To provide a nationwide representative dataset for the study on health impact of air pollution, we combined the data from the Korea National Health and Nutrition Examination Survey with the daily air quality and weather data by matching the date of examination and the residential address of the participants. The database of meteorological factors and air quality as sources of exposure data were estimated using the Community Multiscale Air Quality model. The linkage dataset was merged by three ways; administrative district, si-gun-gu (city, county, and district), and geocode (in latitude and longitude coordinate units) based on the participants’ residential address, respectively. During the study period, the exposure dataset of 85,018 individuals (38,306 men and 46,712 women) whose examination dates were recorded were obtained. According to the definition of exposure period, the dataset was combined with the data on short-term, mid-term, and long-term exposure to air pollutants and the meteorological indices. Calculation of the daily merged dataset’s average air pollution linked by si-gun-gu and geocode units showed similar results. This study generated a daily average of meteorological indices and air pollution exposure dataset for all regions including rural and remote areas in Korea for 11 years. It is expected to provide a platform for the researchers studying the health impact of air pollution and climate change on the representative population and area, which may facilitate the establishment of local health care plans by understanding the residents’ health status at the local as well as national level.

Despite recent progress of numerical air quality models, accurate prediction of fine particulate matter (PM_2.5) is still challenging because of uncertainties in physical and chemical parameterizations, meteorological data, and emission inventory databases. Recent advances in artificial neural networks can be used to overcome limitations in numerical air quality models. In this study, a deep neural network (DNN) model was developed for a 3 d forecasting of 6 h average PM_2.5 concentrations: the day of prediction (D+0), 1 d after prediction (D+1), and 2 d after prediction (D+2). The DNN model was evaluated against the currently operational Community Multiscale Air Quality (CMAQ) modeling system in South Korea. Our study demonstrated that the DNN model outperformed the CMAQ modeling results. The DNN model provided better forecasting skills by reducing the root-mean-squared error (RMSE) by 4.1, 2.2, and 3.0 µg m⁻³ for the 3 consecutive days, respectively, compared with the CMAQ. Also, the false-alarm rate (FAR) decreased by 16.9 %p (D+0), 7.5 %p (D+1), and 7.6 %p (D+2), indicating that the DNN model substantially mitigated the overprediction of the CMAQ in high PM_2.5 concentrations. These results showed that the DNN model outperformed the CMAQ model when it was simultaneously trained by using the observation and forecasting data from the numerical air quality models. Notably, the forecasting data provided more benefits to the DNN modeling results as the forecasting days increased. Our results suggest that our data-driven machine learning approach can be a useful tool for air quality forecasting when it is implemented with air quality models together by reducing model-oriented systematic biases.