Spatiotemporal modelling of PM2.5 concentrations in Lombardy (Italy): a comparative study

This study presents a comparative analysis of three predictive models with an increasing degree of flexibility: hidden dynamic geostatistical models (HDGM), generalised additive mixed models (GAMM), and the random forest spatiotemporal kriging models (RFSTK). These models are evaluated for their eff...

Full description

Saved in:
Bibliographic Details
Published in:Environmental and ecological statistics 2024-06, Vol.31 (2), p.245-272
Main Authors: Otto, Philipp, Fusta Moro, Alessandro, Rodeschini, Jacopo, Shaboviq, Qendrim, Ignaccolo, Rosaria, Golini, Natalia, Cameletti, Michela, Maranzano, Paolo, Finazzi, Francesco, Fassò, Alessandro
Format: Article
Language:English
Subjects:
Air pollution
Algorithms
Biomedical and Life Sciences
Chemistry and Earth Sciences
Comparative analysis
Comparative studies
Computer Science
Datasets
Ecology
Emissions
Health Sciences
Life Sciences
Machine learning
Math. Appl. in Environmental Science
Medicine
Modelling
Outdoor air quality
Partial differential equations
Particulate matter
Physics
Pollutants
Pollution data
Prediction models
Spatiotemporal data
Statistics for Engineering
Statistics for Life Sciences
Theoretical Ecology/Statistics
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study presents a comparative analysis of three predictive models with an increasing degree of flexibility: hidden dynamic geostatistical models (HDGM), generalised additive mixed models (GAMM), and the random forest spatiotemporal kriging models (RFSTK). These models are evaluated for their effectiveness in predicting PM 2.5 concentrations in Lombardy (North Italy) from 2016 to 2020. Despite differing methodologies, all models demonstrate proficient capture of spatiotemporal patterns within air pollution data with similar out-of-sample performance. Furthermore, the study delves into station-specific analyses, revealing variable model performance contingent on localised conditions. Model interpretation, facilitated by parametric coefficient analysis and partial dependence plots, unveils consistent associations between predictor variables and PM 2.5 concentrations. Despite nuanced variations in modelling spatiotemporal correlations, all models effectively accounted for the underlying dependence. In summary, this study underscores the efficacy of conventional techniques in modelling correlated spatiotemporal data, concurrently highlighting the complementary potential of Machine Learning and classical statistical approaches.
ISSN:1352-8505
1573-3009
DOI:10.1007/s10651-023-00589-0