PM10 Spatiotemporal Patterns in Portugal: Functional Data Analysis from 2017 to 2018

Air pollution significantly and severely affects human health, environment, materials, and economy, emerging as a key microclimate and air quality regulation issue. Hence, the spatial and temporal characterization of air pollutants and their relationship with meteorological constraining factors is critical, particularly from a climate change perspective.

Within this context, we present an exploratory statistical assessment combining functional data analysis (FDA) with unsupervised learning algorithms and spatial statistics to extract meaningful information about the main spatiotemporal patterns underlying air pollutant exceedances in mainland Portugal. Air pollutants’ spatial and temporal characterization over Portugal was performed, focusing particularly on the emissions of Particulate Matter (PM) during the major wildfire events in 2017-2018 and based on the Copernicus Atmosphere Monitoring (CAMS) data. Firstly, the temporal evolution of PM concentrations on each CAMS grid node was described as a function of time and outline the main temporal patterns of variability using a functional principal component analysis. Afterwards, CAMS grid nodes are classified according to their spatiotemporal similarities through hierarchical clustering adapted to spatially correlated functional data. Preliminary results show the main spatial patterns of AQ variability and indicate the regions presenting higher PM levels, especially during wildfire events. The present approach shows the potential of existing exploratory tools for spatiotemporal analysis of PM10 data, over regions less covered by the national air quality monitoring network.

Acknowledgements: This work is supported by the Portuguese Fundação para a Ciência e Tecnologia, FCT, I.P./MCTES through national funds (PIDDAC): UID/50019/2025 and LA/P/0068/2020 https://doi.org/10.54499/LA/P/0068/2020); and also on behalf of DHEFEUS -2022.09185.PTDC and the project FAIR- 2022.01660.PTDC).