Fast simulation of Gaussian random fields with flexible correlation models in Euclidean spaces

TL;DR

This paper enhances the spectral turning-bands method for simulating Gaussian random fields, extending it to new correlation models with exact, stable, and efficient algorithms suitable for large-scale spatial data analysis.

Contribution

It introduces extensions of the spectral turning-bands method to Bummer-Tricomi and Gauss-Hypergeometric correlation models with exact stochastic representations and efficient simulation algorithms.

Findings

Algorithms are numerically stable and accurate across various parameters.

Simulation complexity is linear in the number of spectral components.

Applications include bootstrap methods for climate data analysis.

Abstract

The efficient simulation of Gaussian random fields with flexible correlation structures is fundamental in spatial statistics, machine learning, and uncertainty quantification. In this work, we revisit the \emph{spectral turning-bands} (STB) method as a versatile and scalable framework for simulating isotropic Gaussian random fields with a broad range of covariance models. Beyond the classical Mat\'ern family, we show that the STB approach can be extended to two recent and flexible correlation classes that generalize the Mat\'ern model: the Bummer-Tricomi model, which allows for polynomially decaying correlations and long-range dependence, and the Gauss-Hypergeometric model, which admits compactly supported correlations, including the Generalized Wendland family as a special case. We derive exact stochastic representations for both families: a Beta-prime mixture formulation for the Kummer-Tricomi model and complementary Beta- and Gasper-mixture representations for the Gauss-Hypergeometric model. These formulations enable exact, numerically stable, and computationally efficient simulation with linear complexity in the number of spectral components. Numerical experiments confirm the accuracy and computational stability of the proposed algorithms across a wide range of parameter configurations, demonstrating their practical viability for large-scale spatial modeling. As an application, we use the proposed STB simulators to perform parametric bootstrap for standard error estimation and model selection under weighted pairwise composite likelihood in the analysis of a large climate dataset.