Some theory for anisotropic processes on the sphere

Abstract We investigate properties of an existing class of models for Gaussian processes on the sphere that are invariant to shifts in longitude. The class is obtained by applying first-order differential operators to an isotropic process and potentially adding an independent isotropic term. For a particular choice of the operators, we derive explicit forms for the spherical harmonic representation of these processes’ covariance functions. Because the spherical harmonic representation is a spectral one, these forms allow us to draw conclusions about the local properties of the processes. For one, the coefficients in the spherical harmonic representation relate to the equivalence and orthogonality of the measures induced by the models. It turns out that under certain conditions the models will lack consistent parameter estimability even when the process is observed everywhere on the sphere. We also consider the ability of the models to capture isotropic tendencies on the local level, a phenomenon observed in some data.

[1]  J. Holton An introduction to dynamic meteorology , 2004 .

[2]  Mikyoung Jun,et al.  An Approach to Producing Space–Time Covariance Functions on Spheres , 2007, Technometrics.

[3]  Noel A Cressie,et al.  Global statistical analysis of MISR aerosol data: a massive data product from NASA's Terra satellite , 2007 .

[4]  F. Lindgren,et al.  Spatial stochastic models generated by nested stochastic partial differential equations , 2009 .

[5]  T. Gneiting Correlation functions for atmospheric data analysis , 1999 .

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

[7]  A. Yaglom Correlation Theory of Stationary and Related Random Functions I: Basic Results , 1987 .

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

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

[10]  Michael L. Stein,et al.  Some statistical methods for analyzing the TOMS data , 1998 .

[11]  M. Fuentes Approximate Likelihood for Large Irregularly Spaced Spatial Data , 2007, Journal of the American Statistical Association.

[12]  Irene A. Stegun,et al.  Handbook of Mathematical Functions. , 1966 .

[13]  Roger Woodard,et al.  Interpolation of Spatial Data: Some Theory for Kriging , 1999, Technometrics.

[14]  Multi-angle Imaging SpectroRadiometer , 2011 .

[15]  Steven Kay,et al.  Gaussian Random Processes , 1978 .

[16]  M. Stein Space–Time Covariance Functions , 2005 .