2020/04/27 by Mark D. Risser, Risser, Mark
Environmental Science · #Air Quality and Health Impacts #Applications (stat.AP) #Atmospheric and Environmental Gas Dynamics #Climate Change and Health Impacts #Climate variability and models #FOS: Computer and information sciences #Methodology (stat.ME)
paper · pdf · doi:10.48550/arxiv.2005.03658
openalex publication_date 2020/04/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Heat waves resulting from prolonged extreme temperatures pose a significant\nrisk to human health globally. Given the limitations of observations of extreme\ntemperature, climate models are often used to characterize extreme temperature\nglobally, from which one can derive quantities like return values to summarize\nthe magnitude of a low probability event for an arbitrary geographic location.\nHowever, while these derived quantities are useful on their own, it is also\noften important to apply a spatial statistical model to such data in order to,≠.g., understand how the spatial dependence properties of the return values\nvary over space and emulate the climate model for generating additional spatial\nfields with corresponding statistical properties. For these objectives, when\nmodeling global data it is critical to use a nonstationary covariance function.\nFurthermore, given that the output of modern global climate models can be on\nthe order of \O(104), it is important to utilize approximate\nGaussian process methods to enable inference. In this paper, we demonstrate the\napplication of methodology introduced in Risser and Turek (2020) to conduct a\nnonstationary and fully Bayesian analysis of a large data set of 20-year return\nvalues derived from an ensemble of global climate model runs with over 50,000\nspatial locations. This analysis uses the freely available BayesNSGP software\npackage for R.\n