聚类分析算法在大地电磁三维解释中的应用
|
摘要
为了提高对大地电磁三维反演结果的分析和解释能力,完成构造界面识别、异常构造刻画等地质和地球物理解释,本文提出使用非监督的聚类方法分析大地电磁反演结果.根据三维反演模型电阻率值的分布和各种聚类分析方法的特点,选择使用K均值聚类方法对反演模型电阻率值进行聚类分析.在K均值聚类分析过程中,本文采用了RS指数指导选择聚类数目, Kaufman法进行中心初始化.通过对东北地区大地电磁数据三维反演结果使用K均值聚类分析方法,得到了东北地区电性岩石圈的厚度估计,结果表明东北地区岩石圈底部聚类电阻率大约为339Ω·m,其中松辽盆地岩石圈最薄,约为60 km;大兴安岭地区最厚,约为150 km;佳木斯地体厚度约为100 km;而长白山地区岩石圈厚度不易确定,可能受新生代构造活动影响,电阻率明显减小.聚类方法能够有效地帮助对大地电磁三维反演结果中的地质构造进行识别和归类.
In order to improve the ability of analyzing and interpreting the results of magnetotelluric three-dimensional inversion and complete geological and geophysical interpretation such as structural interface recognition and anomalous structures characterization, this paper proposes an unsupervised clustering method to analyze the results of magnetotelluric 3 D inversion. According to the distribution of resistivity values in magnetotelluric 3 D inversion model and the characteristics of various clustering analysis methods, the K-means clustering method is selected to analyze the resistivity values in inversion results. In the process of the K-means clustering method, this paper uses RS index to guide the selection of clustering number and Kaufman method to initialize the clustering centers. By using the K-means clustering methods to analyze the 3 D inversion results of magnetotelluric data in Northeast China, the thickness estimation of electrical lithosphere in Northeast China is obtained. The result shows that the cluster resistivity at the bottom of the lithosphere in Northeast China is about 339 Ω·m. The lithosphere of Songliao Basin is the thinnest, about 60 km; the lithosphere of Great Xing'an Range is the thickest, about 150 km; Jiamusi Block is about 100 km; and the lithosphere thickness of Changbai Mountain Range is not easy to determine, which may be influenced by Cenozoic tectonic activities, and its resistivity is obviously reduced. The result shows that the clustering methods can effectively help to identify and classify the geotectonic structures in the results of magnetotelluric 3 D inversion.
In order to improve the ability of analyzing and interpreting the results of magnetotelluric three-dimensional inversion and complete geological and geophysical interpretation such as structural interface recognition and anomalous structures characterization, this paper proposes an unsupervised clustering method to analyze the results of magnetotelluric 3 D inversion. According to the distribution of resistivity values in magnetotelluric 3 D inversion model and the characteristics of various clustering analysis methods, the K-means clustering method is selected to analyze the resistivity values in inversion results. In the process of the K-means clustering method, this paper uses RS index to guide the selection of clustering number and Kaufman method to initialize the clustering centers. By using the K-means clustering methods to analyze the 3 D inversion results of magnetotelluric data in Northeast China, the thickness estimation of electrical lithosphere in Northeast China is obtained. The result shows that the cluster resistivity at the bottom of the lithosphere in Northeast China is about 339 Ω·m. The lithosphere of Songliao Basin is the thinnest, about 60 km; the lithosphere of Great Xing'an Range is the thickest, about 150 km; Jiamusi Block is about 100 km; and the lithosphere thickness of Changbai Mountain Range is not easy to determine, which may be influenced by Cenozoic tectonic activities, and its resistivity is obviously reduced. The result shows that the clustering methods can effectively help to identify and classify the geotectonic structures in the results of magnetotelluric 3 D inversion.
引文
Audebert M, Clément R, Touze-Foltz N, et al. 2014. Time-lapse ERT interpretation methodology for leachate injection monitoring based on multiple inversions and a clustering strategy (MICS) [J]. Journal of Applied Geophysics, 111: 320-333, doi: 10.1016/j.jappgeo.2014.09.024.
Ball G H, Hall D J. 1965. Isodata: A novel method of data analysis and pattern classification [J]. Isodata A Novel Method of Data Analysis & Pattern Classification.
Bosch M, Zamora M, Utama W. 2002. Lithology discrimination from physical rock properties [J]. Geophysics, 67(2): 573-581, doi: 10.1190/1.1468618.
Eaton D W, Darbyshire F, Evans R L, et al. 2009. The elusive lithosphere-asthenosphere boundary (LAB) beneath cratons [J]. Lithos, 109(1-2): 1-22, doi: 10.1016/j.lithos.2008.05.009.
Forgy E W. 1965. Cluster analysis of multivariate data: efficiency versus interpretability of classification [J]. Biometrics, 21(3): 768-769.
Giuseppe M G D, Troiano A, Patella D, et al. 2018. A geophysical k-means cluster analysis of the Solfatara-Pisciarelli volcano-geothermal system, Campi Flegrei (Naples, Italy) [J]. Journal of Applied Geophysics, 156: 44-54, doi:10.1016/j.jappgeo.2017.06.001.
Giuseppe M G D, Troiano A, Troise C, et al. 2014. k-Means clustering as tool for multivariate geophysical data analysis. An application to shallow fault zone imaging [J]. Journal of Applied Geophysics, 101: 108-115, doi: 10.1016/j.jappgeo.2013.12.004.
Guo Z, Cao Y, Wang X, et al. 2014. Crust and upper mantle structures beneath Northeast China from receiver function studies [J]. Earthquake Science, 27(3): 265-275, doi: 10.1007/s11589- 014- 0076-x.
Halkidi M, Vazirgiannis M, Batistakis Y. 2000. Quality Scheme Assessment in the Clustering Process [J]. Lecture Notes in Computer Science, 1910(1): 265-276, doi: 10.1007/3-540- 45372-5_26.
Han J T, Guo Z Y, Liu W Y, et al. 2018. Deep dynamic process of lithosphere thinning in Songliao basin [J]. Chinese J. Geophys. (in Chinese), 61(6): 2265-2279, doi: 10.6038/cjg2018L0155.
HE L, WU L D, CAI Y C. 2007. Survey of Clustering Algorithms in Data Mining [J]. Application Research of Computers (in Chinese), 24(1): 10-13, doi: 10.3969/j.issn.1001-3695.2007.01.003.
Jain A K. 2010. Data Clustering: 50 Years Beyond K-means [J]. Pattern Recognition Letters, 31(8): 651- 666, doi: 10.1016/j.patrec.2009.09.011.
Katsavounidis I, Jay Kuo C C, Zhang Z. 1994. A new initialization technique for generalized Lloyd iteration [J]. Signal Processing Letters IEEE, 1(10): 144-146, doi: 10.1109/97.329844.
Kaufman L, Rousseeuw P J. 1990. Finding Groups in Data: An Introduction to Cluster Analysis [M]. DBLP,
Li J P. 2018. Deep Electrical Structure and Rheological Analysis of Cenozoic Volcanism in Northeast China:[D]. Changchun:Jilin University.
Lloyd S. 1982. Least squares quantization in PCM [J]. IEEE Transactions on Information Theory, 28(2): 129-137, doi: 10.1109/TIT.1982.1056489.
Macqueen J B. 1967. Some methods for classification and analysis of multivariate observations [C]// 5-Th Berkeley Symposium on Mathematical Statistics and Probability.University of California Press.
Paasche H, Tronicke J. 2007. Cooperative inversion of 2D geophysical data sets: A zonal approach based on fuzzy c-means cluster analysis [J]. Geophysics, 72(3): A35-A39.
Reynolds J M. 1997. An Introduction to Applied and Environmental Geophysics [M]. New York: John Wiley and Sons.
Scitovski S. 2017. A density-based clustering algorithm for earthquake zoning [J]. Computers & Geosciences, 110: 90-95, doi: 10.1016/j.cageo.2017.08.014.
Sharma S C. 1996. Applied Multivariate Techniques [M]. New York: John Wiley and Sons.
Shi Y J, Liu G D, Wu G Y. 1985. Course of Magnetotelluric Sounding (in Chinese) [M]. Beijing: Seismological Press.
Song Y C, Meng H D, O’Grady M J, et al. 2010. The application of cluster analysis in geophysical data interpretation [J]. Computational Geosciences, 14(2): 263-271, doi: 10.1007/s10596- 009-9150-1.
Steinhaus H. 1956. Sur la division des corps matériels en parties [J]. Bull. Acad. Polon. Sci. IV(C1.III), 801-804.
Sun J G, Liu J, Zhao L Y. 2008. Clustering algorithms research [J]. Journal of Software (in Chinese), 19(1): 48- 61, doi: 10.3724/SP.J.1001.2008.00048.
Tibshirani R, Walther G, Hastie T. 2001. Estimating the number of clusters in a data set via the gap statistic [J]. Journal of the Royal Statistical Society, 63(2): 411- 423.
韩江涛, 郭振宇, 刘文玉, 等. 2018. 松辽盆地岩石圈减薄的深部动力学过程[J]. 地球物理学报, 61(06): 2265-2279, doi: 10.6038/cjg2018L0155.
贺玲, 吴玲达, 蔡益朝. 2007. 数据挖掘中的聚类算法综述[J]. 计算机应用研究, 24(1): 10-13, doi: 10.3969/j.issn.1001-3695.2007.01.003.
李建平. 2018. 中国东北地区新生代火山深部电性结构及流变分析[D]. 长春: 吉林大学.
石应骏, 刘国栋, 吴广耀, 等. 1985. 大地电磁测深法教程[M]. 北京: 地质出版社.
孙吉贵, 刘杰, 赵连宇. 2008. 聚类算法研究[J]. 软件学报, 19(1): 48- 61, doi: 10.3724/SP.J.1001.2008.00048.
Ball G H, Hall D J. 1965. Isodata: A novel method of data analysis and pattern classification [J]. Isodata A Novel Method of Data Analysis & Pattern Classification.
Bosch M, Zamora M, Utama W. 2002. Lithology discrimination from physical rock properties [J]. Geophysics, 67(2): 573-581, doi: 10.1190/1.1468618.
Eaton D W, Darbyshire F, Evans R L, et al. 2009. The elusive lithosphere-asthenosphere boundary (LAB) beneath cratons [J]. Lithos, 109(1-2): 1-22, doi: 10.1016/j.lithos.2008.05.009.
Forgy E W. 1965. Cluster analysis of multivariate data: efficiency versus interpretability of classification [J]. Biometrics, 21(3): 768-769.
Giuseppe M G D, Troiano A, Patella D, et al. 2018. A geophysical k-means cluster analysis of the Solfatara-Pisciarelli volcano-geothermal system, Campi Flegrei (Naples, Italy) [J]. Journal of Applied Geophysics, 156: 44-54, doi:10.1016/j.jappgeo.2017.06.001.
Giuseppe M G D, Troiano A, Troise C, et al. 2014. k-Means clustering as tool for multivariate geophysical data analysis. An application to shallow fault zone imaging [J]. Journal of Applied Geophysics, 101: 108-115, doi: 10.1016/j.jappgeo.2013.12.004.
Guo Z, Cao Y, Wang X, et al. 2014. Crust and upper mantle structures beneath Northeast China from receiver function studies [J]. Earthquake Science, 27(3): 265-275, doi: 10.1007/s11589- 014- 0076-x.
Halkidi M, Vazirgiannis M, Batistakis Y. 2000. Quality Scheme Assessment in the Clustering Process [J]. Lecture Notes in Computer Science, 1910(1): 265-276, doi: 10.1007/3-540- 45372-5_26.
Han J T, Guo Z Y, Liu W Y, et al. 2018. Deep dynamic process of lithosphere thinning in Songliao basin [J]. Chinese J. Geophys. (in Chinese), 61(6): 2265-2279, doi: 10.6038/cjg2018L0155.
HE L, WU L D, CAI Y C. 2007. Survey of Clustering Algorithms in Data Mining [J]. Application Research of Computers (in Chinese), 24(1): 10-13, doi: 10.3969/j.issn.1001-3695.2007.01.003.
Jain A K. 2010. Data Clustering: 50 Years Beyond K-means [J]. Pattern Recognition Letters, 31(8): 651- 666, doi: 10.1016/j.patrec.2009.09.011.
Katsavounidis I, Jay Kuo C C, Zhang Z. 1994. A new initialization technique for generalized Lloyd iteration [J]. Signal Processing Letters IEEE, 1(10): 144-146, doi: 10.1109/97.329844.
Kaufman L, Rousseeuw P J. 1990. Finding Groups in Data: An Introduction to Cluster Analysis [M]. DBLP,
Li J P. 2018. Deep Electrical Structure and Rheological Analysis of Cenozoic Volcanism in Northeast China:[D]. Changchun:Jilin University.
Lloyd S. 1982. Least squares quantization in PCM [J]. IEEE Transactions on Information Theory, 28(2): 129-137, doi: 10.1109/TIT.1982.1056489.
Macqueen J B. 1967. Some methods for classification and analysis of multivariate observations [C]// 5-Th Berkeley Symposium on Mathematical Statistics and Probability.University of California Press.
Paasche H, Tronicke J. 2007. Cooperative inversion of 2D geophysical data sets: A zonal approach based on fuzzy c-means cluster analysis [J]. Geophysics, 72(3): A35-A39.
Reynolds J M. 1997. An Introduction to Applied and Environmental Geophysics [M]. New York: John Wiley and Sons.
Scitovski S. 2017. A density-based clustering algorithm for earthquake zoning [J]. Computers & Geosciences, 110: 90-95, doi: 10.1016/j.cageo.2017.08.014.
Sharma S C. 1996. Applied Multivariate Techniques [M]. New York: John Wiley and Sons.
Shi Y J, Liu G D, Wu G Y. 1985. Course of Magnetotelluric Sounding (in Chinese) [M]. Beijing: Seismological Press.
Song Y C, Meng H D, O’Grady M J, et al. 2010. The application of cluster analysis in geophysical data interpretation [J]. Computational Geosciences, 14(2): 263-271, doi: 10.1007/s10596- 009-9150-1.
Steinhaus H. 1956. Sur la division des corps matériels en parties [J]. Bull. Acad. Polon. Sci. IV(C1.III), 801-804.
Sun J G, Liu J, Zhao L Y. 2008. Clustering algorithms research [J]. Journal of Software (in Chinese), 19(1): 48- 61, doi: 10.3724/SP.J.1001.2008.00048.
Tibshirani R, Walther G, Hastie T. 2001. Estimating the number of clusters in a data set via the gap statistic [J]. Journal of the Royal Statistical Society, 63(2): 411- 423.
韩江涛, 郭振宇, 刘文玉, 等. 2018. 松辽盆地岩石圈减薄的深部动力学过程[J]. 地球物理学报, 61(06): 2265-2279, doi: 10.6038/cjg2018L0155.
贺玲, 吴玲达, 蔡益朝. 2007. 数据挖掘中的聚类算法综述[J]. 计算机应用研究, 24(1): 10-13, doi: 10.3969/j.issn.1001-3695.2007.01.003.
李建平. 2018. 中国东北地区新生代火山深部电性结构及流变分析[D]. 长春: 吉林大学.
石应骏, 刘国栋, 吴广耀, 等. 1985. 大地电磁测深法教程[M]. 北京: 地质出版社.
孙吉贵, 刘杰, 赵连宇. 2008. 聚类算法研究[J]. 软件学报, 19(1): 48- 61, doi: 10.3724/SP.J.1001.2008.00048.