Detection of Geochemical Element Assemblage Anomalies Using a Local Correlation Approach

Abstract

As direct prospecting data, geochemical data play an important role in modelling prospect potential. Geochemical element assemblage anomalies are usually reflected by the correlation between elements. Correlation coefficients are computed from the values of two elements, which reflect only the correlation at a global level. Thus, the spatial details of the correlation structure are ignored. In fact, an element combination anomaly often exists in geological backgrounds, such as on a fault zone or within a lithological unit. This anomaly may cause some combination of anomalies that are submerged inside the overall area and thus cannot be effectively extracted. To address this problem, we propose a local correlation coefficient based on spatial neighbourhoods to reflect the global distribution of elements. In this method, the sampling area is first divided into a set of uniform grid cells. A moving window with a size of 3×3 is defined with an integer of 3 to represent the sampling unit. The local correlation in each unit is expressed by the Pearson correlation coefficient. The whole area is scanned by the moving window, which produces a correlation coefficient matrix, and the result is portrayed with a thermal diagram. The local correlation approach was tested on two selected geochemical soil survey sites in Xiao Mountain, Henan Province. The results show that the areas of high correlation are mainly distributed in the fault zone or the known mineral spots. Therefore, the local correlation method is effective in extracting geochemical element combination anomalies.

Keywords

local correlation coefficients / geochemical prospecting / geochemical soil survey / assemblage anomalies / geodata mining

References

[1] Attias H. Independent Factor Analysis. Neural Computation, 1999, 11(4): 803-851.

[2] Attias, H., 1999b. Learning a Hierarchical Belief Network of Independent Factor Analyzers. Proceedings of the 1998 Conference on Advances in Neural Information Processing Systems II. July 1999, 361–367

[3] Attias H. Independent Factor Analysis with Temporally Structured Sources. Advances in Neural Information Processing Systems, 2000, 12: 386-392.

[4] Bendich P, Gasparovic E, Harer J, . Erin C, Brittany F, Lori Z, . Scaffoldings and Spines: Organizing High-Dimensional Data Using Cover Trees, Local Principal Component Analysis, and Persistent Homology. Research in Computational Topology, 2018, Cham: Springer, 93-114.

[5] Chen G X, Cheng Q M. Singularity Analysis Based on Wavelet Transform of Fractal Measures for Identifying Geochemical Anomaly in Mineral Exploration. Computers & Geosciences, 2016, 87: 56-66.

[6] Cheng Q M, Bonham-Carter G, Wang W L, . A Spatially Weighted Principal Component Analysis for Multi-Element Geochemical Data for Mapping Locations of Felsic Intrusions in the Gejiu Mineral District of Yunnan, China. Computers & Geosciences, 2011, 37(5): 662-669.

[7] Cheng Q M, Agterberg F P. Singularity Analysis of Ore-Mineral and Toxic Trace Elements in Stream Sediments. Computers & Geosciences, 2009, 35(2): 234-244.

[8] Coston N, O’Rourke S. Gaussian Fluctuations for Linear Eigenvalue Statistics of Products of Independent Iid Random Matrices. Journal of Theoretical Probability, 2020, 33(1): 1541-1612.

[9] Gielen E, Riutort-Mayol G, Palencia-Jiménez J S, . An Urban Sprawl Index Based on Multivariate and Bayesian Factor Analysis with Application at the Municipality Level in Valencia. Environment and Planning B: Urban Analytics and City Science, 2018, 45(5): 888-914

[10] Hartono, Sitompul O S, Tulus, . Optimization Model of K-Means Clustering Using Artificial Neural Networks to Handle Class Imbalance Problem. IOP Conference Series: Materials Science and Engineering, 2018, 288(1): 012075

[11] Hu S, Luo T, Jing C Y. Principal Component Analysis of Fluoride Geochemistry of Groundwater in Shanxi and Inner Mongolia, China. Journal of Geochemical Exploration, 2013, 135: 124-129.

[12] Isomura T, Toyoizumi T. Independent Component Analysis. Scientific Reports, 2018, 8(1): 1-11

[13] Jiang W C, Li H, Wu J H, . A Newly Found Biotite Syenogranite in the Huangshaping Polymetallic Deposit, South China: Insights into Cu Mineralization. Journal of Earth Science, 2018, 29(3): 537-555.

[14] Li J, Zhuang X G, Querol X, . New Data on Mineralogy and Geochemistry of High-Ge Coals in the Yimin Coalfield, Inner Mongolia, China. International Journal of Coal Geology, 2014, 125(5): 10-21.

[15] Markhvida M, Ceferino L, Baker J W. Modeling Spatially Correlated Spectral Accelerations at Multiple Periods Using Principal Component Analysis and Geostatistics. Earthquake Engineering & Structural Dynamics, 2018, 47(5): 1107-1123.

[16] Niven E B, Deutsch C V. Calculating a Robust Correlation Coefficient and Quantifying Its Uncertainty. Computers & Geosciences, 2012, 40(3): 1-9.

[17] Painsky A, Rosset S, Feder M. Linear Independent Component Analysis over Finite Fields: Algorithms and Bounds. IEEE Transactions on Signal Processing, 2018, 66(22): 5875-5886.

[18] Park S, Kwak N. Independent Component Analysis by Lp-Norm Optimization. Pattern Recognition, 2018, 76(3): 752-760.

[19] Rahimi E, Maghsoudi A, Hezarkhani A. Geochemical Investigation and Statistical Analysis on Rare Earth Elements in Lakehsiyah Deposit, Bafq District. Journal of African Earth Sciences, 2016, 124 139-150.

[20] Scitovski S. A Density-Based Clustering Algorithm for Earthquake Zoning. Computers & Geosciences, 2018, 110: 90-95.

[21] Simon, N., Tibshirani, R., 2013. Comment on “Detecting Novel Associations in Large Data Sets” by Reshef et al., Science Dec 16, 2011. Unpublished (available at wwwstat.stanford.edu/tibs/reshef/comment.pdf)

[22] Smillie C. Salicornia Spp. as a Biomonitor of Cu and Zn in Salt Marsh Sediments. Ecological Indicators, 2015, 56(3b): 70-78.

[23] Song L, Jian J, Tan D J, . Estimate of Heavy Metals in Soil and Streams Using Combined Geochemistry and Field Spectroscopy in Wan-Sheng Mining Area, Chongqing, China. International Journal of Applied Earth Observation and Geoinformation, 2015, 34: 1-9.

[24] Wang J P, Liu Z J, Liu J J, . Trace Element Compositions of Pyrite from the Shuangwang Gold Breccias, Western Qinling Orogen, China: Implications for Deep Ore Prediction. Journal of Earth Science, 2018, 29(3): 564-572.

[25] Wang H C, Cheng Q M, Zuo R G. Spatial Characteristics of Geochemical Patterns Related to Fe Mineralization in the Southwestern Fujian Province (China). Journal of Geochemical Exploration, 2015, 148(7): 259-269.

[26] Xu W, Chang C, Hung Y S, . Order Statistics Correlation Co-efficient as a Novel Association Measurement with Applications to Biosignal Analysis. IEEE Transactions on Signal Processing, 2007, 55(12): 5552-5563.

[27] Xue J H, Lee C, Wakeham S G, . Using Principal Components Analysis (PCA) with Cluster Analysis to Study the Organic Geochemistry of Sinking Particles in the Ocean. Organic Geochemistry, 2011, 42(4): 356-367.

[28] Yu X C, Liu L W, Hu D, . Robust Ordinal Independent Component Analysis (ROICA) Applied to Mineral Resources Prediction. Journal of Jilin University (Earth Science Edition), 2012, 42(3): 872-880

[29] Yu X C, Hu D, Xu J. Blind Source Separation: Theory and Applications, 2014, New York: John Wiley & Sons

[30] Zhang Z W, Shu Q, Yang X Y, . Review on the Tectonic Terranes Associated with Metallogenic Zones in Southeast Asia. Journal of Earth Science, 2019, 30(1): 1-19.

[31] Zuo R G, Wang J, Chen G X, . Identification of Weak Anomalies: A Multifractal Perspective. Journal of Geochemical Exploration, 2015, 148: 12-24.