Machine Learning Attribution of Atmospheric Black Carbon in a Coastal City: Source Variability and the Sea–Land Breeze Recirculation Mechanism

Black carbon (BC) variability in coastal environments reflects the combined effects of emission sources, regional transport, and mesoscale circulation. This study selected the typical coastal city of Qingdao as the study area, collected BC online observation data for one year in 2021, and integrated it with the Aethalometer model, trajectory analysis, and Random Forest (RF) machine learning framework to examine these mechanisms. The annual mean BC concentration was 1.78 ± 1.29 μg m–3, with fossil fuel combustion contributing 88.2% of the total mass, while the biomass burning contribution increased to 19% in winter. Source-resolved attribution showed distinct meteorological responses. Fossil-fuel black carbon (BCff) was more strongly associated with wind-controlled ventilation and transport, whereas biomass burning BC (BCbb) responded more strongly to low-temperature conditions. Under sea–land breeze (SLB) conditions, BC enhancement was most evident at wind speeds of 2.0 to 3.5 m s–1. Diurnal persistence, directional shifts in elevated fossil fuel BC, and interactions between wind speed and wind direction further supported return transport and reaccumulation within coastal circulation. These findings provide process-based insight into source-dependent BC variability and SLB-related reaccumulation in coastal environments.