The air quality in Seoul was gradually improved through quantitative management targeting major emission sources. However, the health effect of PM2.5 exists even at lower concentrations than the national ambient air quality standard. Therefore, in addition to quantitative management targeting major emission sources, qualitative management that reduces hazardous components of PM2.5 is required. In Seoul, to effectively manage PM2.5, concentrations of chemical components of PM2.5 have been measured since 2015. Using these data, this study aimed to derive components that have high health risks by analyzing the health effects of individual components, reviewing related emission sources of the components, and suggesting directions for future air quality management to reduce the health risk of PM2.5.
As the concentration of PM2.5 decreases, the concentration of components also showed decreasing patterns, EC, one of the carbon species showed most significant decreases. Meanwhile, the degree of reductions of OC and ion components are relatively small.
The concentrations of PM components has not been continuously measured. Therefore we estimated missing values applying machine learning techniques. The prediced values well simulate the serial trends and variability of true values except for some metal components such as Ti.
The predicted concentration of PM2.5 chemical components is linked to health data. For health data, the number of daily outpatient visits due to respiratory and cardiovascular diseases was calculated through the customized DB of the National Health Insurance Service. To compare health risk of chemical components of PM2.5, we estimated percent increases of hospital outpatient visit by increase of 1-S.D.(standard deviation) of each chemical component.
OC, EC, NO3-, Na+, Al, and levoglucosan were significantly associated with increases in hospital outpatient visits. For cardiovascular diseases, NO3- and Na+ are significantly associated with increases in hospital visits. When the NO3- concentration increases by 1.91㎍/㎥, outpatient visits to due to hypertension increase by 2.5% in the population aged 85 or older. For respiratory diseases, OC, EC, NO3-, Na+, Al, and levoglucosan are significantly associated with increases in hospital visits. When EC concentration increases by 0.51㎍/㎥, outpatient visits due to respiratory diseases increase by 6.1% in the population less than 20 years old. As a result of decreases in the annual average of short-term exposures of OC, EC, NO3-, Na+, Al, and levoglucosan, the attributable hospital outpatient visit number and fraction were decreased.
The major emission sources of the identified 6 components are mainly road and non-road emissions, biomass combustion, non-commercial combustion, and non-exhaustive emissions. Therefore, to reduce the health effects of PM2.5, it is necessary to reinforce the emission reduction strategy in these emission sources. In addition, in the case of secondary pollutants whose generation mechanism is not yet clear such as Na+ and OC, to reduce health effects, studies for identifying the generation mechanisms and precursor should be conducted. Al is known as one of the key tracers for non-exhaust emissions, but major sources and emissions are not clear. To reduce Al concentration, studies that identify sources and emissions should be conducted, especially for non-exhaust emissions.