Abstract
This study examines the relationship between ship traffic and ambient air pollutant concentrations in Quintero Bay, Chile, an industrial coastal area characterized by intense port activity. The analysis integrates air quality measurements from the national monitoring network (SINCA) with vessel movement data derived from the AIS, combined with advanced time–frequency analysis techniques. Unlike traditional emission inventories, which aim to quantify emissions at the source, SINCA provides real-time ambient pollutant concentrations expressed in µg/m3 and ppb. Therefore, the objective of this study is not to directly estimate ship emissions, but to evaluate how variations in maritime activity co-vary with observed pollutant levels at air quality monitoring stations. The investigation focuses on pollutants commonly associated with shipping activities, including PM10 and PM2.5, NOx, SO2, and CO. A detailed characterization of the vessel fleet operating in the study area was performed using both static and dynamic AIS data. This included ship type classification, estimation of main engine power based on gross-tonnage models, and calculation of engine load distributions derived from AIS-reported vessel speeds. Although these parameters allow the estimation of emission factors, the present work emphasizes identifying temporal associations between ship movements and ambient pollutant concentrations rather than quantifying emissions. Additionally, pollution roses and Conditional Probability Functions (CPF) were used to evaluate the directional origin of high-concentration episodes in relation to prevailing wind conditions. The results show that several episodes of elevated NOx, SO2, and PM10 concentrations exhibit significant coherence with ship activity, particularly at daily time scales influenced by local meteorology. Among the analyzed pollutants, PM10 displays the strongest and most recurrent association with maritime traffic, highlighting the influence of port operations on air quality in Quintero Bay.
Keywords
Get full access to this article
View all access options for this article.
