Why do we need and how should we implement Bayesian kriging methods

Original Paper


The spatial prediction methodology that has become known under the heading of kriging is largely based on the assumptions that the underlying random field is Gaussian and the covariance function is exactly known. In practical applications, however, these assumptions will not hold. Beyond Gaussianity of the random field, lognormal kriging, disjunctive kriging, (generalized linear) model-based kriging and trans-Gaussian kriging have been proposed in the literature. The latter approach makes use of the Box–Cox-transform of the data. Still, all the alternatives mentioned do not take into account the uncertainty with respect to the distribution (or transformation) and the estimated covariance function of the data. The Bayesian trans-Gaussian kriging methodology proposed in the present paper is in the spirit of the “Bayesian bootstrap” idea advocated by Rubin (Ann Stat 9:130–134, 1981) and avoids the unusual specification of noninformative priors often made in the literature and is entirely based on the sample distribution of the estimators of the covariance function and of the Box–Cox parameter. After some notes on Bayesian spatial prediction, noninformative priors and developing our new methodology finally we will present an example illustrating our pragmatic approach to Bayesian prediction by means of a simulated data set.


Kriging Covariance Function Transformation Parameter Predictive Distribution Covariance Parameter 



This work was partially funded by the European Commission, under the Sixth Framework Programme, by the Contract N. 033811 with DG INFSO, action Line IST-2005-2.5.12 ICT for Environmental Risk Management. The views expressed herein are those of the authors and are not necessarily those of the European Commission.


  1. Abrahamsen P (1992) Bayesian kriging for seismic depth conversion of a multilayer reservoir. In: Soares A (ed) Geostatistics Troia 92. Kluwer, Dordrecht, pp 385–398Google Scholar
  2. Abramowitz M, Stegun IA (1965) Handbook of mathematical functions. Dover, New YorkGoogle Scholar
  3. Banerjee S, Carlin BP, Gelfand AE (2004) Hierarchical modeling and analysis for spatial data. Chapman and Hall/CRC, Boca Raton, FloridaGoogle Scholar
  4. Berger JO, De Oliveira V, Sanso B (2001) Objective Bayesian analysis of spatially correlated data. J Am Stat Assoc 96(456):1361–1374CrossRefGoogle Scholar
  5. Box GEP, Cox DR (1964) An analysis of transformations with discussion. J Roy Stat Soc Ser B 26:211–252Google Scholar
  6. Brown PL, Le ND, Zidek JV (1994),Multivariate spatial interpolation and exposure to air pollutants. Can J Stat 22:489–509CrossRefGoogle Scholar
  7. Cressie NAC (1985) Fitting variogram models by weighted least squares. Math Geol 17(5)Google Scholar
  8. Christensen R (1991) Linear models for multivariate time series and spatial data. Springer, BerlinGoogle Scholar
  9. Christensen OF, Diggle PJ, Ribeiro PJ (2001) Analysing positive-valued spatial data: the transformed Gaussian Model. In: Monestiez P, Allard D, Froidevaux R (eds) GeoENV III—geostatistics for environmental applications. Kluwer, Dordrecht, pp 287–298Google Scholar
  10. Cui H, Stein A, Myers DE (1995) Extension of spatial information, Bayesian kriging and updating of prior variogram parameters. Environmetrics, 373–384Google Scholar
  11. De Oliveira V, Kedem B, Short DA (1997) Bayesian prediction of transformed Gaussian random fields. J Am Stat Assoc 92(440):1422–1433CrossRefGoogle Scholar
  12. Diggle JP, Ribeiro JR PJ (2002) Bayesian inference in Gaussian model-based geostatistics. Geogr Environ Model 6(2):129–146CrossRefGoogle Scholar
  13. Diggle PJ, Tawn JA, Moyeed RA (1998) Model-based geostatistics (with discussion). Appl Stat 47:299–350Google Scholar
  14. Ecker MD, Gelfand AE (1997) Bayesian variogram modeling for an isotropic spatial process. Technical Report 97–01, Department of Statistics, University of ConnecticutGoogle Scholar
  15. Gaudard M, Karson M, Linder E, Sinha D (1999) Bayesian spatial prediction. Environ Ecol Stat 6:147–171CrossRefGoogle Scholar
  16. Handcock MS, Stein ML (1993) A Bayesian analysis of kriging. Technometrics 35(4):403–410CrossRefGoogle Scholar
  17. Handcock MS, Wallis JR (1994) An approach to statistical spatio-temporal modeling of meteorological fields, with discussion. J Am Stat Assoc 89:368–378CrossRefGoogle Scholar
  18. Journel AJ, Huijbregts CJ (1978) Mining geostatistics. Academic, New YorkGoogle Scholar
  19. Kitanidis PK (1986) Parameter uncertainty in estimation of spatial functions: Bayesian analysis. Water Resour Res 22:499–507CrossRefGoogle Scholar
  20. Le ND, Zidek JV (1992) Interpolation with uncertain spatial covariance: a Bayesian alternative to kriging. J Multivariate Anal 43:351–374CrossRefGoogle Scholar
  21. Omre H (1987) Bayesian kriging-merging observations and qualified guess in kriging. Math Geol 19:25–39CrossRefGoogle Scholar
  22. Omre H, Halvorsen KB (1989) The Bayesian bridge between simple and universal kriging. Math Geol 21(7):767–786CrossRefGoogle Scholar
  23. Paulo R (2005) Default priors for Gaussian processes. Ann Stat 33:556–582CrossRefGoogle Scholar
  24. Pilz J, Schimek MG, Spöck G (1997) Taking account of uncertainty in spatial covariance estimation. In: Baafi E (ed) Geostatistics proc V int Geost Congr. Kluwer, DordrechtGoogle Scholar
  25. Rivoirard J (1994) Introduction to disjunctive kriging and non-linear geostatistics. Clarendon Press, OxfordGoogle Scholar
  26. Rubin DB (1981) The Bayesian bootstrap. Ann Stat 9:130–134CrossRefGoogle Scholar
  27. Stein ML (1999) Interpolation of spatial data: some Theory of Kriging. Springer, New YorkGoogle Scholar
  28. Spöck G (1997) Die geostatistische Berücksichtigung von a-priori Kenntnissen über die Trendfunktion und die Kovarianzfunktion aus Bayesscher, Minimax und Spektraler Sicht, diploma thesis. University of KlagenfurtGoogle Scholar

Copyright information

© Springer-Verlag 2007

Authors and Affiliations

  1. 1.Department of StatisticsUniversity of KlagenfurtKlagenfurtAustria

Personalised recommendations