Korean Journal of Remote Sensing (대한원격탐사학회지)

Korean Society of Remote Sensing (대한원격탐사학회)
권호 표지
DOI QR코드

DOI QR Code

Analysis of Ozone Formation Sensitivity in the Seoul Metropolitan Area Using TROPOMI FNR (HCHO/NO2 Ratio) from 2019 to 2024

  • Ji-Yun Lee (Department of Earth Environmental and Space Sciences, Chungnam National University) ;
  • Jinah Jang (Greenhouse Gas Inventory and Research Center of Korea) ;
  • Jaemin Kim (Research Institute of Natural Sciences, Chungnam National University) ;
  • Jeong-Ah Yu (Environmental Satellite Center, National Institute of Environmental Research) ;
  • Seung-Yeon Kim (Environmental Satellite Center, National Institute of Environmental Research) ;
  • Kyoung-Hee Sung (Environmental Satellite Center, National Institute of Environmental Research) ;
  • Hyo-Jung Lee (Department of Atmospheric Sciences, Pusan National University) ;
  • Yun Gon Lee (Department of Earth Environmental and Space Sciences, Chungnam National University)
  • Received : 2025.10.20
  • Accepted : 2025.11.24
  • Published : 2025.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

Tropospheric ozone negatively impacts human health and air quality, and its formation is driven by nonlinear interactions between nitrogen oxides (NOx) and volatile organic compounds (VOCs). Therefore, understanding ozone formation mechanisms is essential for establishing effective air quality management strategies. South Korea exhibits spatial variations in ozone and its precursors (HCHO, NO2) due to regional differences in topography and pollutant emissions, with distinct ozone trends being evident even within the Seoul Metropolitan Area-Seoul, Incheon, and Gyeonggi Province. This study divided the Seoul Metropolitan Area into Seoul region and surrounding (Gyeonggi-Incheon) region and assessed regimes of ozone formation sensitivity using satellite-based formaldehyde-to-nitrogen dioxide ratio (FNR) data from June to September during 2019-2024. Three methods for deriving FNR threshold values were compared: high-ozone case-based method (Method 1), concentration-exceedance-limited method (Method 2), and total ozone concentration-based method (Method 3). Based on Method 3, which reflects overall ozone concentrations, regimes of ozone formation sensitivity were analyzed for 25 ground-level ozone observation sites in the Seoul region and 120 in the Gyeonggi (Gyeonggi-Incheon) region. Between 2019 and 2024, the number of VOC-limited regime sites decreased from 23 to 2 in the Seoul region and from 78 to 24 in the Gyeonggi region; conversely, the number of Transitional regime sites increased from 2 to 15 in the Seoul region and from 15 to 46 in the Gyeonggi region. Over the same period, in both the Seoul and Gyeonggi regions from June to September, NO2 concentrations decreased more rapidly than HCHO concentrations, leading to an increase in FNR values. Consequently, sites that were previously in the VOC-limited regime shifted to the Transitional regime, increasing the proportion of areas classified as Transitional. This trend suggests that simultaneous reductions in NOx and VOCs are necessary to effectively mitigate ozone pollution in the Seoul Metropolitan Area. These results enhance the understanding of ozone formation mechanisms and provide a foundation for developing region-specific air quality management policies.

Keywords

1. Introduction

Tropospheric ozone (O₃) adversely affects human health and air quality (Aneja and Li, 1992; Ashmore, 2005; Karnosky et al., 2007; Kim et al., 2020; Xie et al., 2019). In South Korea, previous studies reported that a 10-μg/m³ increase in the maximum daily 8-h average O₃ (MDA8 O₃) during summer is associated with a 0.45% increase in daily mortality (Chen et al., 2014). O₃ is formed through the photochemical oxidation of volatile organic compounds (VOCs) and carbon monoxide (CO) in the presence of nitrogen oxides (NOx; NO + NO₂) (Atkinson, 2000). O₃ formation sensitivity to its precursors—NOx and VOCs—is nonlinear (Kleinman, 1994; Sillman, 1999). Consequently, the effectiveness of emission reduction strategies depends on whether reductions target NOx or VOCs, as responses vary according to O₃ concentration levels (Geddes et al., 2009; Jin et al., 2017; Lee et al., 2021; Li et al., 2013).

Formaldehyde (HCHO)—an intermediate product formed during the oxidation of various VOCs—exhibits a reactivity proportional to its total reaction rate with OH radicals, making it a useful proxy for VOCs (Duncan et al., 2010; Lamsal et al., 2014; Martin et al., 2003). Similarly, nitrogen dioxide (NO₂), which constitutes the majority of atmospheric NOx, serves as an NOx proxy (Martin et al., 2004; Shen et al., 2019; Zhu et al., 2014). Thus, the formaldehyde-to-nitrogen dioxide ratio (FNR) is regarded as a surrogate indicator of the VOCs/NOx ratio and can be used to assess region-specific O₃ formation sensitivity and diagnose the dominant precursors (Jang et al., 2023; Jin et al., 2020; Ren et al., 2022).

Ground-based observation methods can provide accurate in situ diagnoses of O₃ formation sensitivity; however, their spatiotemporal coverage is limited (Wang et al., 2017). Conversely, model-based methods can describe O₃ formation sensitivity over broader spatiotemporal scales; however, their accuracy is constrained by uncertainties in meteorological factors, photochemical reactions, and emission inventories (Liu and Shi, 2021). This study utilized satellite data providing photochemical information with temporal continuity and global representativeness as well as continuous observations of NO₂ and HCHO to calculate the HCHO/NO₂ ratio as an indicator of O₃ formation sensitivity (Ren et al., 2022).

Various methodologies exist for evaluating O₃ formation sensitivity using FNR threshold values based on observational data; however, they can yield different results even under the same spatiotemporal conditions. For example, Jin et al. (2020) classified surface O₃ formation sensitivity based on FNR values and the likelihood of high O₃ events, whereas Jang et al. (2023) and Ren et al. (2022) assessed it through the relationship between FNR values and surface O₃ concentrations. Additionally, Wang et al. (2021) suggested a method based on the statistical characteristics of monthly mean surface O₃ concentrations.

In South Korea, surface O₃ concentrations exhibit large spatial and temporal variability (Jeong et al., 2020; Yeo and Kim, 2021). The Seoul Metropolitan Area (SMA; Seoul, Incheon, and Gyeonggi Province)—characterized by a high density of electronic and chemical industries—is a representative region of anthropogenic pollution (Oak et al., 2025). During summer, O₃ concentrations in the SMA are notably higher than those in other regions with similar meteorological conditions, i.e., temperature and wind speed. This pattern cannot be explained solely by meteorological factors; hence, local surface O₃ formation mechanisms play a critical role in the SMA O₃ problem (Kim et al., 2023).

South Korea exhibits regional differences in topography and pollutant emission characteristics, resulting in spatial variations in ozone and its precursors (HCHO, NO₂) (Jeong et al., 2020). Distinct ozone trends are evident even within the SMA (Kim et al., 2023). In this study, the SMA was divided into Seoul region and surrounding (Gyeonggi–Incheon) region, and the O₃ formation sensitivity from 2019 to 2024 was analyzed using satellite-based FNR data.

The aim was to elucidate the differences in O₃ formation mechanisms between the Seoul and Gyeonggi (Gyeonggi–Incheon) regions and provide foundational data for developing region-specific, tailored air quality management policies that account for regional characteristics.

2. Materials and Methods

2.1. Ground-Based O₃ Observations

Surface O₃ concentrations were analyzed using the hourly average final verified values provided by the Air Quality Information Open System of the Korea Environment Corporation (AirKorea, https://www.airkorea.or.kr/). These data are reported in parts per million (ppm). To ensure data reliability, only urban air-monitoring stations with at least 75% of monthly data (June–September) available during the 2019–2024 period were included. During the study period, 440 urban air-monitoring stations operated across South Korea. The analysis was based on 25 and 120 stations situated in the Seoul and Gyeonggi regions, respectively (Fig. 1).

OGCSBN_2025_v41n6_985_3_f0001.png 이미지

Fig. 1. Locations of AirKorea monitoring sites (white dots) in the Seoul and Gyeonggi regions and spatial distributions of tropospheric NO2 (a) and HCHO (b) from TROPOMI observations during June–September 2019–2024.

The MDA8 O₃ metric is widely used for conducting air quality assessments and health impact studies in several countries and regions, including the United States, Europe, and China (Malley et al., 2017; Turner et al., 2016). Here, the annual changes in surface O₃ concentrations across the SMA were assessed using the MDA8 O₃ values. For each day, 24-h surface O₃ data were averaged over all consecutive 8-h periods, and among the resulting averages, the highest value, i.e., the MDA8 O₃, was calculated based on measurements from all national regulatory monitoring stations within each city. This value is considered representative of the daily O₃ concentration. Furthermore, the analysis of O₃ formation sensitivity regimes was based on hourly average O₃ concentrations measured at 13:30—corresponding to the TROPOspheric Monitoring Instrument (TROPOMI) satellite observation time. These values served as the basis for calculating FNR threshold values.

2.2. Satellite Observations of O₃ Precursors

TROPOMI onboard the Copernicus Sentinel-5 Precursor satellite enables daily global observations of atmospheric constituents (Veefkind et al., 2012). The TROPOMI observation time coincides with the period of peak photochemical O₃ production, making it suitable for O₃ formation sensitivity analyses. During this time, the solar zenith angle is low, and the atmospheric boundary layer is well developed, allowing the instrument to sensitively detect lower HCHO and tropospheric NO₂ concentrations (Jin et al., 2017).

The TROPOMI data used here were Level-2 HCHO and tropospheric NO₂ vertical column data (version 2) obtained from the Earthdata Search portal of NASA (https://search.earthdata. nasa.gov/). To ensure data quality, the following filtering criteria were applied: tropospheric NO₂ quality flag ≥ 0.75; HCHO quality flag ≥ 0.5; cloud fraction ≤ 0.3. The filtered TROPOMI data were regridded by averaging the values within each 0.0625° × 0.0625° grid cell for subsequent analysis. Administrative boundary shapefiles for the Seoul and Gyeonggi regions were used to spatially extract satellite observations within the study area using the open-source GeoPandas library in Python (Fig. 1).

2.3. Methods

Previous studies focused on the “warm season” (i.e., April–September), when O₃ concentrations increase due to active photochemical reactions involving both NOx and VOCs, and considered the relatively short atmospheric lifetime of O₃ (Ren et al., 2022). For the study region (i.e., SMA), data from June to September between 2019 and 2024 were combined, including TROPOMI HCHO and tropospheric NO₂ observations and surface O₃ measurements, to minimize the influence of long-range transport.

O₃ formation sensitivity can be classified into three regimes: In the VOC-limited (or NOx-saturated) regime, where HCHO is low, and NO₂ is high, O₃ concentrations increase with either a reduction in NO₂ or an increase in HCHO. In the NOx-limited regime, where NO₂ is low and HCHO is high, O₃ increases with increasing NO₂. In the Transitional regime, where both precursors are high, O₃ exhibits a peak and increases with simultaneous increases in both HCHO and NO₂ (Jang et al., 2023; Jin et al., 2020; Ren et al., 2022). Based on this theoretical classification, and following previous studies indicating that high O₃ concentrations primarily occur in the Transitional regime (Jang et al., 2023; Jin et al., 2020; Ren et al., 2022; Wang et al., 2021), this study calculated FNR threshold values using three different methods.

Method 1 involved matching observed O₃ values with the corresponding FNR values on the same dates and dividing the FNR range (0–8) into 200 equal-width bins. The ozone exceedance probability—defined as the probability of exceeding the MDA8 O₃ standard of 60 parts per billion by volume (ppbv) in Korea (adjusted from the U.S. standard of 70 ppbv to reflect domestic conditions)—was calculated for each bin. A cubic polynomial regression was subsequently performed, and the FNR value corresponding to the maximum of the fitted curve was defined as the FNR threshold, with the Transitional regime defined as the top 10% of the fitted curve.

Method 2 defined the FNR threshold as the median FNR value corresponding to the top 10% of monthly O₃ concentrations that exceed 60 ppbv, and the Transitional regime was defined as the range within ±30% of this threshold. Method 3 followed a procedure similar to that of Method 1, matching observed O₃ values with the corresponding FNR values on the same dates and dividing the FNR range (0–8) into 200 equal-width bins. Instead of calculating exceedance probabilities, the mean O₃ concentration within each bin was used. A cubic polynomial regression was then applied, and the FNR value corresponding to the maximum of the fitted curve was defined as the FNR threshold, with the Transitional regime defined as the top 10% of the fitted curve.

3. Results

3.1. Temporal Variation Analysis of Observed O₃ and its Precursors

Here, the annual variations of O₃ and its precursors in the Seoul and Gyeonggi regions were analyzed using MDA8 O₃, HCHO, and tropospheric NO₂ vertical column data from June to September (2019–2024) to assess how these values differ from the national averages. MDA8 O₃ serves as a key indicator directly related to human health and environmental regulations, whereas HCHO and NO₂ act as proxies for VOCs and NOx, respectively, playing an important role in O₃ formation sensitivity analysis.

Since 2019, tropospheric NO₂ concentrations have shown a nationwide decline, particularly in the Seoul and Gyeonggi regions. Between 2019 and 2024, levels decreased from 16.98 to 7.28 10¹⁵ molecules/cm² in the Seoul region and from 6.48 to 3.28 10¹⁵ molecules/cm² in the Gyeonggi region. This marked reduction is largely attributed to the Second Nitrogen Oxides (NOx) Reduction Policy under the Metropolitan Air Quality Management Basic Plan (2015–2024) (Metropolitan Air Quality Management Office, 2024). Conversely, HCHO concentrations have continued to decline steadily in the Seoul region, whereas slight upward trends were observed in the Gyeonggi region and nationwide (Fig. 2). O₃ concentrations, expressed in μg/m³, have increased consistently across the country since 2019, particularly in the Seoul and Gyeonggi regions. In the Seoul region, levels rose from 91.00 in 2019 to 121.38 in 2024. Although temporary decreases were recorded in 2020 (87.27) and 2022 (97.04), the overall trend has remained upward, as reflected in the intermediate increase observed in 2021 (105.06).

OGCSBN_2025_v41n6_985_4_f0001.png 이미지

Fig. 2. Temporal variations in MDA8 (maximum daily 8-h average) ozone (O3; μg/m3) calculated from hourly ground-level O3 concentrations, TROPOMI-retrieved HCHO (1015 molecules/cm2), and tropospheric NO3 (1015 molecules/cm2) from June to September (2019–2024) in (a) South Korea,(b) Seoul region, and (c) Gyeonggi region. Ground-level O3, HCHO, and tropospheric NO2 concentrations are represented bycircles, triangles, and squares, respectively. Linear trend lines and corresponding regression equations are included to illustrate interannual pollutant trends.

These variations in precursor concentrations across regions indicate that O₃ formation sensitivity regimes differ according to regional characteristics. Therefore, to capture the regional differences in O₃ formation sensitivity regimes more accurately and enhance the precision of policy evaluation and O₃ formation sensitivity analysis, this study applied region-specific FNR threshold values to analyze O₃ formation sensitivity.

3.2. Assessment of FNR Threshold Variability

Three methods for calculating FNR threshold values were compared and analyzed to assess O₃ formation sensitivity in the Seoul and Gyeonggi regions.

Method 1 estimates the probability of high O₃ events (>60 ppbv) based on the relationship between O₃ and its precursors as a function of FNR values (Jin et al., 2020). In the Seoul region, many FNR bins correspond to a zero probability of O₃ exceedance, which can introduce large fluctuations in the cubic polynomial curve. This results in instability in calculating the FNR threshold and poses challenges in obtaining a reliable value (Fig. 3a).

OGCSBN_2025_v41n6_985_5_f0001.png 이미지

Fig. 3. Thresholds of the TROPOMI HCHO and tropospheric NO2 ratio (FNR) from June to September during 2019–2024. (a, c, e) Results for the Seoul region. (b, d, f) Results for the Gyeonggi region. Panels are arranged in the order of Method 1 to Method 3.

Method 2 focuses on minimizing the influence of meteorological conditions and long-range transport by analyzing only cases where the monthly mean O₃ concentration exceeded 60 ppbv during the peak O₃ formation period from June to September. According to Wang et al. (2021), these exceedance events are primarily due to local photochemical reactions. The median FNR value of these cases is defined as the threshold, and the Transitional regime is set as ±30 % of this value. However, in regions with limited data, such as the Seoul region, interannual variability in the FNR threshold increases, reducing its reliability (Fig. 3c).

Method 3 derives the FNR threshold by performing regression analysis on the mean O₃ concentration within each FNR bin. This approach has the advantage of comprehensively considering O₃ concentrations across all spatiotemporal conditions rather than being limited to specific cases. However, if an FNR bin contains insufficient O₃ observations, the mean O₃ value for that bin may not be representative of the corresponding FNR value.

The comparison of the three methods showed that Methods 1 and 3 yielded similar FNR thresholds and Transitional regime ranges despite their different approaches. In contrast, Method 2 exhibited high variability and relatively low reliability due to limited data (Fig. 3c). Furthermore, Method 1 can be effectively applied in regions with sufficient high O₃ cases and corresponding FNR values; however, its applicability is limited in areas with low O₃ concentrations (Fig. 3a). Therefore, Method 3, which considers O₃ concentrations under all conditions and offers a broad spatial applicability, was selected for the final analysis in this study.

3.3. Analysis of O₃ and its Precursors According to O₃ Formation Sensitivity Regimes

Method 3 was used to classify O₃ formation sensitivity regimes in the Seoul and Gyeonggi regions from 2019 to 2024, applying FNR thresholds and Transitional regime values of 2.3 (range: 1.9–2.7) and 2.7 (range: 2.3–3.1), respectively, and analyzing the corresponding precursor concentrations. Based on the mean FNR values matched to each surface O₃ monitoring station during 2019–2024, the analysis of O₃ formation sensitivity regimes revealed that 14 (56.0%) and 11 (44.0%) stations (n = 25) in the Seoul region fell within the VOC-limited and Transitional regimes, respectively (Fig. 4a). In the Gyeonggi region 56 (46.7%), 38 (31.7%), and 26 (21.7%) stations (n = 120) fell within the VOC-limited, Transitional, and NOx-limited regimes, respectively (Fig. 4c).

OGCSBN_2025_v41n6_985_6_f0001.png 이미지

Fig. 4. Proportions of the VOC-limited, Transitional, and NOx-limited regimes for ground-level O3 based on the TROPOMI HCHO and tropospheric NO2 ratio from June to September during 2019–2024 and corresponding distributions of HCHO and tropospheric NO2 concentrations within each regime in the (a, b) Seoul and (c, d) Gyeonggi regions.

The precursor concentrations within each O₃ formation sensitivity regime were analyzed by region (Figs. 4b, d). HCHO concentrations were relatively consistent across the regimes, whereas NO₂ concentrations tended to decrease with increasing FNR values. This indicates that variations in NO₂ concentrations are closely associated with changes in O₃ formation sensitivity regimes. Furthermore, the spatial distribution analysis presented in Fig. 5 shows that HCHO concentrations were relatively uniform across the Seoul and Gyeonggi regions, whereas NO₂ concentrations were comparatively higher in the VOC-limited and Transitional regimes near industrial complexes in both regions.

OGCSBN_2025_v41n6_985_7_f0001.png 이미지

Fig. 5. Ground-level O3 monitoring sites classified by O3 formation sensitivity regimes based on regional TROPOMI HCHO and tropospheric NO2 ratio thresholds in the Seoul and Gyeonggi regions for June–September during 2019–2024 (indicated by dotcolors). The panels show also the spatial distributions of TROPOMI tropospheric NO2 (a) and HCHO (b) concentrations for the same period.

A comprehensive analysis of the annual proportions of O₃ formation sensitivity regimes (Fig. 6) and the year-to-year changes in O₃ and precursor concentrations (Fig. 2) revealed a notable increase in the proportion of the Transitional regime in 2023 and 2024. During this period, in the Seoul region, HCHO concentrations rose, whereas tropospheric NO₂ levels remained stable, resulting in a corresponding increase in O₃ concentrations. In the Gyeonggi region, HCHO concentrations also increased, whereas NO₂ concentrations decreased, leading to an expansion of the Transitional regime and a concomitant rise in O₃ concentrations.

OGCSBN_2025_v41n6_985_7_f0002.png 이미지

Fig. 6. Annual variations in the proportions of O3 formation sensitivity regimes based onTROPOMI HCHO and tropospheric NO2 ratio thresholds for the (a) Seoul and (b) Gyeonggi regions, with linear trend lines shown separately for each regime. Numbers above the bars indicate the number of O3 monitoring sites in each regime. The Seoul region had 25 sites from 2019–2024, whereas the number of sites in the Gyeonggi region increased from 103 in 2019 to 120 by 2021–2024.

Overall, in the Seoul region, the proportions of the VOC-limited and Transitional regimes decreased and increased, respectively. In the Gyeonggi region, the decrease in the VOC-limited regime occurred at a lower rate compared with that in the Seoul region, whereas the proportions of the Transitional and NOx-limited regimes gradually increased. These results suggest that effective O₃ mitigation in both the Seoul and Gyeonggi regions requires emission reduction policies targeting not only NOx but also VOCs.

4. Discussion and Conclusions

This study analyzed changes in surface MDA8 O₃ concentrations and satellite-based HCHO and tropospheric NO₂ in the Seoul and Gyeonggi regions in June–September from 2019 to 2024 and quantitatively evaluated region-specific O₃ formation sensitivity using FNR. Since 2019, tropospheric NO₂ concentrations have shown a nationwide decline—particularly in the Seoul and Gyeonggi regions. Between 2019 and 2024, levels decreased from 16.98 to 7.28 10¹⁵ molecules/cm² in the Seoul region and from 6.48 to 3.28 10¹⁵ molecules/cm² in the Gyeonggi region. In contrast, HCHO concentrations steadily declined in the Seoul region; however, they showed slight increases in the Gyeonggi region and nationwide. Since 2019, O₃ concentrations, expressed in μg/m³, have been consistently increasing across the country, with particularly pronounced rises occurring in the Seoul and Gyeonggi regions. In the Seoul region, O₃ levels increased from 91 to 121, with the overall trend remaining upward despite temporary decreases in 2020 and 2022. These regional differences reflect variations in precursor emissions and local characteristics, indicating that O₃ formation regimes operate differently across regions.

In this context, the study compared different methods for calculating FNR thresholds to precisely evaluate region-specific O₃ formation sensitivity. Among the three approaches, Method 3 —capable of reflecting O₃ concentrations under various spatiotemporal conditions rather than being limited to specific cases—was deemed the most appropriate for the final analysis. Using this method, the FNR thresholds and Transitional regime values for the Seoul and Gyeonggi regions were determined to be 2.3 (range: 1.9–2.7) and 2.7 (range: 2.3–3.1), respectively, which were then used to classify the O₃ formation sensitivity regimes in each region.

The analysis showed that HCHO concentrations did not differ markedly across O₃ formation sensitivity regimes, whereas NO₂ concentrations tended to decrease with increasing FNR values. This indicates that NO₂ variations contribute to driving transitions between O₃ formation regimes. Spatial distribution analysis confirmed this pattern: HCHO was relatively uniform across the study area, whereas NO₂ concentrations were comparatively higher in the VOC-limited and Transitional regimes near industrial complexes in both regions.

A combined analysis of the annual proportions of sensitivity regimes and precursor concentrations revealed a marked increase in the Transitional regime in 2023 and 2024. During this period, in both the Seoul and Gyeonggi regions, HCHO concentrations increased, whereas NO₂ concentrations decreased, accompanied by a corresponding O₃ concentration rise.

A Transitional regime expansion was observed in both the Seoul and Gyeonggi regions, suggesting that effective O₃ mitigation will require emission reduction policies targeting not only NOx but also VOCs. Nevertheless, this study has certain limitations. The use of satellite-based data may not fully account for external factors, such as aerosol effects, meteorological conditions, long-range transport, and stratospheric intrusions. In addition, uncertainties may arise from mismatches in spatial resolution between satellite and surface observations as well as algorithmic errors. Future studies should aim to achieve more accurate classification of O₃ formation sensitivity regimes by utilizing tropospheric O₃, NO₂, and HCHO data from geostationary satellites, such as GEMS. Overall, these results enhance the understanding of O₃ formation mechanisms and provide a foundation for developing region-specific air quality management policies.

Author Contributions

Conceptualization: Lee J-Y, Lee YG; Data curation: Lee J-Y, Jang JA, Kim JM; Formal analysis: Lee J-Y, Jang JA, Lee YG; Funding acquisition: Lee YG; Investigation: Lee J-Y, Jang JA, Kim JM; Methodology: Lee J-Y, Jang JA, Kim JM, Lee H-J, Lee YG; Project administration: Yu J-A, Kim S-Y, SungK-H; Supervision: Lee YG; Validation: Lee J-Y; Writing–original draft: Lee J-Y, Lee YG; Writing–review and editing: Lee J-Y, Lee YG.

Conflicts of Interest

No potential conflict of interest relevant to this article was reported.

Funding

This research was funded by the National Institute of Environmental Research (NIER) of the Ministry of Environment of the Republic of Korea (Grant No. NIER-2025-04-02-036) and by the Korea Meteorological Institute (Grant No. RS-2025-02221093).

Data Availability Statement

AirKorea surface O₃ data are openly available at https://www.airkorea.or.kr/eng/. TROPOMI Level-2 HCHO and tropospheric NO₂ data are openly available at https://search.earthdata.nasa.gov/.

Acknowledgments

None.

Supplementary Materials

None.

References

  1. Aneja, V.P., and Li, Z., 1992. Characterization of ozone at high elevation in the eastern United States: Trends, seasonal variations, and exposure. Journal of Geophysical Research: Atmospheres, 97(D9),9873-9888.https://doi.org/10.1029/92JD00503
  2. Ashmore, M. R., 2005. Assessing the future global impacts of ozone on vegetation. Plant, Cell & Environment, 28(8), 949-964. https://doi.org/10.1111/j.1365-3040.2005.01341.x
  3. Atkinson, R., 2000. Atmospheric chemistry of VOCs and NOx. Atmospheric Environment, 34(12-14),2063-2101. https://doi.org/10.1016/S1352-2310(99)00460-4
  4. Chen, R., Cai, J., Meng, X., Kim, H., Honda, Y., Guo, Y. L., et al., 2014. Ozone and daily mortality rate in 21 cities of East Asia: How does season modify the association?.American Journal of Epidemiology, 180(7), 729–736. https://doi.org/10.1093/aje/kwu183
  5. Duncan, B. N., Yoshida, Y., Olson,J. R., Sillman, S., Martin, R. V., Lamsal, L., et al., 2010. Application of OMI observations to a space-based indicator of NOx and VOC controls on surface ozone formation. Atmospheric Environment, 44(18), 2213–2223. https://doi.org/10.1016/j.atmosenv.2010.03.010
  6. Geddes, J. A., Murphy, J. G.,and Wang, D.K., 2009. Long term changes in nitrogen oxides and volatile organic compounds in Toronto and the challenges facing local ozone control. Atmospheric Environment, 43(21), 3407-3415. https://doi.org/10.1016/j.atmosenv.2009.03.053
  7. Jang, J., Lee, Y. G., Yu, J.-A., Sung, K.-H., and Kim, S.-M., 2023. Characteristic analysis of tropospheric ozone sensitivity from the satellite-based HCHO/NO2 ratio in South Korea. Korean Journal of Remote Sensing, 39(5–1), 563–576. https://doi.org/10.7780/kjrs.2023.39.5.1.8
  8. Jeong, Y., Lee, H. W., and Jeon, W., 2020. Regional differences of primary meteorological factors impacting O3 variability in South Korea. Atmosphere, 11(1), 74. https://doi.org/10.3390/atmos11010074
  9. Jin, X., Fiore, A., Boersma, K. F., De Smedt, I., and Valin, L., 2020. Inferring changes in summertime surface Ozone–NOx–VOC chemistry over US urban areas from two decades of satellite and ground-based observations. Environmental Science & Technology, 54(11), 6518–6529. https://doi.org/ 10.1021/acs.est.9b07785
  10. Jin, X., Fiore, A. M., Murray, L. T., Valin, L. C., Lamsal, L.N., Duncan, B., et al., 2017. Evaluating a space-based indicator of surface ozone-NOx-VOC sensitivity over midlatitude source regions and application to decadal trends.Journal of Geophysical Research: Atmospheres, 122(19), 10439-10461. https://doi.org/10.1002/2017JD026720
  11. Karnosky, D.F., Skelly, J.M., Percy, K. E., and Chappelka, A.H., 2007. Perspectives regarding 50 years of research on effects of tropospheric ozone air pollution on US forests. Environmental Pollution, 147(3),489-506. https://doi.org/10.1016/j.envpol.2006.08.043
  12. Kim, S.-W., Kim, K.-M., Jeong, Y., Seo, S., Park, Y., and Kim, J., 2023. Changes in surface ozone in South Korea on diurnal to decadal timescales for theperiod of 2001-2021. Atmospheric Chemistry and Physics, 23, 12867-12886. https://doi.org/10.5194/acp-23-12867-2023
  13. Kim, S.-Y., Kim,E., and Kim, W.J., 2020. Health effects of ozone on respiratory diseases. Tuberculosis and Respiratory Diseases, 83(Supple 1), S6–S11. https://doi.org/10.4046/trd.2020.0154
  14. Kleinman, L.I., 1994. Low and high NOx tropospheric photochemistry. Journal of Geophysical Research: Atmospheres, 99(D8), 16831–16838. https://doi.org/10.1029/94JD01028
  15. Lamsal, L. N., Krotkov, N. A., Celarier, E. A., Swartz, W. H., Pickering, K.E., Bucsela, E.J., et al., 2014. Evaluation of OMI operational standard NO2 column retrievals using in situ and surface-based NO2 observations. Atmospheric Chemistry andPhysics, 14(21), 11587-11609. https://doi.org/10.5194/acp-14-11587-2014
  16. Lee, H.-J., Chang, L.-S., Jaffe, D.A., Bak, J., Liu,X., Gonzalez Abad, G., et al., 2021. Ozone continues to increase in East Asia despite decreasing NO2: Causes and abatements. Remote Sensing, 13(11), 2177. https://doi.org/103390/rs13112177 103390/rs13112177
  17. Li, Y., Lau, A.K. H., Fung,J.C.H., Zheng, J., and Liu, S., 2013. Importance of NOx control for peak ozone reduction in the Pearl River Delta region. Journal of Geophysical Research: Atmospheres, 118(16), 9428-9443. https://doi.org/10.1002/jgrd.50659
  18. Liu, C., and Shi, K.,2021. A review on methodology in O3-NOx-VOC sensiivity study. Environinental Pollution, 291,118249. https://doi.org/101016/j.envpol.2021.118249 101016/j.envpol.2021.118249
  19. Malley, C.S., Henze, D.K., Kuylenstierna, J.C.I., Vallack, H. W., Davila, Y., Anenberg, S. C., et al., 2017. Updated global estimates of respiratory mortality in adults≥30 years of age attributable to long-term ozone exposure. Environmental Health Perspectives, 125(8), 087021.
  20. Martin, R. V., Fiore, A. M., and Van Donkelaar, A., 2004. Space-based diagnosis of surface ozone sensitivity to anthropogenic emissions. Geophysical Research Letters, 31(6), L06120. https://doi.org/10.1029/2004GL019416
  21. Martin, R. V, Jacob, D. J., Chance, K., Kurosu, T. P., Palmer, P. I., and Evans, M. J., 2003. Global inventory of nitrogen oxide emissions constrained by space-based observations of NO2 columns. Journal of Geophysical Research: Atmospheres, 108(D17), 4537. https://doi.org/10.1029/2003jd003453
  22. Metropolitan Air Quality Management Office, 2024. The 2nd metropolitan air quality management master plan (2025-2029). Ministry of Climate, Energy and Environment.
  23. Oak, Y. J., Jacob, D. J., Pendergrass, D. C., Dang, R., Colombi, N. K., Chong, H., et al., 2025. Air quality trends and regimes in South Korea inferred from 2015–2023 surface and satellite observations. Atmospheric Chemistry and Physics, 25(5), 3233–3252. https://doi.org/10.5194/acp-25-3233-2025
  24. Ren, J., Guo, F., and Xie, S., 2022. Diagnosing ozone-NOx-VOC sensitivity and revealing causes of ozone increases in China based on 2013–2021 satellite retrievals.Atmospheric Chemistry and Physics, 22(22), 15035–15047. https://doi.org/10.5194/acp-22-15035-2022
  25. Shen, L., Jacob, D. J., Zhu, L., Zhang, Q., Zheng, B., Sulprizio, M. P., et al., 2019. The 2005–2016 trends of formaldehyde columns over China observed by satellites: Increasing anthropogenic emissions of volatile organic compounds and decreasing agricultural fire emissions. Geophysical Research Letters, 46(8), 4468–4475. https://doi.org/10.1029/2019GL082172
  26. Sillman, S., 1999. The relation between ozone, NOx and hydrocarbons in urban and polluted rural environments. Atmospheric Environment, 33(12), 1821–1845. https://doi.org/10.1016/S1352-2310(98)00345-8
  27. Turner, M. C., Jerrett, M., Pope III, C. A., Krewski, D., Gapstur, S.M.,Diver, W. R., et al., 2016. Long-term ozone exposure and mortality in a large prospective study. American Journal of Respiratory and Critical Care Medicine, 193(10), 1134-1142. https://doi.org/10.1164/ICCm.201508-1633OC
  28. Veefkind, J. P., Aben, I., McMullan, K., Förster, H., De Vries, J., Otter, G., et al., 2012. TROPOMI on the ESA Sentinel-5 Precursor: A GMES mission for global observations of the atmospheric composition for climate, air quality and ozone layer applications. Remote Sensing of Environment, 120, 70–83. https://doi.org/10.1016/j.rse.2011.09.027
  29. Wang, T., Xue, L., Brimblecombe, P., Lam, Y. F., Li, L., and Zhang, L., 2017. Ozone pollution in China: A review of concentrations, meteorological influences, chemical precursors, and effects. Science of The Total Environment, 575, 1582–1596. https://doi.org/10.1016/j.scitotenv.2016.10.081
  30. Wang, W., van der A,. Ding, J., van Weele, M., and Cheng, T., 2021. Spatial and temporal changes of the ozone sensitivity in China based on satellite and ground-based observations. Atmospheric Chemistry and Physics, 21(9), 7253-7269. https://doi.org/10.5194/acp-21-7253-2021
  31. Xie, Y., Dai, H., Zhang, Y., Wu, Y., Hanaoka, T., and Masui, T., 2019. Comparison of health and economic impacts of PM2.5 and ozone pollution in China. Environment International, 130, 104881. https://doi.org/10.1016/j.envint.2019.05.075
  32. Yeo, M. J., and Kim, Y. P., 2021. Long-term trends of surface ozone in Korea. Journal of Cleaner Production, 294, 125352. https://doi.org/10.1016/j.jclepro.2020.125352
  33. Zhu, L., Jacob, D. J., Mickley, L. J., Marais, E. A., Cohan, D. S., Yoshida, Y., et al., 2014. Anthropogenic emissions of highly reactive volatile organic compounds in eastern Texas inferred from oversampling of satellite (OMI) measurements of HCHO columns. Environmental Research Letters, 9(11), 114004. https://doi.org/10.1088/1748-9326/9/11/114004