Abstract
Aerosols play an important role in extreme air pollution events and climate change. Despite decades of effort, air pollution models often struggle to simulate heavy pollution events accurately due to complex aerosol chemical compositions and spatial-temporal variations. Recently, high-resolution aerosol mass spectrometer (AMS) measurements in Dublin revealed that organic aerosols (OA) dominate composition (>50%) during heavy pollution episodes, with PM1 concentrations exceeding 200 μg m⁻³ on November 19, 2016, and January 22, 2017. This study utilizes the WRF-Chem model to simulate these episodes. We find that standard emission datasets (EDGAR, EMEP, etc.) fail to replicate these events. However, by enhancing local emissions with diurnal variation and temperature dependence, the model successfully rebuilds the pollution peaks. These episodes are primarily attributed to local peat burning, highlighting significant underestimations of organic carbon (OC) in current inventories.
Introduction
Predicting "heavy pollution events"—short-term, extreme spikes in particulate matter—remains a significant challenge for atmospheric science. While regional models effectively track average air quality, they frequently miss dangerous peaks that impact public health. In Dublin, Ireland, these events are characterized by a high proportion of submicron organic aerosols (OA). Standard global and regional emission inventories often fail to capture the magnitude of these events, suggesting a significant gap in our understanding of local emission sources and their temporal dynamics.
Methodology
To address the discrepancies between observed and modeled pollution levels, this study employed the WRF-Chem v3.9 (Weather Research and Forecasting model coupled with Chemistry) framework. The chemical evolution was simulated using the CRIMECH v2 gas-phase mechanism and the MADE/SORGAM aerosol scheme.
A critical component of the methodology involved a sensitivity analysis of Planetary Boundary Layer (PBL) schemes. The YSU scheme was identified as the most effective for Dublin’s coastal-urban environment, accurately simulating the atmospheric "lid" that traps pollutants. The most significant advancement was the "enhancement" of local emissions. Rather than relying on static inventory values, researchers incorporated diurnal variations and temperature-dependent scaling, reflecting the increased use of solid fuels for residential heating during cold periods.
Meteorological Context and Model Validation
The most severe pollution episodes in Dublin typically coincide with "stagnation" events. These are characterized by low wind speeds and reduced boundary layer heights, which prevent the dispersion of locally generated pollutants.
As shown in the meteorological analysis, the pollution peaks on November 19, 2016, and January 22, 2017, occurred during periods of low temperature and stable atmospheric conditions. These conditions drive both an increase in residential heating demand and a decrease in the atmosphere's capacity to dilute emissions.
Key Insights: The Role of Residential Peat Burning
The study identifies residential peat and wood burning as "regionally significant" sources of organic aerosols. Unlike many European counterparts that have transitioned to cleaner energy, solid fuel use remains prevalent in Ireland.
Current inventories like MapEire may significantly underestimate these emissions. For instance, while census-based data might suggest low average concentrations, the localized impact of a single household burning wood or peat is immense. The discrepancy suggests that "secondary" heating—lighting a fire for comfort despite having central heating—is a major contributor missing from standard datasets.
Results and Implications
By incorporating temperature-dependent emissions, the refined WRF-Chem model successfully reproduced PM1 spikes exceeding 200 μg/m³. The results confirm that while long-range transport contributes to background levels, extreme urban smog events in Dublin are primarily driven by local residential activity.
These findings have broader implications for climate modeling. If models underestimate organic aerosols, they likely underestimate the associated negative radiative forcing (cooling effect). This could lead to an overestimation of global warming trends in current climate projections.
Conclusion
The failure of standard models to predict heavy pollution events in Dublin highlights the need for more dynamic emission inventories. Integrating top-down measurements (satellite and ground-based) with bottom-up inventories is essential for improving early warning systems and accurately forecasting air quality. Future work should focus on refining the speciation and temporal profiles of solid fuel emissions across various urban landscapes.
Summary updated