High-order Statistics of Spatial Random Fields: Exploring Spatial Cumulants for Modeling Complex Non-Gaussian and Non-linear Phenomena

The spatial distributions of earth science and engineering phenomena under study are currently predicted from finite measurements and second-order geostatistical models. The latter models can be limiting, as geological systems are highly complex, non-Gaussian, and exhibit non-linear patterns of spatial connectivity. Non-linear and non-Gaussian high-order geostatistics based on spatial connectivity measures, namely spatial cumulants, are proposed as a new alternative modeling framework for spatial data. This framework has two parts. The first part is the definition, properties, and inference of spatial cumulants—including understanding the interrelation of cumulant characteristics with the in-situ behavior of geological entities or processes, as examined in this paper. The second part is the research on a random field model for simulation based on its high-order spatial cumulants.Mathematical definitions of non-Gaussian spatial random functions and their high-order spatial statistics are presented herein, stressing the notion of spatial cumulants. The calculation of spatial cumulants with spatial templates follows, including anisotropic experimental cumulants. Several examples of two- and three-dimensional images, including a diamond bearing kimberlite pipe from the Ekati Mine in Canada, are analyzed to assess the relations between cumulants and the spatial behavior of geological processes. Spatial cumulants of orders three to five are shown to capture directional multiple-point periodicity, connectivity including connectivity of extreme values, and spatial architecture. In addition, they provide substantial information on geometric characteristics and anisotropy of geological patterns. It is further shown that effects of complex spatial patterns are seen even if only subsets of all cumulant templates are computed. Compared to second-order statistics, cumulant maps are found to include a wealth of additional information from underlying geological patterns. Further work seeks to integrate this information in the predictive capabilities of a random field model.

[1]  M. David.  Handbook of Applied Advanced Geostatistical Ore Reserve Estimation , 1987 .

[2]  Noel A Cressie,et al.  Statistics for Spatial Data. , 1992 .

[3]  Andre G. Journel,et al.  Spatial Connectivity: From Variograms to Multiple-Point Measures , 2003 .

[4]  P. Kitanidis.  Introduction to Geostatistics: Applications in Hydrogeology , 1997 .

[5]  J. Chilès,et al.  Geostatistics: Modeling Spatial Uncertainty , 1999 .

[6]  A. Shiryaev.  Some Problems in the Spectral Theory of Higher-Order Moments. I , 1960 .

[7]  J. Caers,et al.  Conditional Simulation with Patterns , 2007 .

[8]  Sebastien Strebelle,et al.  Conditional Simulation of Complex Geological Structures Using Multiple-Point Statistics , 2002 .

[9]  Pablo Fosalba,et al.  Gravitational Evolution of the Large-Scale Probability Density Distribution: The Edgeworth and Gamma Expansions , 2000 .

[10]  R. M. Srivastava,et al.  Multivariate Geostatistics: Beyond Bivariate Moments , 1993 .

[11]  Haakon Tjelmeland,et al.  Directional Metropolis : Hastings Updates for Posteriors with Nonlinear Likelihoods , 2005 .

[12]  C. Deutsch,et al.  Geostatistics Banff 2004 , 2005 .

[13]  R. Reese.  Geostatistics for Environmental Scientists , 2001 .

[14]  B. Matérn.  Spatial variation : Stochastic models and their application to some problems in forest surveys and other sampling investigations , 1960 .

[15]  L. Hu,et al.  Multiple-Point Simulations Constrained by Continuous Auxiliary Data , 2008 .

[16]  Jerry M. Mendel,et al.  Tutorial on higher-order statistics (spectra) in signal processing and system theory: theoretical results and some applications , 1991, Proc. IEEE.

[17]  Paul Switzer,et al.  Filter-Based Classification of Training Image Patterns for Spatial Simulation , 2006 .

[18]  M. Rosenblatt,et al.  ASYMPTOTIC THEORY OF ESTIMATES OF kTH-ORDER SPECTRA. , 1967, Proceedings of the National Academy of Sciences of the United States of America.

[19]  Feng Zhang,et al.  A High Order Cumulants Based Multivariate Nonlinear Blind Source Separation Method , 2005, Machine Learning.

[20]  Geoffrey S. Watson,et al.  Geostatistical Ore Reserve Estimation. , 1978 .

[21]  M. Rosenblatt.  Stationary sequences and random fields , 1985 .

[22]  Mike Rees,et al.  5. Statistics for Spatial Data , 1993 .

[23]  Daniel M. Tetzlaff,et al.  Stationarity Scores on Training Images for Multipoint Geostatistics , 2009 .

[24]  Colin Daly,et al.  Higher Order Models using Entropy, Markov Random Fields and Sequential Simulation , 2005 .

[25]  Jef Caers,et al.  Representing Spatial Uncertainty Using Distances and Kernels , 2009 .

[26]  Noel A. C. Cressie,et al.  Statistics for Spatial Data: Cressie/Statistics , 1993 .

[27]  B. Harris.  Spectral Analysis Of Time Series , 1967 .

[28]  Alexandre Boucher,et al.  Applied Geostatistics with SGeMS: A User's Guide , 2009 .

[29]  E.Gaztanaga,et al.  Gravitational Evolution of the Large-Scale Probability Density Distribution: The Edgeworth&Gamma Expansions , 1999 .

[30]  T. Nowicki,et al.  The geology of kimberlite pipes of the Ekati property, Northwest Territories, Canada , 2004 .

[31]  R. Dimitrakopoulos,et al.  High-order Stochastic Simulation of Complex Spatially Distributed Natural Phenomena , 2010 .

[32]  R. Dimitrakopoulos,et al.  Two-dimensional Conditional Simulations Based on the Wavelet Decomposition of Training Images , 2009 .

[33]  Håkon Tjelmeland,et al.  Markov Random Fields with Higher‐order Interactions , 1998 .

[34]  S. Dreiss,et al.  Hydrostratigraphic interpretation using indicator geostatistics , 1989 .

[35]  Alexandre Boucher,et al.  Considering complex training images with search tree partitioning , 2009, Comput. Geosci..

[36]  C. L. Nikias,et al.  Higher-order spectra analysis : a nonlinear signal processing framework , 1993 .

[37]  D. Cox.  Regression Models and Life-Tables , 1972 .

[38]  Alexandre Boucher,et al.  A SGeMS code for pattern simulation of continuous and categorical variables: FILTERSIM , 2008, Comput. Geosci..

[39]  Timothy C. Coburn,et al.  Geostatistics for Natural Resources Evaluation , 2000, Technometrics.