1. Home
  2. Issues
  3. Volume 20, Issue 4 (2022): Special Issue: Large-Scale Spatial Data Science
  4. Geostatistics for Large Datasets on Riem ...

Journal of Data Science


Geostatistics for Large Datasets on Riemannian Manifolds: A Matrix-Free Approach
Volume 20, Issue 4 (2022): Special Issue: Large-Scale Spatial Data Science, pp. 512–532
Pub. online: 3 November 2022      Type: Statistical Data Science      Open accessOpen Access
Received
2 August 2022
Accepted
17 October 2022
Published
3 November 2022

Abstract

Large or very large spatial (and spatio-temporal) datasets have become common place in many environmental and climate studies. These data are often collected in non-Euclidean spaces (such as the planet Earth) and they often present nonstationary anisotropies. This paper proposes a generic approach to model Gaussian Random Fields (GRFs) on compact Riemannian manifolds that bridges the gap between existing works on nonstationary GRFs and random fields on manifolds. This approach can be applied to any smooth compact manifolds, and in particular to any compact surface. By defining a Riemannian metric that accounts for the preferential directions of correlation, our approach yields an interpretation of the nonstationary geometric anisotropies as resulting from local deformations of the domain. We provide scalable algorithms for the estimation of the parameters and for optimal prediction by kriging and simulation able to tackle very large grids. Stationary and nonstationary illustrations are provided.

Supplementary material

 Supplementary Material
The code used to perform the maximum likelihood estimation in Section 5.2 is available at https://github.com/mike-pereira/matrix-free-mle.

References

 
Abdulah S, Ltaief H, Sun Y, Genton MG, Keyes DE (2018). ExaGeoStat: A high performance unified software for geostatistics on manycore systems. IEEE Transactions on Parallel and Distributed Systems, 29(12): 2771–2784.
 
Borovitskiy V, Azangulov I, Terenin A, Mostowsky P, Deisenroth M, Durrande N (2021). Matérn Gaussian processes on graphs. In: International Conference on Artificial Intelligence and Statistics, 2593–2601. PMLR.
 
Borovitskiy V, Terenin A, Mostowsky P, Deisenroth M (2020). Matérn Gaussian processes on Riemannian manifolds. Advances in Neural Information Processing Systems, 33: 12426–12437.
 
Carrizo Vergara R, Allard D, Desassis N (2022). A general framework for SPDE-based stationary random fields. Bernoulli, 28(1): 1–32.
 
Chilès JP, Delfiner P (2012). Geostatistics: Modeling Spatial Uncertainty. 2nd Edition. Wiley Series In Probability and Statistics.
 
Emery X, Porcu E (2019). Simulating isotropic vector-valued Gaussian random fields on the sphere through finite harmonics approximations. Stochastic Environmental Research and Risk Assessment, 33(8): 1659–1667.
 
Fouedjio F, Desassis N, Rivoirard J (2016). A generalized convolution model and estimation for non-stationary random functions. Spatial Statistics, 16: 35–52.
 
Fouedjio F, Desassis N, Romary T (2015). Estimation of space deformation model for non-stationary random functions. Spatial Statistics, 13: 45–61.
 
Fuglstad GA, Lindgren F, Simpson D, Rue H (2015a). Exploring a new class of non-stationary spatial Gaussian random fields with varying local anisotropy. Statistica Sinica, 25: 115–133.
 
Fuglstad GA, Simpson D, Lindgren F, Rue H (2015b). Does non-stationary spatial data always require non-stationary random fields? Spatial Statistics, 14: 505–531.
 
Gershgorin S (1931). Über die Abgrenzung der Eigenwerte einer matrix. Izv. Akad. Nauk. USSR Otd. Fiz.-Mat. Nauk, 7: 749–754.
 
Gneiting T (2013). Strictly and non-strictly positive definite functions on spheres. Bernoulli, 19(4): 1327–1349.
 
Han I, Malioutov D, Shin J (2015). Large-scale log-determinant computation through stochastic Chebyshev expansions. In: International Conference on Machine Learning, 908–917. PMLR.
 
Heaton MJ, Datta A, Finley AO, Furrer R, Guinness J, Guhaniyogi R, et al. (2019). A case study competition among methods for analyzing large spatial data. Journal of Agricultural, Biological and Environmental Statistics, 24(3): 398–425.
 
Higdon D, Swall J, Kern J (1999). Non-stationary spatial modeling. Bayesian Statistics, 6(1): 761–768.
 
Huang C, Zhang H, Robeson SM (2011). On the validity of commonly used covariance and variogram functions on the sphere. Mathematical Geosciences, 43(6): 721–733.
 
Huang H, Abdulah S, Sun Y, Ltaief H, Keyes DE, Genton MG (2021). Competition on spatial statistics for large datasets. Journal of Agricultural, Biological and Environmental Statistics, 26(4): 580–595.
 
Hutchinson MF (1989). A stochastic estimator of the trace of the influence matrix for Laplacian smoothing splines. Communications in Statistics-Simulation and Computation, 18(3): 1059–1076.
 
Jeong J, Jun M, Genton MG (2017). Spherical process models for global spatial statistics. Statistical Science, 32(4): 501–513.
 
Jost J (2008). Riemannian Geometry and Geometric Analysis. Springer.
 
Lang A, Pereira M (2021). Galerkin–Chebyshev approximation of Gaussian random fields on compact riemannian manifolds. arXiv preprint: https://arxiv.org/abs/2107.02667.
 
Lang A, Schwab C (2015). Isotropic Gaussian random fields on the sphere: Regularity, fast simulation and stochastic partial differential equations. The Annals of Applied Probability, 25(6): 3047–3094.
 
Lantuéjoul C, Freulon X, Renard D (2019). Spectral simulation of isotropic Gaussian random fields on a sphere. Mathematical Geosciences, 51(8): 999–1020.
 
Lee JM (2013). Introduction to Smooth Manifolds. Springer.
 
Lindgren F, Rue H, Lindström J (2011). An explicit link between Gaussian fields and Gaussian Markov random fields: The stochastic partial differential equation approach. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 73(4): 423–498.
 
Marinucci D, Peccati G (2011). Random Fields on the Sphere: Representation, Limit Theorems and Cosmological Applications. London Mathematical Society Lecture Note Series. Cambridge University Press.
 
Nocedal J, Wright S (2006). Numerical Optimization. Springer Science & Business Media.
 
Paciorek CJ, Schervish MJ (2006). Spatial modelling using a new class of nonstationary covariance functions. Environmetrics, 17(5): 483–506.
 
Pereira M (2019). Generalized random fields defined on Riemannian manfolds: Theory and practice, Ph.D. thesis, MINES ParisTech, PSL University.
 
Pereira M, Desassis N (2019). Efficient simulation of Gaussian Markov random fields by Chebyshev polynomial approximation. Spatial Statistics, 31: 100359.
 
Pereira M, Desassis N, Magneron C, Palmer N (2020). A matrix-free approach to geostatistical filtering. arXiv preprint: https://arxiv.org/abs/2004.02799.
 
Perrin O, Senoussi R (2000). Reducing non-stationary random fields to stationarity and isotropy using a space deformation. Statistics & Probability Letters, 48(1): 23–32.
 
Porcu E, Furrer R, Nychka D (2021). 30 years of space–time covariance functions. Wiley Interdisciplinary Reviews: Computational Statistics, 13(2): e1512.
 
Porcu E, Mateu J, Christakos G (2009). Quasi-arithmetic means of covariance functions with potential applications to space–time data. Journal of Multivariate Analysis, 100(8): 1830–1844.
 
Powell MJ (1994). A direct search optimization method that models the objective and constraint functions by linear interpolation. In: Advances in Optimization and Numerical Analysis, 51–67. Springer.
 
Rayner NA, Auchmann R, Bessembinder J, Brönnimann S, Brugnara Y, Capponi F, et al. (2020). The EUSTACE project: Delivering global, daily information on surface air temperature. Bulletin of the American Meteorological Society, 101(11): E1924–E1947.
 
Rue H, Held L (2005). Gaussian Markov Random Fields: Theory and Applications. Chapman and Hall/CRC.
 
Saad Y (2003). Iterative Methods for Sparse Linear Systems. SIAM.
 
Sampson PD, Guttorp P (1992). Nonparametric estimation of nonstationary spatial covariance structure. Journal of the American Statistical Association, 87(417): 108–119.
 
Simon D (2013). Evolutionary Optimization Algorithms. John Wiley & Sons.
 
Solin A, Särkkä S (2020). Hilbert space methods for reduced-rank Gaussian process regression. Statistics and Computing, 30(2): 419–446.
 
Tong YL (2012). The Multivariate Normal Distribution. Springer Science & Business Media.
 
Williams CK, Rasmussen CE (2006). Gaussian Processes for Machine Learning. MIT press, Cambridge, MA.

Related articles PDF XML
Related articles PDF XML

Copyright
2022 The Author(s). Published by the School of Statistics and the Center for Applied Statistics, Renmin University of China.
Open access article under the CC BY license.

Keywords
anisotropy finite elements Gaussian process Laplace-Beltrami operator nonstationarity

Funding
The authors acknowledge the support of the Mines Paris / INRAE chair “Geolearning”.

Metrics
since February 2021
927

Article info
views

438

PDF
downloads


Share


RSS