Intercity-scale linkage of cold-season PM _2.5 and emissions over South Korea: structural diagnostics from scalable spatial decomposition

Abstract Although national emission control policies have contributed to long-term reductions in PM 2.5 levels, responses across neighboring cities remain heterogeneous, complicating the evaluation of policy effectiveness. This study diagnoses the spatial structure and examines temporal evolution of intercity-scale (tens of kilometers) and neighborhood-scale (a few kilometers) PM 2.5 components, separated from background variation including long-range transport, by applying a recently proposed spatial decomposition framework based on spatial coherence to long-term air quality monitoring network data in South Korea (2016–2023). Within this framework, a hierarchical averaging scheme uses coherence-based neighboring stations to define the intercity-scale variability of background-removed PM 2.5 concentrations. Focusing on the cold season (DJFM), spatial diagnosis reveals dominant intercity-scale contributions in the Seoul Metropolitan Area (SMA) and neighboring industrialized areas, highlighting the effectiveness of air control zone-based policies. In contrast, neighborhood-scale contributions predominate in high-PM 2.5 stations, particularly in the Southeastern Area (SEA), indicating the need for localized interventions. Temporal analysis reveals a structural transition beginning in 2019/20 DJFM, characterized by simultaneous trend reversals: intercity-scale components shifted from increasing to decreasing in the SMA, an area with intensive urban and industrial emissions, while the opposite occurred in the SEA, a coastal city cluster. This transition coincided with the implementation of the Seasonal PM Management policy and the COVID-19 lockdown. The variations in the intercity-scale component broadly align with residual emission intensities of PM 2.5 and NO x , defined as deviations from national emission trends based on the Clean Air Policy Support System emission inventory. These results reveal a structural shift in intercity-scale PM 2.5 variability and its linkage to emissions, offering a scalable diagnostic tool for evaluating policy effectiveness and underscoring the need for scale-specific air quality management strategies.