Multifractal analysis and random forest algorithm for mineral prospecting in the Habahe gold deposit, Xinjiang

doi:10.13745/j.esf.sf.2025.4.62

Abstract

Abstract:

In the era of big data, machine learning-based intelligent mineral prediction methods have been widely applied. The integration of non-linear theory and techniques, such as fractal and multifractal approaches, into intelligent mineral prospecting research provides new perspectives and technical support. This study focuses on the Habahe gold deposit in Xinjiang, China, and establishes a four-factor information prospecting model based on regional structure, mineralization alteration, magnetic anomalies, and induced polarization anomalies. An intelligent prediction workflow is implemented, by combining the multifractal method with the random forest algorithm. The S-A multifractal filtering technique and local singularity analysis are employed to separate the background variations of regional geophysical and geochemical data from superimposed anomalies, enabling the extraction of concealed information indicative of deep mineralization. The C-Nsum multifractal model is applied to reveal the hidden nonlinear characteristics of gold content in drilling data and determine anomaly thresholds. Subsequently, the random forest algorithm and SHAP method is utilized for comprehensive information integration and feature contribution evaluation, achieving quantitative prediction of gold mineral resources. This approach delineated three prospective mineralization targets, which were validated through drilling, demonstrating the effectiveness of multifractal theory in quantitative mineral prediction within the Habahe gold deposit area. The results provide a robust basis for subsequent mineral exploration efforts.

Key words: S-A multifractal model, local singularity analysis, random forest, mineral prospectivity, machine learning, Habahe gold deposit

CLC Number:

JIAN Fuyuan, ZHANG Ziming, DONG Yuelin, ZHANG Wenjing, HAO Fengyun, WANG Yiming, WANG Yu, ZHANG Zhenjie. Multifractal analysis and random forest algorithm for mineral prospecting in the Habahe gold deposit, Xinjiang[J]. Earth Science Frontiers, 2025, 32(4): 78-94.

Figures/Tables 17

Fig.1 A simplified geological map of Habahe gold deposit area

Fig.2 Metallogenic model of Jinba gold deposit. Modified after [47].

Fig.3 Random forest classification principle

Fig.4 C-Nsum fractal double logarithmic diagram

Fig.5 Kriging interpolation results

Fig.6 S-A multifractal filtering results

Fig.7 Magnetic anomaly

Fig.8 Apparent resistivity

Fig.9 Apparent polarizability

Fig.10 Structural features

Fig.11 Mineral prospecting results of Habahe gold deposit

Fig.12 Confusion matrix of training set and prediction set

Fig.13 ROC curve of random forest model

Fig.14 SHAP-based Feature Importance

Fig.15 Metallogenic prospective area of Habahe gold deposit base

Fig.16 The division basis of metallogenic prospective area

Fig.17 Drilling verification

References 87

[1]	赵鹏大. 成矿定量预测与深部找矿[J]. 地学前缘, 2007, 14(5): 1-10.
[2]	王世称, 成秋明, 范继璋. 金矿资源综合信息评价方法[M]. 长春: 吉林科学技术出版社, 1990.
[3]	刘艳鹏, 朱立新, 周永章. 大数据挖掘与智能预测找矿靶区实验研究: 卷积神经网络模型的应用[J]. 大地构造与成矿学, 2020, 44(2): 192-202.
[4]	TAHMASEBI P, HEZARKHANI A. A hybrid neural networks-fuzzy logic-genetic algorithm for gradeestimation[J]. Computers & Geosciences, 2012, 42: 18-27.
[5]	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.
[6]	王语, 周永章, 肖凡, 等. 基于成矿条件数值模拟和支持向量机算法的深部成矿预测: 以粤北凡口铅锌矿为例[J]. 大地构造与成矿学, 2020, 44(2): 222-230.
[7]	WOODHEAD J, LANDRY M. Harnessing thepower of artificial intelligence and machine learning in mineral exploration: opportunities and cautionary notes[J]. SEG Discovery, 2021(127): 19-31.
[8]	肖克炎, 娄德波, 孙莉, 等. 全国重要矿产资源潜力评价一些基本预测理论方法的进展[J]. 吉林大学学报(地球科学版), 2013, 43(4): 1073-1082.
[9]	王世称, 陈永清. 金矿综合信息成矿系列预测理论体系[J]. 黄金地质, 1995, 1(1): 1-7.
[10]	成秋明. 非线性成矿预测理论: 多重分形奇异性-广义自相似性-分形谱系模型与方法[J]. 地球科学: 中国地质大学学报, 2006, 31(3): 337-348.
[11]	张振杰, 成秋明, 杨玠, 等. 机器学习与成矿预测: 以闽西南铁多金属矿预测为例[J]. 地学前缘, 2021, 28(3): 221-235. DOI
[12]	肖克炎, 樊铭静, 孙莉, 等. 矿床成矿系列综合信息预测理论方法及其应用[J]. 地球学报, 2023, 44(5): 769-780.
[13]	ZUO R G. Geodata science-based mineral prospectivity mapping: a review[J]. Natural Resources Research, 2020, 29(6): 3415-3424.
[14]	翟裕生. 试论矿床成因的基本模型[J]. 地学前缘, 2014, 21(1): 1-8.
[15]	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.
[16]	YOUSEFI M, CARRANZA E J M, KREUZER O P, et al. Data analysis methods forprospectivity modelling as applied to mineral exploration targeting: state-of-the-art and outlook[J]. Journal of Geochemical Exploration, 2021, 229: 106839.
[17]	周永章, 王俊, 左仁广, 等. 地质领域机器学习、深度学习及实现语言[J]. 岩石学报, 2018, 34(11): 3173-3178.
[18]	周永章, 肖凡. 管窥人工智能与大数据地球科学研究新进展[J]. 地学前缘, 2024, 31(4): 1-6. DOI
[19]	季斌, 周涛发, 袁峰. 遗传算法优化支持向量机矿产预测方法[J]. 测绘科学, 2015, 40(10): 106-109.
[20]	GAO M, WANG G W, XU Y C, et al. 3D mineral exploration Cu-Zn targeting with multi-sourcegeoscience datasets in the Weilasituo-Bairendaba district, Inner Mongolia, China[J]. Frontiers in Earth Science, 2023, 11: 1102640.
[21]	SENANAYAKE I P, KIEM A S, HANCOCK G R, et al. Aspatial data-driven approach for mineral prospectivity mapping[J]. Remote Sensing, 2023, 15(16): 4074.
[22]	ZUO R G, XIONG Y H, WANG Z Y, et al. A new generation of artificial intelligence algorithms for mineral prospectivity mapping[J]. Natural Resources Research, 2023, 32(5): 1859-1869.
[23]	刘承照, 韩帅, 李明超, 等. 耦合PCA-SVM算法的金矿矿床规模预测分析研究[J]. 地学前缘, 2019, 26(4): 138-145. DOI
[24]	马瑶, 赵江南. 机器学习方法在矿产资源定量预测应用研究进展[J]. 地质科技通报, 2021, 40(1): 132-141.
[25]	向杰, 陈建平, 肖克炎, 等. 基于机器学习的三维矿产定量预测: 以四川拉拉铜矿为例[J]. 地质通报, 2019, 38(12): 2010-2021.
[26]	付光明, 严加永, 罗凡, 等. 基于随机森林算法的三维成矿预测: 以赣东北朱溪钨矿床外围为例[J]. 地质论评, 2021, 67(增刊1): 275-276.
[27]	王世称, 艾宪森, 李忠, 等. 综合信息矿产预测理论与方法体系[R]. 长春: 吉林大学, 2005.
[28]	SCHMIDHUBER J. Deep learning in neural networks: an overview[J]. Neural Networks, 2015, 61: 85-117. PMID
[29]	李沐思, 陈丽蓉, 谢飞, 等. 面向地球化学异常识别的深度学习算法对比研究[J]. 物探与化探, 2023, 47(1): 179-189.
[30]	CHEN Z Y, XIONG Y H, YIN B J, et al. Recognizing geochemical patterns related to mineralization using a self-organizing map[J]. Applied Geochemistry, 2023, 151: 105621.
[31]	LV X M, WANG G W. GIS-based mineral prospectivity mapping using machine learning methods: a case study from Duobaoshan ore district, northeastern China[J]. Ore Geology Reviews, 2024, 175: 106352.
[32]	MANDELBROT B. How long is the coast of Britain? Statistical self-similarity and fractional dimension[J]. Science, 1967, 156(3775): 636-638. PMID
[33]	MANDELBROT B B. Les objets fractals: forme, hasard et dimension[C]. Paris: FLAMMARION, 1975.
[34]	CHENG Q M, AGTERBERG F P, BALLANTYNE S B. The separation of geochemical anomalies from background by fractal methods[J]. Journal of Geochemical Exploration, 1994, 51(2): 109-130.
[35]	CHENG Q M. A new model for quantifying anisotropic scale invariance and for decomposition of mixing patterns[J]. Mathematical Geology, 2004, 36(3): 345-360.
[36]	CHENG Q M. Multifractal distribution of eigenvalues and eigenvectors from 2D multiplicative cascade multifractal fields[J]. Mathematical Geology, 2005, 37(8): 915-927.
[37]	CHENG Q M. Mapping singularities with stream sediment geochemical data for prediction of undiscovered mineral deposits in Gejiu, Yunnan Province, China[J]. Ore Geology Reviews, 2007, 32(1/2): 314-324.
[38]	牛旭刚, 周宏, 张文纲, 等. S-A多重分形法在地球化学异常圈定中的对比应用: 以甘肃文康地区为例[J]. 地质与勘探, 2023, 59(4): 817-827.
[39]	PARSA M, MAGHSOUDI A, YOUSEFI M, et al. Multifractal interpolation and spectrum-area fractal modeling of stream sediment geochemical data: implications for mapping exploration targets[J]. Journal of African Earth Sciences, 2017, 128: 5-15.
[40]	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.
[41]	李晓晖, 袁峰, 贾蔡, 等. 基于地统计学插值方法的局部奇异性指数计算比较研究[J]. 地理科学, 2012, 32(2): 136-142. DOI
[42]	李晓晖, 袁峰, 贾蔡, 等. 基于反距离加权和克里格插值的S-A多重分形滤波对比研究[J]. 测绘科学, 2012, 37(3): 87-89, 46.
[43]	陈志军, 成秋明, 陈建国. 基于栅格数据的局部奇异性分析迭代方法及其软件实现[J]. 地质学刊, 2014, 38(4): 613-622.
[44]	唐杰, 王文磊, 袁长江. 多重分形局部奇异性分析方法在中国西藏多龙矿集区深层次地球化学异常识别中的应用[J]. 地质科技通报, 2024, 43(5): 296-310.
[45]	万丽, 刘慧, 曾祥健. 普朗斑岩型铜矿床成矿元素多重分形特征及其矿化强度指示[J]. 吉林大学学报(地球科学版), 2021, 51(4): 1054-1063.
[46]	沈远超, 申萍, 曾庆栋, 等. 新疆阿尔泰金矿带主要金矿类型、成矿规律及成矿预测[J]. 地质与勘探, 2004, 40(5): 1-5.
[47]	LI S D, CHEN C, GAO L L, et al. Fluid inclusions and H-O-C-S-Pb isotopic systematics of the Jinba gold deposit, NW China: implications for ore genesis[J]. Frontiers in Earth Science, 2021, 9: 638375.
[48]	LAURENT-CHARVET S, CHARVET J, SHU L S, et al. Palaeozoic late collisional strike-slip deformations in Tianshan and Altay, eastern Xinjiang, NW China[J]. Terra Nova, 2002, 14(4): 249-256.
[49]	段旦, 韩国胜, 刘东升. 新疆金坝金矿床地质地球化学特征及找矿标志[J]. 世界有色金属, 2019(14): 66-69.
[50]	薛小龙. 新疆哈巴河县哲兰德金矿矿床地质特征及成因探讨[J]. 世界有色金属, 2018(4): 150, 152.
[51]	王颖维, 徐九华, 边春静, 等. 新疆哈巴河县金坝金矿地质特征与流体包裹体研究[J]. 矿物学报, 2015, 35(增刊1): 625.
[52]	李培树. 新疆哈巴河县金坝金矿南矿带成矿成因浅析[J]. 西部探矿工程, 2019, 31(9): 193-196.
[53]	李永, 周刚, 柴凤梅. 阿尔泰南缘哈巴河岩体LA-ICP-MS锆石定年及地质意义[J]. 新疆地质, 2012, 30(2): 146-151.
[54]	李伟, 张玉坛. 阿勒泰哈巴河金矿带重要矿床成因分析和研究[J]. 西部探矿工程, 2024, 36(8): 105-108.
[55]	刘诚, 王平户, 李先军, 等. 新疆托库孜巴依金矿带中蚀变闪长岩型金矿体地质特征[J]. 科技信息, 2011(20): 730-732.
[56]	李光明, 沈远超, 刘铁兵, 等. 新疆阿尔泰南缘托库孜巴依金矿成矿演化: 石英脉系、同位素地球化学及其Ar-Ar年代学证据[J]. 矿床地质, 2007, 26(1): 15-32.
[57]	陈华勇, 陈衍景, 刘玉琳. 新疆额尔齐斯金矿带的成矿作用及其与中亚型造山作用的关系[J]. 中国科学(D辑: 地球科学), 2000, 30(增刊1): 38-44.
[58]	覃毅学, 孔华, 段旦, 等. 新疆托库孜巴依和金坝矿床闪长岩与金成矿关系: 锆石定年与载金矿物特征约束[J]. 矿床地质, 2023, 42(2): 381-395.
[59]	XU X Y, WANG H L, LI P, et al. Geochemistry and geochronology of Paleozoic intrusions in the Nalati (Narati) area in western Tianshan, Xinjiang, China: implications for Paleozoic tectonic evolution[J]. Journal of Asian Earth Sciences, 2013, 72: 33-62.
[60]	张敏, 牛向龙, 莫凌超, 等. 阿尔泰南缘托库孜巴依金矿床流体包裹体特征及成矿流体演化[J]. 岩石学报, 2020, 36(4): 1171-1185.
[61]	XIAO W J, WINDLEY B F, HUANG B C, et al. End-Permian to mid-Triassic termination of theaccretionary processes of the southern Altaids: implications for the geodynamic evolution, Phanerozoic continental growth, and metallogeny of Central Asia[J]. International Journal of Earth Sciences, 2009, 98(6): 1189-1217.
[62]	KRÖNER A, ALEXEIEV D V, ROJAS-AGRAMONTE Y, et al. Mesoproterozoic (Grenville-age) terranes in the Kyrgyz North Tianshan: zircon ages and Nd-Hf isotopic constraints on the origin and evolution of basement blocks in the southern central Asian Orogen[J]. Gondwana Research, 2013, 23(1): 272-295.
[63]	徐九华, 张国瑞, 谢玉玲, 等. 阿尔泰山南缘赛都金矿床的构造-成矿流体及其演化[J]. 岩石矿物学杂志, 2009, 28(2): 141-151.
[64]	占深, 田忠锋, 宋梦莹, 等. 新疆哈巴河金矿带典型矿床稳定同位素特征[J]. 新疆地质, 2016, 34(4): 476-480.
[65]	DING Q F, WU C Z, SANTOSH M, et al. H-O, S and Pb isotope geochemistry of the Awanda gold deposit in southern Tianshan, Central Asian orogenic belt: implications for fluid regime and metallogeny[J]. Ore Geology Reviews, 2014, 62: 40-53.
[66]	SIBSON R H, ROBERT F, POULSEN K H. High-angle reverse faults, fluid-pressure cycling, and mesothermal gold-quartz deposits[J]. Geology, 1988, 16(6): 551.
[67]	王颖维, 徐九华, 卫晓锋, 等. 新疆金坝金矿成矿作用的地质和流体包裹体证据[J]. 地球科学, 2018, 43(9): 3112-3125.
[68]	成秋明. 空间模式的广义自相似性分析与矿产资源评价[J]. 地球科学: 中国地质大学学报, 2004, 29(6): 733-743.
[69]	成秋明, 张生元, 左仁广, 等. 多重分形滤波方法和地球化学信息提取技术研究与进展[J]. 地学前缘, 2009, 16(2): 185-198.
[70]	CHENG Q M. Non-linear theory and power-law models for information integration and mineral resources quantitative assessments[J]. Mathematical Geosciences, 2008, 40(5): 503-532.
[71]	MAO Z, PENG S, LAI J, et al. Fractal study of geochemical prospecting data in south area of Fenghuanshan copper deposit, Tongling, Anhui[J]. Journal of Earth Sciences and Environment, 2004, 26(4): 11-14.
[72]	HASSANPOUR S, AFZAL P. Application of concentration-number(C-N) multifractal modeling for geochemical anomaly separation in Haftcheshmeh porphyry system, NW Iran[J]. Arabian Journal of Geosciences, 2013, 6(3): 957-970.
[73]	ALI E H M, KHAKZAD A, LOTFI M, et al. Prospecting of Ni mineralization using fractal models based onlithogeochemical data in Patang area, eastern Iran[J]. Arabian Journal of Geosciences, 2015, 8(11): 9667-9677.
[74]	BREIMAN L. Random forests[J]. Machine Learning, 2001, 45(1): 5-32.
[75]	LUNDBERG S M, LEE S I. A unified approach to interpreting model predictions[C]// Proceedings of the 31st international conference on neural information processing systems. Red Hook: Curran Associates Inc., 2017: 4768-4777.
[76]	LUNDBERG S M, ERION G, CHEN H, et al. From local explanations to global understanding with explainable AI for trees[J]. Nature Machine Intelligence, 2020, 2(1): 56-67. DOI PMID
[77]	崔婷婷, 安雪莲, 孙德亮, 等. 基于SHAP的可解释机器学习的滑坡易发性评价模型[J]. 成都理工大学学报(自然科学版), 2025, 52(1): 153-172.
[78]	STIRNBERG R, CERMAK J, KOTTHAUS S, et al. Meteorology-driven variability of air pollution (PM₁) revealed with explainable machine learning[J]. Atmospheric Chemistry and Physics, 2021, 21(5): 3919-3948.
[79]	王世称, 陈永良. 大型、超大型金矿床综合信息成矿预测标志[J]. 黄金地质, 1999, 5(1): 1-5.
[80]	罗潇, 王彦国, 葛坤朋, 等. 基于方向tilt-Euler的三维磁数据快速反演[J]. 地球物理学报, 2021, 64(6): 2127-2140. DOI
[81]	韩梅, 张春灌, 李想, 等. 利用重磁资料研究楚科奇边缘地构造特征[J]. 地球物理学进展, 2022, 37(3): 945-951.
[82]	侯重初, 蔡宗熹, 刘奎俊. 从偶层位出发建立曲面上的位场转换解释系统[J]. 地球物理学报, 1985, 28(4): 410-418.
[83]	蔡宗熹. 关于磁性界面拟合反演公式及其推广[J]. 物化探计算技术, 1984, 6(2): 9-18.
[84]	杨林, 王庆飞, 赵世宇, 等. 造山型金矿构造控矿作用[J]. 岩石学报, 2023, 39(2): 277-292.
[85]	杨俊杰. 新疆阿尔泰克兰多金属矿集区成矿规律与成矿预测[D]. 西安: 长安大学, 2023.
[86]	曹根深, 张宇, 陈华勇. 造山型金矿床黄铁矿微量元素对成矿机制的指示[J]. 岩石学报, 2023, 39(8): 2330-2346.
[87]	BREWER C A, PICKLE L. Evaluation of methods for classifying epidemiological data on choropleth maps in series[J]. Annals of the Association of American Geographers, 2002, 92(4): 662-681.