## Abstract

This paper presents the characterization of the covariance matrix function of a Gaussian or second-order elliptically contoured vector random field on the sphere which is stationary, isotropic, and mean square continuous. This characterization involves an infinite sum of the products of positive definite matrices and Gegenbauer’s polynomials, and may not be available for other non-Gaussian vector random fields on spheres such as a *χ*
^{2} or log-Gaussian vector random field. We also offer two simple but efficient constructing approaches, and derive some parametric covariance matrix structures on spheres.

This is a preview of subscription content, access via your institution.

## References

Bapat RB, Raghavan TES (1997) Nonnegative matrices and applications. Cambridge University Press, Cambridge

Bingham NH (1973) Positive definite functions on spheres. Proc Camb Philos Soc 73:145–156

Calder CA (2007) Dynamic factor process convolution models for multivariate space–time data with application to air quality assessment. Environ Ecol Stat 14:229–247

Chen D, Menegatto A, Sun X (2003) A necessary and sufficient condition for strictly positive definite functions on spheres. Proc Am Math Soc 131:2733–2740

Chilés, JP, Delfiner P (1999) Geostatistics: modeling spatial uncertainty. Wiley, New York

Choi J, Reich B, Fuentes M, Davis J (2009) Multivariate spatial-temporal modeling and predication of speciated fine particles. J Statist Theory Pract 3:407–418

Christensen JPR, Ressel P (1978) Functions operating on positive definite matrices and a theorem of Schoenberg. Trans Am Math Soc 243:89–95

Cohen A, Johns RH (1969) Regression on a random field. J Am Stat Assoc 64:1172–1182

Cressie N (1993) Statistics for spatial data, revised edn. Wiley, New York

Dimitrakopoulos RD, Mustapha H, Gloaguen E (2010) High-order statistics of spatial random fields: exploring spatial cumulants for modeling complex non-Gaussian and non-linear phenomena. Math Geosci 42:65–99

Du J, Ma C (2011) Spherically invariant vector random fields in space and time. IEEE Trans Signal Process 59:5921–5929

Du J, Ma C (2012) Variogram matrix functions for vector random fields with second-order increments. Math Geosci 44:411–425

Feller W (1971) An introduction to probability theory and its applications, vol. II, 2nd edn. Wiley, New York

Gaspari G, Cohn SE (1999) Construction of correlations in two and three dimensions. Q J R Meteorol Soc 125:723–757

Gaspari G, Cohn E, Guo J, Pawson S (2006) Construction and application of covariance functions with variable length-fields. Q J R Meteorol Soc 132:1815–1838

Haas TC (1998) Multivariate spatial prediction in the presence of non-linear trend and covariance non-stationarity. Environmetrics 7:145–165

Hannan EJ (1966) Spectral analysis for geophysical data. Geophys J R Astron Soc 11:225–236

Hannan EJ (1969) Fourier methods and random processes. Bull Internat Statist Inst 42:475–496

Hannan EJ (1970) Multiple time series. Wiley, New York

Huang C, Yao Y, Cressie N, Hsing T (2009) Multivariate intrinsic random functions for cokriging. Math Geosci 41:887–904

Huang C, Zhang H, Robeson SM (2011) On the validity of commonly used covariance and variogram functions on the sphere. Math Geosci 43:721–733

Johns RH (1963a) Stochastic processes on a sphere. Ann Math Stat 34:213–218

Johns RH (1963b) Stochastic processes on a sphere as applied to meteorological 500-millibar forecasts. Proceedings of the symposium of time series analysis, vol 119. Wiley, New York

Jun M (2011) Non-stationary cross-covariance models for multivariate processes on a globe. Scand J Stat 38:726–747

Le ND, Zidek JV (2006) Statistical analysis of environmental space–time processes. Springer, New York

Ma C (2011a) Vector random fields with second-order moments or second-order increments. Stoch Anal Appl 29:197–215

Ma C (2011b) Covariance matrices for second-order vector random fields in space and time. IEEE Trans Signal Process 59:2160–2168

Ma C (2011c) Covariance matrix functions of vector

*χ*^{2}random fields in space and time. IEEE Trans Commun 59:2254–2561Mangulis V (1965) Handbook of series for scientists and engineers. Academic Press, New York

Mardia KV (1988) Multi-dimensional multivariate Gaussian Markov random fields with application to image processing. J Multivar Anal 24:265–284

Matérn B (1986) Spatial variation, 2nd edn. Springer, New York

Matheron G (1989) The internal consistency of models in geostatistics. In: Armstrong M (ed) Geostatistics, vol 1. Kluwer Academic, Dordrecht, pp 21–38

McLeod MG (1986) Stochastic processes on a sphere. Phys Earth Planet Inter 43:283–299

Roy R (1972) Spectral analysis for random field on the circle. J Appl Probab 9:745–757

Roy R (1973) Estimation of the covariance function of a homogeneous process on the sphere. Ann Stat 1:780–785

Roy R (1976) Spectral analysis for a random process on the sphere. Ann Inst Stat Math 28:91–97

Sain R, Cressie N (2007) A spatial model for multivariate lattice data. J Econom 140:226–259

Sain R, Furrer R, Cressie N (2011) A spatial analysis of multivariate output from regional climate models. Ann Appl Stat 5:150–175

Schoenberg I (1942) Positive definite functions on spheres. Duke Math J 9:96–108

Szegö G (1959) Orthogonal polynomials. Amer. Math. Soc. Colloq. Publ., vol 23. Amer. Math. Soc., Providence

Tebaldi C, Lobell DB (2008) Towards probabilistic projections of climate change impacts on global crop yields. Geophys Res Lett 35:L08705. doi:10.1029/2008GL033423

Trenberth KE, Shea DJ (2005) Relationships between precipitation and surface temperature. Geophys Res Lett 32:L14703. doi:10.1029/2005GL022760

Watson GN (1944) A treatise on the theory of Bessel functions, 2nd edn. Cambridge University Press, London

Weaver A, Courtier P (2001) Correlation modelling on the sphere using a generalized diffusion equation. Q J R Meteorol Soc 127:1815–1846

Widder DV (1946) The Laplace transform. Princeton Univ. Press, Princeton

Yadrenko AM (1983) Spectral theory of random fields, Optimization Software, New York.

Yaglom AM (1987) Correlation theory of stationary and related random functions, vol. I. Springer, New York

Zidek JV, Sun W, Le ND (2000) Designing and integrating composite networks for monitoring multivariate Gaussian pollution fields. J R Stat Soc, Ser C, Appl Stat 49:63–79

## Acknowledgements

This work is supported in part by U.S. Department of Energy under Grant DE-SC0005359, in part by the Kansas NSF EPSCoR under Grant EPS0903806, and in part by a Kansas Technology Enterprise Corporation grant. The author would like to thank an associate editor and two anonymous reviewers for their valuable comments and suggestions which helped to improve the presentation of this paper.

## Author information

### Affiliations

### Corresponding author

## Appendix: Proof of Theorem 2

### Appendix: Proof of Theorem 2

Suppose that **C**(*θ*) is the covariance matrix function of an *m*-variate stationary and isotropic random field \(\{ \mathbf {Z}(\mathbf{x}), \mathbf{x}\in \mathbb{S}^{\infty}\}\). Then *C*
_{
ij
}(*θ*)=*C*
_{
ji
}(*θ*),*θ*∈[0,*π*], *i*,*j*=1,…,*m*. According to Theorem 2 of Schoenberg (1942), for each *k*∈{1,…,*m*}, the direct covariance function *C*
_{
kk
}(*θ*) of a component random field \(\{ Z_{k} (\mathbf{x}), \mathbf{x}\in\mathbb{S}^{\infty}\}\) possesses an ultraspherical expansion

where {*b*
_{
n
}(*k*,*k*),*n*=0,1,…} is a summable sequence of nonnegative numbers. Similarly, for *i*≠*j*, a scalar random field \(\{ Z_{i} (\mathbf {x})+Z_{j} (\mathbf{x} ), \mathbf{x}\in\mathbb{S}^{\infty}\}\) has the covariance function *C*
_{
ii
}(*θ*)+*C*
_{
jj
}(*θ*)+2*C*
_{
ij
}(*θ*) that possesses an ultraspherical expansion

and a scalar random field \(\{ Z_{i} (\mathbf{x})-Z_{j} (\mathbf{x}), \mathbf {x}\in\mathbb{S}^{\infty}\}\) has the covariance function *C*
_{
ii
}(*θ*)+*C*
_{
jj
}(*θ*)−2*C*
_{
ij
}(*θ*) that possesses an ultraspherical expansion

Taking the difference of Eqs. (12) and (13), we obtain

where \(b_{n} (i, j) = \frac{1}{4} (e^{+}_{n} (i, j)-e^{-}_{n} (i, j)), n = 0, 1, \ldots, i, j =1, 2, \ldots, m\). It follows from Eqs. (11) and (14) that **C**(*θ*) adopts an expansion like Eq. (6). Therefore, it remains to show that **B**
_{
n
}=(*b*
_{
n
}(*i*,*j*))_{
m×m
} is positive definite, for each nonnegative integer *n*. To this end, suppose that *a*
_{1},…,*a*
_{
m
} are arbitrary real numbers. Using these constants, we formulate a scalar random field \(\{ \sum_{k=1}^{m} a_{k} Z_{k} (\mathbf{x}), \mathbf{x}\in\mathbb {S}^{\infty}\}\). This is a second-order random field on \(\mathbb{S}^{\infty}\) with covariance function

where the second equality follows from Eqs. (11) and (14). Applying Theorem 2 of Schoenberg (1942) to the last expansion, we obtain that all its coefficients **a**′**B**
_{
n
}
**a** are nonnegative, that is, each **B**
_{
n
}=(*b*
_{
n
}(*i*,*j*))_{
m×m
} is positive definite.

On the other hand, let **C**(*θ*) be an *m*×*m* matrix function with expansion of Eq. (6). For each *n*≥0, it is known (Schoenberg 1942) that cos^{n}
*θ* is positive definite on \(\mathbb{S}^{\infty}\), and thus **1**cos^{n}
*θ* is a covariance matrix function, where **1** is an *m*×*m* matrix with all entries equal 1. By Theorem 6 of Ma (2011b), the Hadamard product of **B**
_{
n
} and **1**cos^{n}
*θ*, which is the same as **B**
_{
n
}cos^{n}
*θ*, is also a covariance matrix function on \(\mathbb{S}^{\infty}\). So is **C**(*θ*), the sum of the sequence {**B**
_{
n
}cos^{n}
*θ*,*n*=0,1,2,…}.

## Rights and permissions

## About this article

### Cite this article

Ma, C. Stationary and Isotropic Vector Random Fields on Spheres.
*Math Geosci* **44, **765–778 (2012). https://doi.org/10.1007/s11004-012-9411-8

Received:

Accepted:

Published:

Issue Date:

### Keywords

- Absolutely monotone function
- Cross covariance
- Covariance matrix function
- Direct covariance
- Elliptically contoured random field
- Gaussian random field
- Gegenbauer’s polynomials
- Positive definite matrix