Meteorology-driven variability of air pollution (PM<sub>1</sub>) revealed with explainable machine learning

Abstract

Abstract. Air pollution, in particular high concentrations of particulate matter smaller than 1 µm in diameter (PM1), continues to be a major health problem, and meteorology is known to substantially influence atmospheric PM concentrations. However, the scientific understanding of the ways in which complex interactions of meteorological factors lead to high-pollution episodes is inconclusive. In this study, a novel, data-driven approach based on empirical relationships is used to characterize and better understand the meteorology-driven component of PM1 variability. A tree-based machine learning model is set up to reproduce concentrations of speciated PM1 at a suburban site southwest of Paris, France, using meteorological variables as input features. The model is able to capture the majority of occurring variance of mean afternoon total PM1 concentrations (coefficient of determination (R2) of 0.58), with model performance depending on the individual PM1 species predicted. Based on the models, an isolation and quantification of individual, season-specific meteorological influences for process understanding at the measurement site is achieved using SHapley Additive exPlanation (SHAP) regression values. Model results suggest that winter pollution episodes are often driven by a combination of shallow mixed layer heights (MLHs), low temperatures, low wind speeds, or inflow from northeastern wind directions. Contributions of MLHs to the winter pollution episodes are quantified to be on average ∼5 µg/m3 for MLHs below <500 m a.g.l. Temperatures below freezing initiate formation processes and increase local emissions related to residential heating, amounting to a contribution to predicted PM1 concentrations of as much as ∼9 µg/m3. Northeasterly winds are found to contribute ∼5 µg/m3 to predicted PM1 concentrations (combined effects of u- and v-wind components), by advecting particles from source regions, e.g. central Europe or the Paris region. Meteorological drivers of unusually high PM1 concentrations in summer are temperatures above ∼25 ∘C (contributions of up to ∼2.5 µg/m3), dry spells of several days (maximum contributions of ∼1.5 µg/m3), and wind speeds below ∼2 m/s (maximum contributions of ∼3 µg/m3), which cause a lack of dispersion. High-resolution case studies are conducted showing a large variability of processes that can lead to high-pollution episodes. The identification of these meteorological conditions that increase air pollution could help policy makers to adapt policy measures, issue warnings to the public, or assess the effectiveness of air pollution measures.

Factors Underlying Spatiotemporal Variations in Atmospheric PM2.5 Concentrations in Zhejiang Province, China

“…A positive SHAP value means that the prediction (PM 2.5 ) based on the corresponding influencing factor is above the mean value (the average PM 2.5 ). The relative importance of each variable is represented by their mean absolute SHAP values [52]. Advantages of the SHAP algorithm include: (1) global interpretabilitythe collective SHAP value can identify positive or negative relationships for each variable, and the global importance of different features can be calculated by computing their respective absolute SHAP values; (2) local interpretability-each feature acquires its own corresponding spatial SHAP value in different locations; this resolves the limitation of traditional methods of evaluating relative importance whereby results are obtained across the entire region or population but not for each pixel and individual [53].…”

Effect of rainfall-induced diabatic heating over southern China on the formation of wintertime haze on the North China Plain

“…15). This implies that emissions are the source of haze over the NCP, and meteorological factors worsen haze (e.g., Wang et al, 2015;Stirnberg et al, 2021).…”

“…These haze events are caused by emissions of pollutants combined with unfavorable meteorological conditions (Yang et al, 2016;Cai et al, 2017;Zhang et al, 2021). Although emissions play an important role in the generation of haze, numerous studies have suggested that meteorology is also a significant factor in the occurrence of extreme haze events (Quan et al 2011;Wang et al 2015;Gao et al 2016;Stirnberg et al, 2021). For instance, Dang and Liao (2019) found that large interannual variations in the frequency and intensity of severe winter haze days were driven mainly by changes in meteorology.…”

Using an Explainable Machine Learning Approach to Characterize Earth System Model Errors: Application of SHAP Analysis to Modeling Lightning Flash Occurrence

“…Toms et al (2020) further explored how two different methods, layerwise relevance propagation and backward optimization, can be used to glean scientifically relevant information from neural network predictions of variability in the Earth System. Stirnberg et al (2021) use an alternative method, SHapley Additive exPlanations (SHAP) applied to boosted regression trees, to quantify the importance of various meteorological drivers on particulate matter concentrations.…”