Assessing the Mass Balance-based Inverse Modeling Methods to Constrain NOx Emissions in South Korea

Recently, O3 concentration in South Korea has increased rapidly compared to other countries. Therefore, there is an urgent need to identify the scientific causes of this phenomenon. Because NOx (NO + NO2) is a key air pollutant, acting as the main precursor in the photochemical production of O3, the accurate calculation of NOx emissions is essential for analyzing O3 levels. In order to calculate accurate NOx emissions in South Korea, this study constrains NOx emissions using mass balance-based inverse modeling methods and examines the optimization of inverse modeling conditions via a pseudo-observation test. We applied the following inverse modeling methods: Basic Mass Balance (BMB), Finite Difference Mass Balance (FDMB), and Iterative Finite Difference Mass Balance (IFDMB) in model simulation, and compared their efficiencies in correcting NOx emissions. We performed a numerical simulation of air quality using the Community Multi-scale Air Quality (CMAQ) model to calculate the NO2 columns and use the simulated results to constrain the NOx emissions. The pseudo-observation test was employed in this study to assess the estimated NOx emissions via comparison with synthetic observations based on real emissions data. We evaluated the inverse modeling conditions suitable for South Korea by comparing NOx emission errors before and after inverse modeling according to the season, modeling resolution, and regridding methodologies. Looking at the inverse modeling results, overall, IFDMB was the most effective method in constraining NOx emissions in South Korea among the three methods. The accuracy of inverse modeling was the highest in the summer season, and the 9 km resolution modeling was the most efficient in the region. Moreover, it is important to apply a suitable regridding method because the results of inverse modeling were different depending on the regridding methodologies. This study aimed to quantitatively evaluate the effect of the inverse modeling using pseudo-observation not actual. It is expected to be helpful in future studies using actual satellite observation data.

This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2020R1A6A1A03044834).