A nonstationary spatial model of PM2.5 with localized transfer learning from numerical model output

TL;DR

This paper introduces a scalable Bayesian nonstationary spatial model for PM2.5 that leverages localized transfer learning from numerical model outputs to improve air pollution inference and prediction.

Contribution

It develops a novel nonstationary covariance modeling approach using localized transfer learning from numerical models within a scalable Bayesian framework.

Findings

Enhanced spatial prediction accuracy for PM2.5.

Effective integration of numerical model outputs with sparse observational data.

Scalable Bayesian implementation for large spatial datasets.

Abstract

Ambient air pollution measurements from regulatory monitoring networks are routinely used to support epidemiologic studies and environmental policy decision making. However, regulatory monitors are spatially sparse and preferentially located in areas with large populations. Numerical air pollution model output can be leveraged into the inference and prediction of air pollution data combining with measurements from monitors. Nonstationary covariance functions allow the model to adapt to spatial surfaces whose variability changes with location like air pollution data. In the paper, we employ localized covariance parameters learned from the numerical output model to knit together into a global nonstationary covariance, to incorporate in a fully Bayesian model. We model the nonstationary structure in a computationally efficient way to make the Bayesian model scalable.