Bayesian nonparametric clustering for spatio-temporal data, with an application to air pollution

TL;DR

This paper reviews Bayesian nonparametric clustering methods tailored for spatio-temporal environmental data, demonstrating their effectiveness on air pollution data to identify meaningful pollution patterns and inform policy.

Contribution

It provides a comprehensive review of Bayesian nonparametric clustering approaches for spatio-temporal data, with a focus on spatial product partition models applied to air quality monitoring.

Findings

Bayesian clustering effectively identifies pollution patterns

Spatial models improve clustering accuracy

Application to Italian PM10 data reveals key pollution zones

Abstract

Air pollution is a major global health hazard, with fine particulate matter (PM10) linked to severe respiratory and cardiovascular diseases. Hence, analyzing and clustering spatio-temporal air quality data is crucial for understanding pollution dynamics and guiding policy interventions. This work provides a review of Bayesian nonparametric clustering methods, with a particular focus on their application to spatio-temporal data, which are ubiquitous in environmental sciences. We first introduce key modeling approaches for point-referenced spatio-temporal data, highlighting their flexibility in capturing complex spatial and temporal dependencies. We then review recent advancements in Bayesian clustering, focusing on spatial product partition models, which incorporate spatial structure into the clustering process. We illustrate the proposed methods on PM10 monitoring data from Northern Italy, demonstrating their ability to identify meaningful pollution patterns. This review highlights the potential of Bayesian nonparametric methods for environmental risk assessment and offers insights into future research directions in spatio-temporal clustering for public health and environmental science.