1. Bayesian kriging Instead of estimating the parameters, we put a prior distribution on them, and update the distribution using the data. Model: Prior: Posterior:

2. geoR Default covariance model exponential Prior is assigned to  and. The latter assumed zero unless specified. The distributions can be discretized. Default prior on  is flat (if not specified, assumed constant). (Lots of different assignments are possible)

3. Prior/posterior of 

4. Variogram estimates mean median mode

5. Predictive mean

6. Kriging variances BayesClassical range (52,307) range (170,278)

7. 2. Covariances NRCSE

8. Valid covariance functions Bochner’s theorem: The class of covariance functions is the class of positive definite functions C: Why?

9. Spectral representation By the spectral representation any isotropic continuous correlation on R d is of the form By isotropy, the expectation depends only on the distribution G of. Let Y be uniform on the unit sphere. Then

10. Isotropic correlation J v (u) is a Bessel function of the first kind and order v. Hence and in the case d=2 (Hankel transform)

11. The Bessel function J 0 –0.403

12. The exponential correlation A commonly used correlation function is  (v) = e –v/ . Corresponds to a Gaussian process with continuous but not differentiable sample paths. More generally,  (v) = c(v=0) + (1-c)e –v/  has a nugget c, corresponding to measurement error and spatial correlation at small distances. All isotropic correlations are a mixture of a nugget and a continuous isotropic correlation.

13. The squared exponential Usingyields corresponding to an underlying Gaussian field with analytic paths. This is sometimes called the Gaussian covariance, for no really good reason. A generalization is the power(ed) exponential correlation function,

14. The spherical Corresponding variogram nugget sillrange

16. The Matérn class where is a modified Bessel function of the third kind and order . It corresponds to a spatial field with  –1 continuous derivatives  =1/2 is exponential;  =1 is Whittle’s spatial correlation; yields squared exponential.

18. Some other covariance/variogram families NameCovarianceVariogram Wave Rational quadratic LinearNone Power lawNone

19. Recall Method of moments: square of all pairwise differences, smoothed over lag bins Problems: Not necessarily a valid variogram Not very robust Estimation of variograms

20. A robust empirical variogram estimator (Z(x)-Z(y)) 2 is chi-squared for Gaussian data Fourth root is variance stabilizing Cressie and Hawkins:

21. Least squares Minimize Alternatives: fourth root transformation weighting by 1/  2 generalized least squares

22. Maximum likelihood Z~N n ( ,  )  =  [  (s i -s j ;  )] =  V(  ) Maximize and  maximizes the profile likelihood

23. A peculiar ml fit

24. Some more fits

25. All together now...

26. Asymptotics Increasing domain asymptotics: let region of interest grow. Station density stays the same Bad estimation at short distances, but effectively independent blocks far apart Infill asymptotics: let station density grow, keeping region fixed. Good estimates at short distances. No effectively independent blocks, so technically trickier

27. Stein’s result Covariance functions C 0 and C 1 are compatible if their Gaussian measures are mutually absolutely continuous. Sample at {s i, i=1,...,n}, predict at s (limit point of sampling points). Let e i (n) be kriging prediction error at s for C i, and V 0 the variance under C 0 of some random variable. If lim n V 0 (e 0 (n))=0, then

28. Global processes Problems such as global warming require modeling of processes that take place on the globe (an oriented sphere). Optimal prediction of quantities such as global mean temperature need models for global covariances. Note: spherical covariances can take values in [-1,1]–not just imbedded in R 3. Also, stationarity and isotropy are identical concepts on the sphere.

29. Isotropic covariances on the sphere Isotropic covariances on a sphere are of the form where p and q are directions,  pq the angle between them, and P i the Legendre polynomials. Example: a i =(2i+1)  i

30. Global temperature Global Historical Climatology Network 7280 stations with at least 10 years of data. Subset with 839 stations with data 1950-1991 selected.

31. Isotropic correlations