Fast and reliable source identification of criteria air pollutants in an industrial city

Abstract

Most of industrial cities in developed countries use automatic station networks for monitoring of Criteria Air Pollutants (CAPs), and the tremendous amount of data acquired are often used to verify that the concentrations are within safety levels and to warn the population in the other case. Furthermore, these data can be used to investigate the source-receptor relationship. In this study, the data were collected from automatic monitoring stations in Ulsan, the most industrialized city in South Korea. The dataset consists of hourly concentrations of five CAPs (SO2, CO, O3, NO2, and PM10) recorded at 13 monitoring stations in the city during a full year (March 2011-February 2012). Different types of sources were identified by studying the temporal (daily and seasonal) trends and spatial distributions of CAPs with wind directions. It was confirmed that SO2 pollution in Ulsan originated mostly from local industrial areas, whereas CO and NO2 were also substantially influenced by mobile sources. The high PM10 levels resulted from both local industries and traffic sources as well as from remote sources. The originality of this work comes from the study of the high episodes of pollutions on a case by case basis as well as on average data. Moreover, rather simple statistical tools developed in this study can be used for a real-time diagnosis of the local pollution in large urban and industrial areas.