The Korea-United States Air Quality Study (KORUS-AQ) took place in spring 2016 to better understand air pollution in Korea. In support of KORUS-AQ, 2554 whole air samples (WAS) were collected aboard the NASA DC-8 research aircraft and analyzed for 82 C1–C10 volatile organic compounds (VOCs) using multi-column gas chromatography. Together with fast-response measurements from other groups, the air samples were used to characterize the VOC composition in Seoul and surrounding regions, determine which VOCs are major ozone precursors in Seoul, and identify the sources of these reactive VOCs. (1) The WAS VOCs showed distinct signatures depending on their source origins. Air collected over Seoul had abundant ethane, propane, toluene and n-butane while plumes from the Daesan petrochemical complex were rich in ethene, C2–C6 alkanes and benzene. Carbonyl sulfide (COS), CFC-113, CFC-114, carbon tetrachloride (CCl4) and 1,2-dichloroethane were good tracers of air originating from China. CFC-11 was also elevated in air from China but was surprisingly more elevated in air over Seoul. (2) Methanol, isoprene, toluene, xylenes and ethene were strong individual contributors to OH reactivity in Seoul. However methanol contributed less to ozone formation based on photochemical box modeling, which better accounts for radical chemistry. (3) Positive Matrix Factorization (PMF) and other techniques indicated a mix of VOC source influences in Seoul, including solvents, traffic, biogenic, and long-range transport. The solvent and traffic sources were roughly equal using PMF, and the solvents source was stronger in the KORUS-AQ emission inventory. Based on PMF, ethene and propene were primarily associated with traffic, and toluene, ethylbenzene and xylenes with solvents, especially non-paint solvents for toluene and paint solvents for ethylbenzene and xylenes. This suggests that VOC control strategies in Seoul could continue to target vehicle exhaust and paint solvents, with additional regulations to limit the VOC content in a variety of non-paint solvents.

We examine O3 production and its sensitivity to precursor gases and boundary layer mixing in Korea by using a 3-D global chemistry transport model and extensive observations during the KORea-US cooperative Air Quality field study in Korea, which occurred in May–June 2016. During the campaign, observed aromatic species onboard the NASA DC-8 aircraft, especially toluene, showed high mixing ratios of up to 10 ppbv, emphasizing the importance of aromatic chemistry in O3 production. To examine the role of VOCs and NOx in O3 chemistry, we first implement a detailed aromatic chemistry scheme in the model, which reduces the normalized mean bias of simulated O3 mixing ratios from –26% to –13%. Aromatic chemistry also increases the average net O3 production in Korea by 37%. Corrections of daytime PBL heights, which are overestimated in the model compared to lidar observations, increase the net O3 production rate by ~10%. In addition, increasing NOx emissions by 50% in the model shows best performance in reproducing O3 production characteristics, which implies that NOx emissions are underestimated in the current emissions inventory. Sensitivity tests show that a 30% decrease in anthropogenic NOx emissions in Korea increases the O3 production efficiency throughout the country, making rural regions ~2 times more efficient in producing O3 per NOx consumed. Simulated O3 levels overall decrease in the peninsula except for urban and other industrial areas, with the largest increase (~6 ppbv) in the Seoul Metropolitan Area (SMA). However, with simultaneous reductions in both NOx and VOCs emissions by 30%, O3 decreases in most of the country, including the SMA. This implies the importance of concurrent emission reductions for both NOx and VOCs in order to effectively reduce O3 levels in Korea.

Gaseous emissions from vehicles contribute substantially to air pollution and climate change. Vehicular emissions also contain secondary pollutants produced via chemical reactions that occur between the emitted gases and atmospheric air. This study aims at understanding patterns concerning emission of regulated, greenhouse, and precursor gases, which demonstrate potential for secondary aerosol (SA) formation, from vehicles incorporating different engine technologies—multi-point injection (MPI) and gasoline direct injection (GDI)—and using different fuels—gasoline, liquefied petroleum gas (LPG), and diesel. Drive cycles from the National Institute of Environmental Research (NIER) were used in this study. Results obtained from drive cycle tests demonstrate a decline in aggregate gas emissions corresponding to an increase in average vehicle speed. CO₂ accounts for more than 99% of aggregate gaseous emissions. In terms of concentration, CO and NH₃ form predominantly non-CO₂ emissions from gasoline and LPG vehicles, whereas nitrogen oxides (NOx) and non-methane hydrocarbons (NMHC) dominate diesel-vehicle emissions. A higher percentage of SO₂ is emitted from diesel vehicles compared to their gasoline- and LPG-powered counterparts. EURO-5- and EURO-6-compliant vehicles equipped with diesel particulate filters (DPFs) tend to emit higher amounts of NO₂ compared to EURO-3-compliant vehicles, which are not equipped with DPFs. Vehicles incorporating GDI tend to emit less CO₂ compared to those incorporating MPI, albeit at the expense of increased CO emissions. The authors believe that results reported in this paper concerning regulated and unregulated pollutant-emission monitoring can contribute towards an accurate evaluation of both primary and secondary air-pollution scenarios in Korea.

We propose a method to adjust contributions from upwind emissions to downwind PM_2.5 concentrations to account for the differences between observed and simulated PM_2.5 concentrations in an upwind area. Emissions inventories (EI) typically have a time lag between the inventory year and the release year. In addition, traditional emission control policies and social issues such as the COVID-19 pandemic cause steady or unexpected changes in anthropogenic emissions. These uncertainties could result in overestimation of the emission impacts of upwind areas on downwind areas if emissions used in modeling for the upwind areas were larger than the reality. In this study, South Korea was defined as the downwind area while other regions in Northeast Asia including China were defined as the upwind areas to evaluate applicability of the proposed adjustment method. We estimated the contribution of emissions released from the upwind areas to PM_2.5 concentrations in South Korea from 2015 to 2020 using a three-dimensional photochemical model with two EIs. In these two simulations for 2015-2020, the annual mean foreign contributions differed by 4.1-5.5 µg/m³. However, after adjustment, the differences decreased to 0.4-1.1 µg/m³. The adjusted annual mean foreign contributions were 12.7 and 8.8 µg/m³ during 2015-2017 and 2018-2020, respectively. Finally, we applied the adjustment method to the COVID-19 pandemic period to evaluate the applicability for short-term episodes. The foreign contribution of PM_2.5 during the lockdown period in China decreased by 30% after adjustment and the PM_2.5 normalized mean bias in South Korea improved from 15% to -4%. This result suggests that the upwind contribution adjustment can be used to alleviate the uncertainty of the emissions inventory used in air quality simulations. We believe that the proposed upwind contribution adjustment method can help to correctly understand the contributions of local and upwind emissions to PM_2.5 concentrations in downwind areas.

With global anthropogenic black carbon (BC) emissions increasing, automobiles are significantly contributing as the major source of emissions. However, the appropriate regulations of BC emissions from vehicles are not in place. This study examined BC emissions following fuel types (gasoline, liquefied petroleum gas (LPG), and diesel) and engine combustion (gasoline direct injection (GDI) and multi-port injection (MPI) for gasoline vehicles) with emission regulations. To this end, chassis dynamometer and aethalometer (AE33) were used. Driving modes created by the National Institute of Environmental Research (NIER) and emission certification modes (CVS-75 and NEDC) for vehicles in Korea were used to determine BC emissions for various vehicle speeds. In addition, the contributions of biomass and coal combustion to the data of AE33 were analyzed to determine the possibility of tracking the BC sources. MPI, LPG, and EURO 6 with diesel particulate filter (DPF) vehicles emitted the lowest BC emissions in NIER modes. Among gasoline vehicles, MPI vehicles showed the lower BC content in PM emissions. Also, older vehicles in MPI vehicles emitted the high PM and BC emissions. The BC emissions of EURO 3 vehicles without DPF were the highest as the results of previous studies, and it was found that as emissions regulations were tightened, the level of BC results of diesel vehicles became similar with MPI vehicles. The average absorption Ångström exponent (AAE) from difference emissions sources were biomass combustion (oak wood) > coal combustion (the power plant stack) > automobile emissions (gasoline, LPG, diesel).

There is a pressing need for tools that can rapidly predict ${\mathrm{PM}}_{2.5}$ concentrations and assess health impacts under various emission scenarios, aiding in the selection of optimal mitigation strategies. Traditional chemical transport models (CTMs) like CMAQ are accurate but computationally intensive, limiting practical scenario analysis. To address this, we propose a novel method integrating a conditional U-Net surrogate model with health impact assessments, enabling swift estimation of ${\mathrm{PM}}_{2.5}$ concentrations and related health effects. The U-Net model was trained with 2019 South Korean ${\mathrm{PM}}_{2.5}$ data, including precursor emissions and boundary conditions. Our model showed high accuracy and significant efficiency, reducing processing times while maintaining reliability. By combining this surrogate model with the EPA’s BenMAP-CE tool, we estimated potential premature deaths under various emission reduction scenarios in South Korea, extending projections to 2050 to account for demographic changes. Additionally, we assessed the required ${\mathrm{PM}}_{2.5}$ emission reductions needed to counteract the increase in premature deaths due to an aging population. This integrated framework offers an efficient, user-friendly tool that bridges complex air quality modeling with practical policy evaluation, supporting the development of effective strategies to reduce ${\mathrm{PM}}_{2.5}$-related health risks and estimate economic benefits.