Jinseok Kim

Fine particulate matter (PM2.5) pollution poses significant health risks, necessitating accurate predictions for effective management measures for the mitigator of PM2.5. While the Community Multiscale Air Quality (CMAQ) model is widely utilized to simulate PM2.5 concentrations, its computational demands often limit its application in testing various emission reduction scenarios. This study introduces a data-driven approach to emulate CMAQ simulations through AI surrogate model with a conditional U-Net architecture. The model, trained on a dataset of CMAQ-simulated PM2.5 concentrations from across South Korea in 2013, incorporates region-specific emission source activity data as inputs. It accurately predicts hourly changes in PM2.5 concentrations, facilitating the estimation of monthly and annual average concentrations under diverse emission reduction scenarios. Demonstrating robust performance in replicating the spatial and temporal patterns of CMAQ simulations, this surrogate model significantly cuts computational costs and times. Our findings offer policymakers a practical and efficient tool for evaluating the effectiveness of emission control strategies and enhancing decision-making processes in air quality management.

When predicting the health impacts of PM_2.5 from future air quality changes, it is crucial to consider both air quality improvements and population aging. This study divided future emission scenarios into a base and control scenario to project air quality from 2015 to 2030 and assess health outcomes. The GUIDE model, an Integrated Assessment Model (IAM), was used to estimate future emissions, while the CMAQ (Chemical Transport Model) and BenMAP (Health Impact Model) evaluated health impacts resulting from changes in air quality in Korea. The study focused on the impact of population aging on future health outcomes. Both scenarios showed improved PM_2.5 concentrations, with the control scenario showing more substantial improvements due to stronger policy measures. When applying current age patterns, health impacts decreased as PM_2.5 concentrations decreased. However, when considering future population aging, health impacts increased despite improved air quality. The results excluding aging show that the number of premature deaths due to cardiovascular disease and all other causes caused by PM_2.5 is 18,413 in the base year, while in the future control scenario, the number decreases to 11,729. In contrast, when aging is taken into account, the number of premature deaths increases to 23,037. This finding suggests that, although PM_2.5 concentrations are expected to decline, the increasing proportion of elderly individuals will exacerbate health risks. Therefore, accounting for aging population trends is essential when studying the health impacts of future air quality changes.

When predicting the health impacts of PM2.5 from future air quality changes, it’s crucial to consider both air quality improvements and population aging. This study divided future emission scenarios into a Base and Control scenario to project air quality from 2015 to 2030 and assess health outcomes. The GUIDE model, an Integrated Assessment Model (IAM), was used to estimate future emissions, while the CMAQ (Chemical Transport Model) and BenMAP (Health Impact Model) evaluated health impacts resulting from changes in air quality in Korea. The study focused on the impact of population aging on future health outcomes. Both scenarios showed improved PM2.5 concentrations, with the Control scenario seeing more substantial improvements due to stronger policy measures. When applying current age patterns, health impacts decreased as PM2.5 concentrations. However, when considering future population aging, health impacts increased despite improved air quality. This finding suggests that, although PM2.5 concentrations are expected to decline, the increasing proportion of elderly individuals will exacerbate health risks. Therefore, accounting for aging population trends is essential when studying the health impacts of future air quality changes.

There is a pressing need for tools that can rapidly predict PM2.5 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 PM2.5 concentrations and related health effects. The U-Net model was trained with 2019 South Korean PM2.5 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 PM2.5 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 PM2.5-related health risks and estimate economic benefits.

The GEMS MAP of Air Pollution (GMAP) 2021 field campaign for South Korea was conducted in October-November 2021 to understand the changes in air quality after the KORUS-AQ field study and to support efficient pollution management for ozone and aerosol. Extensive aircraft and ground network observations from the campaign offer an opportunity to reduce model-observation disagreements. This study examines these issues using model evaluation against the GMAP 2021 observations and intercomparisons between models. Four regional and one global chemistry transport model using identical anthropogenic emissions participated in the model intercomparison study. Based on the KORUSv5.0 emission inventory that supported the KORUS-AQ campaign, GMAP/SIJAQv2.0 emission inventory was developed to reflect the latest emission trends of major countries affecting South Korea’s air quality.From the results of the model using Global Forecast System (GFS) and final (FNL) Operational Global Analysis data during and after the campaign, the accuracy of the modeling results using FNL was higher than that of GFS, which is a result of informing that the accurate meteorological input data is important for the prediction of aerosol and ozone. In comparisons of simulated versus observed (AirKorea network) CO, O3, NO2, SO2, and PM2.5 concentrations in surface air averaged for the campaign period, the models successfully reproduced observed pollutants in surface air but similar to the results in KORUS-AQ showed low biases for carbon monoxide (CO), implying that there were possible missing CO sources in the inventory in East Asia. Observations show the highest values in the Seoul Metropolitan Area (SMA) and industrial regions except for O3, which is strongly titrated by high NOx levels from traffic emissions. Relative contributions to air quality in South Korea by local and long-range transport pollution influences were classified using the Brute Force Method (BFM) for the campaign period. Observed aerosol chemical composition at the Olympic Park ground site showed that inorganic components (nitrate, sulfate, ammonium) contributed to PM2.5 by 83% during the transboundary dominant case. On the other hand, in the case of local dominant, the contribution of organic carbonaceous aerosol was 42% to PM2.5, indicating a clear difference between the two cases. In model simulations, there is a difference in the ratio between organic and inorganic aerosol, but the difference between the two cases was well simulated. And models showed a tendency to simulate Elemental Carbon (EC) at a concentration more than twice as high as observed due to the effect of emissions. From the model evaluation, we find that ensemble results of multiple models show the most consistent results with observations during the campaign period. In addition to improving the accuracy of individual models and emission inventory, evaluation of model accuracy according to ensemble techniques is necessary to improve forecast results. This research was supported by the FRIEND(Fine Particle Research Initiative in East Asia Considering National Differences) Project through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT (Grant No.: 2020M3G1A1114615)