Cross-Covariance Functions for Multivariate Geostatistics

Abstract

Continuously indexed datasets with multiple variables have become ubiquitous in the geophysical, ecological, environmental and climate sciences, and pose substantial analysis challenges to scientists and statisticians. For many years, scientists developed models that aimed at capturing the spatial behavior for an individual process; only within the last few decades has it become commonplace to model multiple processes jointly. The key difficulty is in specifying the cross-covariance function, that is, the function responsible for the relationship between distinct variables. Indeed, these cross-covariance functions must be chosen to be consistent with marginal covariance functions in such a way that the second-order structure always yields a nonnegative definite covariance matrix. We review the main approaches to building cross-covariance models, including the linear model of coregionalization, convolution methods, the multivariate Matérn and nonstationary and space-time extensions of these among others. We additionally cover specialized constructions, including those designed for asymmetry, compact support and spherical domains, with a review of physics-constrained models. We illustrate select models on a bivariate regional climate model output example for temperature and pressure, along with a bivariate minimum and maximum temperature observational dataset; we compare models by likelihood value as well as via cross-validation co-kriging studies. The article closes with a discussion of unsolved problems.

Published at http://dx.doi.org/10.1214/14-STS487 in the Statistical Science (http://www.imstat.org/sts/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (2)

• Some covariance models based on normal scale mixtures
• Covariance approximation for large multivariate spatial data sets with an application to multiple climate model errors

Cited by in corpus (50)

• Learning Constitutive Relations from Indirect Observations Using Deep Neural Networks
• The rational SPDE approach for Gaussian random fields with general smoothness
• Highly Scalable Bayesian Geostatistical Modeling via Meshed Gaussian Processes on Partitioned Domains
• Recent Developments in Complex and Spatially Correlated Functional Data
• Spectral Mixture Kernels for Multi-Output Gaussian Processes
• Multivariate type G Matérn stochastic partial differential equation random fields
• Spatial Blind Source Separation
• Spatial Factor Modeling: A Bayesian Matrix-Normal Approach for Misaligned Data
• When Doesn't Cokriging Outperform Kriging?
• Spatio-temporal bivariate statistical models for atmospheric trace-gas inversion
• Computer model calibration with large non-stationary spatial outputs: application to the calibration of a climate model
• High-dimensional Multivariate Geostatistics: A Bayesian Matrix-Normal Approach
• Relative Positional Encoding for Transformers with Linear Complexity
• Matérn Class Tensor-Valued Random Fields and Beyond
• Bayesian Regression and Classification Using Gaussian Process Priors Indexed by Probability Density Functions
• Blind source separation for non-stationary random fields
• Modeling massive highly-multivariate nonstationary spatial data with the basis graphical lasso
• Spatio-Temporal Cross-Covariance Functions under the Lagrangian Framework with Multiple Advections
• Optimal designs for some bivariate cokriging models
• Asymmetric Matrix-Valued Covariances for Multivariate Random Fields on Spheres
• Semiparametric Estimation of Cross-covariance Functions for Multivariate Random Fields
• Test of the Latent Dimension of a Spatial Blind Source Separation Model
• Spatial Multivariate Trees for Big Data Bayesian Regression
• Nonlinear blind source separation exploiting spatial nonstationarity
• Random Forests for dependent data
• Estimation of Spatial Deformation for Nonstationary Processes via Variogram Alignment
• Bayesian nonparametric generative modeling of large multivariate non-Gaussian spatial fields
• Modeling Multivariate Spatial-Temporal Data with Latent Low-Dimensional Dynamics
• Weakly-supervised Multi-output Regression via Correlated Gaussian Processes
• Finite Population Survey Sampling: An Unapologetic Bayesian Perspective
• Physics-Informed Gaussian Process Regression for Probabilistic States Estimation and Forecasting in Power Grids
• Multi-group Gaussian Processes
• The promise and perils of point process models of political events
• Information Flow Rate for Cross-Correlated Stochastic Processes
• On Cokriging, Neural Networks, and Spatial Blind Source Separation for Multivariate Spatial Prediction
• Gridding and Parameter Expansion for Scalable Latent Gaussian Models of Spatial Multivariate Data
• Multivariate Gaussian Process Regression for Multiscale Data Assimilation and Uncertainty Reduction
• Large Linear Multi-output Gaussian Process Learning
• Extremes of locally stationary Gaussian and chi fields on manifolds
• Approximate Marginal Posterior for Log Gaussian Cox Processes
• A Copula Model for Non-Gaussian Multivariate Spatial Data
• Stochastic simulation of predictive space-time scenarios of wind speed using observations and physical models
• Bayesian covariance modeling of multivariate spatial random fields
• Inside-out cross-covariance for spatial multivariate data
• A memory-free spatial additive mixed modeling for big spatial data
• High Performance Multivariate Geospatial Statistics on Manycore Systems
• Flexible Validity Conditions for the Multivariate Matérn Covariance in any Spatial Dimension and for any Number of Components
• A spatial dependence graph model for multivariate spatial hybrid processes
• Hierarchical Bayesian Nearest Neighbor Co-Kriging Gaussian Process Models; An Application to Intersatellite Calibration
• Joint Asymptotics for Estimating the Fractal Indices of Bivariate Gaussian Processes