Why is ozone in South Korea and the Seoul metropolitan area so high and increasing?

Surface ozone pollution in South Korea has increased over the past 2 decades, despite efforts to decrease emissions, and is pervasively in exceedance of the maximum daily 8 h average (MDA8) standard of 60 ppb. Here, we investigate the 2015–2019 trends in surface ozone and NO2 concentrations over South Korea and the Seoul metropolitan area (SMA), focusing on the 90th percentile MDA8 ozone as an air quality metric. We use a random forest algorithm to remove the effect of meteorological variability on the 2015–2019 trends and find an ozone increase of up to 1.5 ppb a−1 in April–May, while NO2 decreases by 22 %. Global 3-D atmospheric chemistry model simulations including recent chemical updates can successfully simulate surface ozone over South Korea and China as well as the very high free-tropospheric ozone observed above 2 km altitude (mean 75 ppb in May–June) and can reproduce the observed 2015–2019 emission-driven ozone trend over the SMA including its seasonality. Further investigation of the model trend for May, when meteorology-corrected ozone and its increase are the highest, reveals that a decrease in South Korea NOx emissions is the main driver for the SMA ozone increase. Although this result implies that decreasing volatile organic compound (VOC) emissions is necessary to decrease ozone, we find that ozone would still remain above 80 ppb even if all anthropogenic emissions in South Korea were shut off. China contributes only 8 ppb to this elevated South Korea background, and ship emissions contribute only a few parts per billion. Zeroing out all anthropogenic emissions in East Asia in the model indicates a remarkably high external background of 56 ppb, consistent with the high concentrations observed in the free troposphere, implying that the air quality standard in South Korea is not practically achievable unless this background external to East Asia can be decreased.