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 Mat\'{e}rn 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.Comment: 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

The European mountain cryosphere: A review of past, current and future issues

“…On this basis, other applications of non-separable spatiotemporal covariance models (Gneiting et al, 2007;Genton and Kleiber, 2015), have great potential for further research in the alpine cryosphere. …”

Matérn Class Tensor-Valued Random Fields and Beyond

“…The paper by [11] reviews the main approaches to building multivariate correlation and covariance structures, including the multivariate Matérn models.…”

Multivariate localization methods for ensemble Kalman filtering

“…, N . Here, B is the beta function (Porcu et al, 2013;Genton and Kleiber, 2015). To illustrate the differences between the shape of the Gaspari-Cohn and the Askey functions, we show the Gaspari-Cohn function for c = 25 and the univariate Askey function for c = 50 and ν = 1, .…”

“…Kleiber and Porcu (2015) constructed nonstationary correlation functions with compact support for multivariate random fields. Genton and Kleiber (2015) reviewed approaches to building models for covariances between two different variables such as compactly supported correlation functions for multivariate Gaussian random fields.…”