Development of PM2.5 management strategy in the industrial city of Ulsan: a comprehensive approach to understanding the risk and sources of PM2.5

Abstract

Particulate matter (PM) with an aerodynamic diameter of less than 2.5 μm (PM2.5) is major air pollutants in northeast Asia, with their primary sources being fuel combustion from industrial activity, heating, transportation, and power generation facilities. In addition, gaseous precursors such as sulfur oxides (SOX), nitrogen oxides (NOX), and volatile organic compounds (VOCs) generate secondary aerosols through chemical reactions. The major sources for high PM2.5 events in metropolitan cities are generally local vehicle emissions and secondary formation. Long-range atmospheric transport (LRAT) is also a major reason for high PM events in East Asian megacities such as the Seoul Metropolitan Area, which are frequently influenced by Asian continental outflow. On the other hand, industrial cities are often more strongly influenced by emissions from local industrial activity. The metropolitan city of Ulsan, with a population of 1.13 million, is located in the southeast of the Korean Peninsula. The east coast of Ulsan has a number of petrochemical, nonferrous, automobile, and shipbuilding facilities. Thus, the air quality in Ulsan is greatly influenced by the emissions of various air pollutants from these industrial complexes. However, studies on investigations of PM2.5 pollution in Ulsan have been limited, and there is a need to address the health of residents in an industrial city facing air pollution issues. In this study, we aim to develop comprehensive PM2.5 management strategies by considering source identification, human health assessment, and emission reduction. Currently, episodes of high levels of PM2.5 frequently occur in South Korea as a result of both local emissions and the LRAT of yellow dust and haze events from the Asian continent. Therefore, we investigated the characteristics of PM2.5 pollution episodes semi-continuous measurements obtained from the Yeongnam intensive air quality monitoring station (YN station) in Ulsan. The major source of PM2.5 for the pollution period during winter was LRAT from eastern China and North Korea. The industrial facilities in Ulsan were also responsible for the elevated PM2.5 concentration in winter. The major source of PM2.5 for the pollution period during summer was the local industrial facilities and ship emissions. In addition, secondary formation was enhanced by the air stagnation, high relative humidity, and low PBL height. The influence of thermal power stations and national industrial areas in southern coastal cities was also identified. Moreover, to comprehensively understand the pollution of primary and secondary PM2.5, we developed a technique by combining monitoring and modeling methods to map the spatial distribution of PM2.5. The monitoring data for PM2.5 components and precursors, as well as the air dispersion and receptor modeling data for Ulsan, South Korea were used. It was revealed that the petrochemical and non-ferrous industrial complexes are primary sources of PM2.5 in Ulsan. Similar levels between primary and secondary sulfate were observed, while nitrate concentrations were more influenced by secondary formation rather than primary emissions. Ammonium sulfate concentrations were significantly influenced by industrial activities, while ammonium nitrate concentrations were influenced by both industrial and urban emissions. Significant contributions to SOA formation were observed from the automobile, shipbuilding, and petrochemical industrial complexes, with aromatic compounds such as BTEX playing a significant role. Based on these findings, optimized strategies for managing PM2.5 were proposed for urban areas in individual districts and industrial complexes in Ulsan. However, it should be noted that the estimated concentration of SOA may be underestimated due to the limited number of analyzed VOC species. Particulate matter (PM) contains hazardous air pollutants (HAPs) that may adversely affect human health. In particular, residents living in an industrial city are seriously concerned about the health risks associated with major drivers of cancer risk, such as polycyclic aromatic hydrocarbons (PAHs) and heavy metals. Therefore, in this study, a novel index called the comprehensive air-risk index (CARI) was developed, which represents the human health risks associated with HAPs. Furthermore, to enhance the spatiotemporal resolution of CARI, a machine-learning approach was implemented using measurement data for PAHs and heavy metals. Over the course of eight years, the risk of PAHs decreased, whereas the risk of heavy metals exhibited a different trend in Ulsan. In addition, the trends of PM2.5 concentration and risk can differ in Ulsan. Earlier studies have also documented elevated concentrations of highly toxic heavy metals in PM2.5 in Ulsan. CARI, CAI and AQHI displayed different seasonal patterns. Hence, in large industrial cities, the proportion of HAPs within PM2.5 and meteorological conditions bears greater significance in health risk assessment than the mass concentration of PM2.5. The effectiveness of CARI in reflecting the health risks associated with HAPs in an industrial city can be conclusively affirmed. Utilizing machine learning and the novel risk index, CARI allows for the identification of priority risk areas in an industrial city at high spatio-temporal resolution. While the average CARI was higher in petrochemical and nonferrous industrial areas, it did not surpass the ‘Unhealthy’ threshold of 150, and there were areas that only exceeded the ‘Unhealthy for sensitive groups’ level of 100. In addition, the Onsan and Yaeum districts were identified as high-risk areas, despite having low population density, as they are primarily industrial areas. Finally, to propose the most effective PM2.5 emission reduction policies in Ulsan, machine learning approach was considered based on reduction scenarios. Emission reduction scenarios were simulated for the major components of PM2.5, including SO4 2-, NO3 -, NH4 +, OC, and EC, which contribute significantly to the PM2.5 mass concentration. When the concentrations of five components were uniformly reduced, PM2.5 mass concentration exhibited the most significant decrease in the scenario where OC concentration was reduced. Moreover, 'Bad' days determined by PM2.5 concentration (exceed 35 µg/m3) also showed the greatest decreased in the scenario with decreased OC concentration. SO2 and meteorological conditions were found to be the main factors for CARI related to human health risks caused by HAPs. In addition to SO2, NH4 +, OC, and EC showed high importance. When simulating the concentration reduction scenarios for these four components, the CARI decreased the most when the SO2 concentration was reduced. Moreover, 'Unhealthy' days determined by CARI (exceed 150) also showed the greatest decreased in the scenario with decreased SO2 concentration. This means that SO2 emission should be prioritized for control to reduce the risk of PM2.5. In conclusion, OC-related (i.e. VOCs) emissions should be controlled to efficiently reduce PM2.5 mass concentration, and SO2-related emissions (i.e. industrial activity) should be controlled to reduce human health risk. These studies provide basic information for improving air quality and will benefit the residents of Ulsan. It can also be applied to other major cities around the world to help improve global air quality.