduction flux is a key parameter determining the mineralization potential of arc magmas, providing geodynamic evidence for the critical role of sediment subduction in magmatic-hydrothermal porphyry mineralization. The study also quantitatively evaluates the spatiotemporal distribution patterns and resource potential of porphyry copper deposits in the Andes belt across different geological periods. Therefore, the development of spatiotemporal coupled metallogenic prediction theories and methods can provide important insights and unique perspectives for understanding material recycling and resource effects in deep time, predicting the spatiotemporal distribution of potential mineral resources, thereby guiding mineral exploration.

Key words: data-driven discovery, artificial intelligence, porphyry ore deposits, mineral resource prediction, Earth system science

中图分类号:

P628

陈国雄, 张越鹏, 罗磊, 夏庆霖, 成秋明. 数据驱动斑岩型矿床时空预测模型[J]. 地学前缘, 2025, 32(4): 46-59.

CHEN Guoxiong, ZHANG Yuepeng, LUO Lei, XIA Qinglin, CHENG Qiuming. Data-driven spatio-temporal prediction model of porphyry deposits[J]. Earth Science Frontiers, 2025, 32(4): 46-59.

图/表 10

图1 时空数据耦合的斑岩成矿预测研究思路

Fig.1 Research framework of data-driven spatio-temporal prediction of porphyry Cu deposits

图2 全球斑岩型铜矿床时空分布图

Fig.2 Spatio-temporal distribution of global porphyry copper deposits

图3 深时板块俯冲动力学参数

Fig.3 Dynamic parameters related to plate subduction in deep time

表1 板块俯冲动力学参数

Table 1 Key parameters of plate subduction dynamics

动力学参数	单位
俯冲汇聚速度	cm/a
俯冲板块体积	km³/a
到俯冲带距离	°
洋壳年龄	Ma
碳酸盐岩厚度	m
深海沉积物厚度	m
洋壳CO₂含量	%

图4 安第斯带斑岩成矿和不成矿二分类机器学习模型的SHAP值分布(蜂群图)及深时斑岩型矿床形成背后的俯冲动力学参数分析

Fig.4 SHAP summary plot of plate subduction dynamic parameters for machine learning-based discriminant between ore deposits and non-ore deposits in Andes belt

图5 安第斯带斑岩型矿床与板块动力学参数时空关系 a—俯冲汇聚速度;b—俯冲板块体积;c—碳酸盐岩厚度;d—洋壳CO2含量;e—洋壳年龄;f—深海沉积物厚度。

Fig.5 Spatio-temporal relationship between porphyry copper deposits and plate subduction dynamics in Andes belt

图6 斑岩成矿岩体地球化学成分与俯冲动力学参数相关性热图

Fig.6 Correlation matrix between ore-productive igneous geochemistry and plate subduction dynamics paramters

图7 安第斯带深时斑岩型铜矿床分布。 a—已发现矿床年龄分布;b—预测成矿概率密度。

Fig.7 Temporal distribution of porphyry copper deposits in Andes belt

图8 安第斯带斑岩型铜矿床时空分布预测。

Fig.8 Spatio-temporal prediction of porphyry copper deposits in Andes belt a—20~10 Ma;b—40~30 Ma。

图9 安第斯带深时斑岩型铜矿床构造扩散模拟与资源潜力评估 a—优化目标函数(深时矿床统计分布);b—矿床数量时深分布;c—模型输入和输出;d—深部矿床数量分布(现今)。

Fig.9 Tectonic diffusion modeling and mineral resource appraisal of porphyry copper deposits in Andes belt

参考文献 73

[1]	成秋明. 什么是数学地球科学及其前沿领域?[J]. 地学前缘, 2021, 28(3): 6-25. DOI
[2]	翟明国, 杨树锋, 陈宁华, 等. 大数据时代: 地质学的挑战与机遇[J]. 中国科学院院刊, 2018, 33(8): 825-831.
[3]	郭华东, 王力哲, 陈方, 等. 科学大数据与数字地球[J]. 科学通报, 2014, 59(12): 1047-1054.
[4]	李德仁, 姚远, 邵振峰. 智慧城市中的大数据[J]. 武汉大学学报(信息科学版), 2014, 39(6): 631-640.
[5]	赵鹏大. 大数据时代呼唤各科学领域的数据科学[J]. 中国科技奖励, 2014(9): 29-30.
[6]	CHEN M, QIAN Z, BOERS N, et al. Collaboration between artificial intelligence and Earth science communities for mutual benefit[J]. Nature Geoscience, 2024, 17(10): 949-952.
[7]	ZHAO T J, WANG S, OUYANG C J, et al. Artificial intelligence for geoscience: progress, challenges, and perspectives[J]. The Innovation, 2024, 5(5): 100691.
[8]	HEY T, TANSLEY J, TOLLE E. The fourth paradigm: data intensive scientific discovery[M]. Redmond: Microsoft Research, 2009.
[9]	吴冲龙, 刘刚. 大数据与地质学的未来发展[J]. 地质通报, 2019, 38(7): 1081-1088.
[10]	BRISTOL R S, EULISS N H, BOOTH N L, et al. Science strategy for core science systems in the U.S. Geological Survey, 2013-2023[R]. Reston: USGS, 2012.
[11]	CHENG Q M, OBERHÄNSLI R, ZHAO M L. A new international initiative for facilitating data-driven Earth science transformation[J]. Geological Society, London, Special Publications, 2020, 499(1): 225-240.
[12]	WANG C S, HAZEN R M, CHENG Q M, et al. The deep-time digital earth program: data-driven discovery in geosciences[J]. National Science Review, 2021, 8(9): nwab027.
[13]	成秋明. 极端地质事件定量模拟与预测[J]. 中国科学:地球科学, 2022, 52(6): 975-991.
[14]	BERGEN K J, JOHNSON P A, DE HOOP M V, et al. Machine learning for data-driven discovery in solid Earth geoscience[J]. Science, 2019, 363(6433): eaau0323.
[15]	FAN J X, SHEN S Z, ERWIN D H, et al. A high-resolution summary of Cambrian to Early Triassic marine invertebrate biodiversity[J]. Science, 2020, 367(6475): 272-277.
[16]	HULBERT C, ROUET-LEDUC B, JOHNSON P A, et al. Similarity of fast and slow earthquakes illuminated by machine learning[J]. Nature Geoscience, 2019, 12(1): 69-74. DOI
[17]	REICHSTEIN M, CAMPS-VALLS G, STEVENS B, et al. Deep learning and process understanding for data-driven Earth system science[J]. Nature, 2019, 566(7743): 195-204.
[18]	CHEN G X, CHENG Q M, LYONS T W, et al. Reconstructing Earth’s atmospheric oxygenation history using machine learning[J]. Nature Communications, 2022, 13: 5862.
[19]	王世称. 综合信息矿产预测理论与方法体系新进展[J]. 地质通报, 2010, 29(10): 1399-1403.
[20]	AGTERBERG F P. Computer programs for mineral exploration[J]. Science, 1989, 245(4913): 76-81. PMID
[21]	BROWN W M, GEDEON T D, GROVES D I, et al. Artificial neural networks: a new method for mineral prospectivity mapping[J]. Australian Journal of Earth Sciences, 2000, 47(4): 757-770.
[22]	CHENG Q, AGTERBERG F P. Fuzzy Weights of evidence method and its application in mineral potential mapping[J]. Natural Resources Research, 1999, 8:27-35.
[23]	陈建平, 吕鹏, 吴文, 等. 基于三维可视化技术的隐伏矿体预测[J]. 地学前缘, 2007, 14(5): 54-62.
[24]	毛先成. 三维数字矿床与隐伏矿体立体定量预测研究[D]. 长沙: 中南大学, 2006.
[25]	肖克炎, 李楠, 孙莉, 等. 基于三维信息技术大比例尺三维立体矿产预测方法及途径[J]. 地质学刊, 2012, 36(3): 229-236.
[26]	袁峰, 李晓晖, 张明明, 等. 隐伏矿体三维综合信息成矿预测方法[J]. 地质学报, 2014, 88(4): 630-643.
[27]	周涛发, 袁峰, 张明明, 等. 三维地质模拟在深部找矿勘探中的应用[J]. 安徽地质, 2011, 21(2): 100-104.
[28]	CHEN Y L, WU W. Mapping mineral prospectivity using an extreme learning machine regression[J]. Ore Geology Reviews, 2017, 80: 200-213.
[29]	ZUO R G, XIONG Y H, WANG J, et al. Deep learning and its application in geochemical mapping[J]. Earth-Science Reviews, 2019, 192: 1-14. DOI
[30]	BARBER N D, EDMONDS M, JENNER F, et al. Amphibole control on copper systematics in arcs: insights from the analysis of global datasets[J]. Geochimica et Cosmochimica Acta, 2021, 307: 192-211.
[31]	CHIARADIA M. Copper enrichment in arc magmas controlled by overriding plate thickness[J]. Nature Geoscience, 2014, 7(1): 43-46.
[32]	WROBEL-DAVEAU J C, NICOLL G R. Plate tectonics as a tool for global screening of magmatic arcs and predictions for related porphyry deposits[J]. Economic Geology, 2022, 117(6): 1429-1443.
[33]	WU C, CHEN G X, CHEN H Y. Unraveling the link between worldwide adakite-like rocks and porphyry Cu deposits[J]. Chemical Geology, 2025, 673: 122521.
[34]	CHEN G X, HUANG N, WU G P, et al. Mineral prospectivity mapping based on wavelet neural network and Monte Carlo simulations in the Nanling W-Sn metallogenic province[J]. Ore Geology Reviews, 2022, 143: 104765.
[35]	LI Q K, CHEN G X, WANG D T. Mineral prospectivity mapping using semi-supervised machine learning[J]. Mathematical Geosciences, 2025, 57(2): 275-305.
[36]	翟裕生. 论成矿系统[J]. 地学前缘, 1999, 6(1): 13-27.
[37]	翟裕生. 地球系统科学与成矿学研究[J]. 地学前缘, 2004, 11(1): 1-10.
[38]	CAWOOD P A, HAWKESWORTH C J. Temporal relations between mineral deposits and global tectonic cycles[J]. Geological Society, London, Special Publications, 2015, 393(1): 9-21.
[39]	陈华勇, 吴超. 俯冲带斑岩铜矿系统成矿机理与主要挑战[J]. 中国科学: 地球科学, 2020, 50(7): 865-886.
[40]	成秋明. 洋中脊动力学与俯冲带地震-岩浆-成矿事件远程效应[J]. 地学前缘, 2024, 31(1): 1-14. DOI
[41]	侯增谦. 斑岩Cu-Mo-Au矿床: 新认识与新进展[J]. 地学前缘, 2004, 11(1): 131-144.
[42]	COOKE D R, HOLLINGS P, WALSHE J L. Giant porphyry deposits: characteristics, distribution, and tectonic controls[J]. Economic Geology, 2005, 100(5): 801-818.
[43]	RICHARDS J P. Giant ore deposits formed by optimal alignments and combinations of geological processes[J]. Nature Geoscience, 2013, 6(11): 911-916.
[44]	PARK J W, CAMPBELL I H, CHIARADIA M, et al. Crustal magmatic controls on the formation of porphyry copper deposits[J]. Nature Reviews Earth and Environment, 2021, 2(8): 542-557.
[45]	LEE C A, TANG M. How to make porphyry copper deposits[J]. Earth and Planetary Science Letters, 2020, 529: 115868.
[46]	孙卫东, 凌明星, 杨晓勇, 等. 洋脊俯冲与斑岩铜金矿成矿[J]. 中国科学: 地球科学, 2010, 40(2): 127-137.
[47]	LAMONT T N, LOADER M A, ROBERTS N M W, et al. Porphyry copper formation driven by water-fluxed crustal melting during flat-slab subduction[J]. Nature Geoscience, 2024, 17(12): 1306-1315.
[48]	王瑞, 张京渤, 罗晨皓, 等. 深部过程和物质架构对大陆碰撞带Cu-REE成矿系统的控制: 以冈底斯和三江碰撞带为例[J]. 地学前缘, 2024, 31(1): 211-225. DOI
[49]	SILLITOE R H. Porphyry copper systems[J]. Economic Geology, 2010, 105(1): 3-41.
[50]	DICKEN C L, DUNLAP P, PARKS H L, et al. Spatial database for a global assessment of undiscovered copper resources: chapter Z in global mineral resource assessment[R]. Reston: USGS, 2016.
[51]	MÜLLER R D, DUTKIEWICZ A. Oceanic crustal carbon cycle drives 26-million-year atmospheric carbon dioxide periodicities[J]. Science Advances, 2018, 4(2): eaaq0500.
[52]	MÜLLER R D, CANNON J, QIN X D, et al. GPlates: building a virtual earth through deep time[J]. Geochemistry, Geophysics, Geosystems, 2018, 19(7): 2243-2261.
[53]	MATHER B R, MÜLLER R D, ZAHIROVIC S, et al. Deep time spatio-temporal data analysis using PyGPlates with PlateTectonicTools and GPlately[J]. Geoscience Data Journal, 2024, 11(1): 3-10.
[54]	DIAZ-RODRIGUEZ J, MÜLLER R D, CHANDRA R. Predicting the emplacement of Cordilleran porphyry copper systems using a spatio-temporal machine learning model[J]. Ore Geology Reviews, 2021, 137: 104300.
[55]	SETON M, MÜLLER R D, ZAHIROVIC S, et al. A global data set of present-day oceanic crustal age and seafloor spreading parameters[J]. Geochemistry, Geophysics, Geosystems, 2020, 21(10): e2020GC009214.
[56]	DUTKIEWICZ A, MÜLLER R D, CANNON J, et al. Sequestration and subduction of deep-sea carbonate in the global ocean since the Early Cretaceous[J]. Geology, 2019, 47(1): 91-94.
[57]	DUTKIEWICZ A, MÜLLER R D, WANG X, et al. Predicting sediment thickness on vanished ocean crust since 200 ma[J]. Geochemistry, Geophysics, Geosystems, 2017, 18(12): 4586-4603.
[58]	LOUCKS R R. Distinctive composition of copper-ore-forming arcmagmas[J]. Australian Journal of Earth Sciences, 2014, 61(1): 5-16.
[59]	LIANG Q B, CHEN G X, LUO L, et al. Appraising the porphyry Cu fertility using apatite trace elements: a machine learning method[J]. Journal of Geochemical Exploration, 2025, 270: 107664.
[60]	LUO L, CHEN G X, LI Z H. Identifying tectonic settings of porphyry copper deposits using zircon trace elements: a semi-supervised machine learning method[J]. Ore Geology Reviews, 2024, 171: 106170.
[61]	LLOYD S. Least squares quantization in PCM[J]. IEEE Transactions on Information Theory, 1982, 28(2): 129-137.
[62]	GREENACRE M, GROENEN P J F, HASTIE T, et al. Principal component analysis[J]. Nature Reviews Methods Primers, 2022, 2(1):100.
[63]	HEARST M A, DUMAIS S T, OSUNA E, et al. Support vector machines[J]. IEEE Intelligent Systems and Their Applications, 1998, 13(4): 18-28.
[64]	VAN ENGELEN J E, HOOS H H. A survey on semi-supervised learning[J]. Machine Learning, 2020, 109(2): 373-440.
[65]	CHEN G X, KUSKY T, LUO L, et al. Hadean tectonics: insights from machine learning[J]. Geology, 2023, 51(8): 718-722.
[66]	LUNDBERG S M, LEE S. A unified approach to interpreting model predictions[C]// Proceedings of the 30th conference on Neural Information Processing Systems, Long Beach, California. Red Hook: Curran Associates, Inc., 2017, 30: 4765-4774.
[67]	RIBEIRO M T, SINGH S, GUESTRIN C. “Why should I trust you?”:explaining the predictions of any classifier[C]// Proceedings of the 2016 conference of the North American Chapter of the Association for Computational Linguistics:demonstrations, San Diego, California. Stroudsburg: USAACL, 2016: 1135-1144.
[68]	KAWAMOTO T, YOSHIKAWA M, KUMAGAI Y, et al. Mantle wedge infiltrated with saline fluids from dehydration and decarbonation of subducting slab[J]. Proceedings of the National Academy of Sciences of the United States of America, 2013, 110(24): 9663-9668. DOI PMID
[69]	CAI R H, LIU J G, PEARSON D G, et al. Oxidation of the deep big mantle wedge by recycled carbonates: constraints from highly siderophile elements and osmium isotopes[J]. Geochimica et Cosmochimica Acta, 2021, 295: 207-223.
[70]	RIELLI A, TOMKINS A G, NEBEL O, et al. Evidence of sub-arc mantle oxidation by sulphur and carbon[J]. Geochemical Perspectives Letters, 2017: 124-132.
[71]	QIU K F, ROMER R L, LONG Z Y, et al. The role of an oxidized lithospheric mantle in gold mobilization[J]. Science Advances, 2024, 10(41): eado6262.
[72]	KESLER S E, WILKINSON B H. Earth’s copper resources estimated from tectonic diffusion of porphyry copper deposits[J]. Geology, 2008, 36(3): 255.
[73]	LUO L, CHEN G X, XIA Q L. Tectonic diffusion estimates of global porphyry molybdenum resources[J]. Natural Resources Research, 2022, 31(2): 751-766.