A MULTI-RESOLUTION MODEL FOR NON-GAUSSIAN RANDOM FIELDS ON A SPHERE WITH APPLICATION TO IONOSPHERIC ELECTROSTATIC POTENTIALS.

Gaussian random fields have been one of the most popular tools for analyzing spatial data. However, many geophysical and environmental processes often display non-Gaussian characteristics. In this paper, we propose a new class of spatial models for non-Gaussian random fields on a sphere based on a multi-resolution analysis. Using a special wavelet frame, named spherical needlets, as building blocks, the proposed model is constructed in the form of a sparse random effects model. The spatial localization of needlets, together with carefully chosen random coefficients, ensure the model to be non-Gaussian and isotropic. The model can also be expanded to include a spatially varying variance profile. The special formulation of the model enables us to develop efficient estimation and prediction procedures, in which an adaptive MCMC algorithm is used. We investigate the accuracy of parameter estimation of the proposed model, and compare its predictive performance with that of two Gaussian models by extensive numerical experiments. Practical utility of the proposed model is demonstrated through an application of the methodology to a data set of high-latitude ionospheric electrostatic potentials, generated from the LFM-MIX model of the magnetosphere-ionosphere system.

[1]  T. Gneiting.  Strictly and non-strictly positive definite functions on spheres , 2011, 1111.7077.

[2]  Timothy Fuller-Rowell,et al.  On the importance of E‐field variability for Joule heating in the high‐latitude thermosphere , 1995 .

[3]  G. Peccati,et al.  Random Fields on the Sphere: Representation, Limit Theorems and Cosmological Applications , 2011 .

[4]  O. Barndorff-Nielsen,et al.  Models for non-Gaussian variation, with applications to turbulence , 1979, Proceedings of the Royal Society of London. A. Mathematical and Physical Sciences.

[5]  A. Raftery,et al.  Probabilistic forecasts, calibration and sharpness , 2007 .

[6]  Timothy Fuller-Rowell,et al.  Electric field variability associated with the Millstone Hill electric field model , 2000 .

[7]  Yuan Xu,et al.  Approximation Theory and Harmonic Analysis on Spheres and Balls , 2013 .

[8]  M. Fan.  A Note on Spherical Needlets , 2015, 1508.05406.

[9]  Tomoko Matsuo,et al.  Mesoscale and large‐scale variability in high‐latitude ionospheric convection: Dominant modes and spatial/temporal coherence , 2013 .

[10]  S. Godsill,et al.  Bayesian variable selection and regularization for time–frequency surface estimation , 2004 .

[11]  Montserrat Fuentes,et al.  Isotropic covariance functions on spheres: Some properties and modeling considerations , 2016, J. Multivar. Anal..

[12]  Pencho Petrushev,et al.  Localized Tight Frames on Spheres , 2006, SIAM J. Math. Anal..

[13]  A. Raftery,et al.  Strictly Proper Scoring Rules, Prediction, and Estimation , 2007 .

[14]  M. West.  On scale mixtures of normal distributions , 1987 .

[15]  Marc G. Genton,et al.  Tukey g-and-h Random Fields , 2017 .

[16]  Mark F. J. Steel,et al.  Non-Gaussian Bayesian Geostatistical Modeling , 2006 .

[17]  K. Gorski,et al.  HEALPix: A Framework for High-Resolution Discretization and Fast Analysis of Data Distributed on the Sphere , 2004, astro-ph/0409513.

[18]  K. Atkinson,et al.  Spherical Harmonics and Approximations on the Unit Sphere: An Introduction , 2012 .

[19]  J. M. Ruohoniemi,et al.  Large-scale imaging of high-latitude convection with Super Dual Auroral Radar Network HF radar observations , 1998 .

[20]  Michael Wiltberger,et al.  Structure of High Latitude Currents in Magnetosphere-Ionosphere Models , 2016 .

[21]  Philip Sura,et al.  Climatology of Non-Gaussian Atmospheric Statistics , 2013 .

[22]  Tomoko Matsuo,et al.  Modes of high‐latitude electric field variability derived from DE‐2 measurements: Empirical Orthogonal Function (EOF) analysis , 2001 .

[23]  John Lyon,et al.  The Lyon-Fedder-Mobarry (LFM) global MHD magnetospheric simulation code , 2004 .

[24]  Mikyoung Jun,et al.  Nonstationary covariance models for global data , 2008, 0901.3980.

[25]  Stephan R. Sain,et al.  Emulating and calibrating the multiple‐fidelity Lyon–Fedder–Mobarry magnetosphere–ionosphere coupled computer model , 2015 .

[26]  Tomoko Matsuo,et al.  High‐latitude ionospheric electric field variability and electric potential derived from DE‐2 plasma drift measurements: Dependence on IMF and dipole tilt , 2003 .

[27]  Simon George Shepherd,et al.  Statistical characteristics of small‐scale spatial and temporal electric field variability in the high‐latitude ionosphere , 2012 .

[28]  Stephan R. Sain,et al.  Feature‐based validation of the Lyon‐Fedder‐Mobarry magnetohydrodynamical model , 2016 .

[29]  Matthew J. Heaton,et al.  Parameter tuning for a multi-fidelity dynamical model of the magnetosphere , 2013, 1303.6992.

[30]  Christophe Andrieu,et al.  A tutorial on adaptive MCMC , 2008, Stat. Comput..

[31]  Feng Liang,et al.  Bayesian function estimation using continuous wavelet dictionaries , 2009 .

[32]  Michael L. Stein,et al.  Asymptotically Efficient Prediction of a Random Field with a Misspecified Covariance Function , 1988 .

[33]  Robert S. Womersley,et al.  Efficient Spherical Designs with Good Geometric Properties , 2017, 1709.01624.

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

[35]  D. Weimer,et al.  Models of high‐latitude electric potentials derived with a least error fit of spherical harmonic coefficients , 1995 .

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

[37]  Pekka Janhunen,et al.  Assessment of ionospheric Joule heating by GUMICS-4 MHD simulation, AMIE, and satellite-based statistics: towards a synthesis , 2005 .

[38]  N. Cressie,et al.  Fixed Rank Filtering for Spatio-Temporal Data , 2010 .

[39]  Marc G. Genton,et al.  Cross-Covariance Functions for Multivariate Geostatistics , 2015, 1507.08017.

[40]  Michael L. Stein,et al.  Spatial variation of total column ozone on a global scale , 2007, 0709.0394.

[41]  T. Gneiting,et al.  Comparing Density Forecasts Using Threshold- and Quantile-Weighted Scoring Rules , 2011 .

[42]  Debashis Paul,et al.  Modeling Tangential Vector Fields on a Sphere , 2016, Journal of the American Statistical Association.

[43]  Richard H. Jones,et al.  Stochastic Processes on a Sphere , 1963 .

[44]  N. Cressie,et al.  Fixed rank kriging for very large spatial data sets , 2008 .

[45]  Anand Natarajan,et al.  Gaussian vs non‐Gaussian turbulence: impact on wind turbine loads , 2014 .

[46]  R. Hunsucker,et al.  The High-Latitude Ionosphere and Its Effects on Radio Propagation , 2002 .

[47]  D. Bolin,et al.  Geostatistical Modelling Using Non‐Gaussian Matérn Fields , 2015 .

[48]  Gyorgy Terdik,et al.  Angular Spectra for non-Gaussian Isotropic Fields , 2013, 1302.4049.

[49]  Tomoko Matsuo,et al.  SuperDARN assimilative mapping , 2013 .

[50]  P. Baldi,et al.  Asymptotics for spherical needlets , 2006, math/0606599.

[51]  Tomoko Matsuo,et al.  Effects of high-latitude ionospheric electric field variability on global thermospheric Joule heating and mechanical energy transfer rate , 2007 .

[52]  Henning Omre,et al.  T-distributed Random Fields : A Parametric Model for Heavy-tailed Random Fields , 2006 .

[53]  B. Kedem,et al.  Bayesian Prediction of Transformed Gaussian Random Fields , 1997 .

[54]  H. Rue,et al.  An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach , 2011 .

[55]  B. Mallick,et al.  Fast sampling with Gaussian scale-mixture priors in high-dimensional regression. , 2015, Biometrika.

[56]  P. A. Damiano,et al.  A statistical study of magnetosphere–ionosphere coupling in the Lyon–Fedder–Mobarry global MHD model , 2011 .

[57]  I. J. Schoenberg.  Positive definite functions on spheres , 1942 .