Korean Journal of Remote Sensing (대한원격탐사학회지)

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

DOI QR Code

Assessments of the GEMS NO2 Products Using Ground-Based Pandora and In-Situ Instruments over Busan, South Korea

  • Serin Kim (Major of Spatial Information Engineering, Division of Earth and Environmental System Sciences, Pukyong National University) ;
  • Ukkyo Jeong (Major of Spatial Information Engineering, Division of Earth and Environmental System Sciences, Pukyong National University) ;
  • Hanlim Lee (Major of Spatial Information Engineering, Division of Earth and Environmental System Sciences, Pukyong National University) ;
  • Yeonjin Jung (Major of Spatial Information Engineering, Division of Earth and Environmental System Sciences, Pukyong National University) ;
  • Jae Hwan Kim (Department of Atmospheric Science, Pusan National University)
  • Received : 2024.01.03
  • Accepted : 2024.01.30
  • Published : 2024.02.28
Download PDF
⟨ Previous Next ⟩

Abstract

Busan is the 6th largest port city in the world, where nitrogen dioxide (NO2) emissions from transportation and port industries are significant. This study aims to assess the NO2 products of the Geostationary Environment Monitoring Spectrometer (GEMS) over Busan using ground-based instruments (i.e., surface in-situ network and Pandora). The GEMS vertical column densities of NO2 showed reasonable consistency in the spatiotemporal variations, comparable to the previous studies. The GEMS data showed a consistent seasonal trend of NO2 with the Korea Ministry of Environment network and Pandora in 2022, which is higher in winter and lower in summer. These agreements prove the capability of the GEMS data to monitor the air quality in Busan. The correlation coefficient and the mean bias error between the GEMS and Pandora NO2 over Busan in 2022 were 0.53 and 0.023 DU, respectively. The GEMS NO2 data were also positively correlated with the ground-based in-situ network with a correlation coefficient of 0.42. However, due to the significant spatiotemporal variabilities of the NO2, the GEMS footprint size can hardly resolve small-scale variabilities such as the emissions from the road and point sources. In addition, relative biases of the GEMS NO2 retrievals to the Pandora data showed seasonal variabilities, which is attributable to the air mass factor estimation of the GEMS. Further studies with more measurement locations for longer periods of data can better contribute to assessing the GEMS NO2 data. Reliable GEMS data can further help us understand the Asian air quality with the diurnal variabilities.

Keywords

1. Introduction

Airpollutionis knownto be associatedwithsignificantproportions of cardiopulmonary mortalities, lung cancer, and asthma (WHO, 2013; United States Environmental ProtectionAgency, 2024), and tropospheric nitrogen dioxide (NO2) is one of the significant species of health issues (Chen et al., 2012). NO2 also plays a key role in atmospheric chemistry asit is one of the major precursors of aerosols, tropospheric ozone, and hydroxyl (OH) radicals (Squizzato et al., 2013).Due to the shortlifetime in the troposphere (typically less than a few hours), spatial variabilities of NO2 near sources are often distinguishable from high-resolution satellite instruments (Liu et al., 2016). Networked ground-based in-situ instruments have provided reliable air quality data over long periods. However,such networks are often sparse in remote areas and developing countries, and limited spatial information can be provided (Jeong and Hong, 2021a).

Satellite-based instruments have successfully complemented such limitations of the ground-based air quality monitoring networks by providing information on aerosols and trace gases (Burrows et al., 1999; Levelt et al., 2006). Since October 2017, the TROPOspheric Monitoring Instrument (TROPOMI) onboard the Sentinel-5 Precursor of the European Space Agency (ESA) has provided very high resolution of NO2 retrievals (5.5 × 3.5 km2)that canmonitorsmallerscaleNO2 emissions and dispersions compared to the previous sensors (Boersma et al., 2018). More recently, theGeostationaryEnvironmentMonitoring Spectrometer (GEMS) has provided unprecedented hourly variations of atmospheric pollutants(e.g., aerosols, NO2, SO2, O3, and HCHO) over Asia since 2020. Various studies have utilized satellite data to derive information on surface air quality, including NO2 and CO (Jeong et al., 2021a; 2021b, and references therein).

Busan is the largest port city in South Korea and 6th in the world, with a population of over 3.4 million, where NO2 emissions from road vehicles and ships are significant (Statistics Korea, 2024). Using in-situ records, Yoo and Park (2004) analyzed O3 and NO2 concentrations during a high ozone period from 1998 to 1999. They concluded that air quality in Busan is affected by local emissions as well as meteorological conditions. However, to our knowledge, comprehensive analysis of air quality in Busan using satellite and ground-based instruments is still insufficient. This study utilized multiple platforms, including the GEMS, surface in-situ network, and ground-based remote sensing instruments, to better understand their merit and limitations. Air quality data from satellite and ground-based instruments can complement each other and,therefore, contribute to managing the air quality in Busan.

2. Materials and Methods

2.1. GEMS

The Geostationary Environment Monitoring Spectrometer (GEMS) onboard the GeostationaryKoreaMulti-Purpose Satellite (GEO-KOMPSAT-2B) mission was launched by the National Institute of Environmental Research (NIER) of South Korea in February 2020. The GEMS hourly observes Asia (5°S–45°N in latitude and 75–145°E in longitude) during day time with about 3.5 × 8 km2 of spatial (Kim et al., 2020). The GEMS retrieves aerosols and trace gases(e.g., total and tropospheric ozone, NO2, SO2, HCHO, andCHOCHO)from the UV-visible (300–500 nm) spectral radiances with a spectral resolution of about 0.6 nm (Kim et al., 2020). In the first step, the GEMS NO2 operational algorithm utilizes the Differential Optical Absorption Spectroscopy (DOAS) technique with a 432–450 nm spectral window to calculate slant column density (SCD). In the next step, the NO2 SCD values are converted into total NO2 vertical column density (VCD) based on the air massfactor(AMF) calculations using the vectorlinearized discrete ordinate radiative transfer(VLIDORT) model (Spurr and Christi, 2014) version 2.6 (Park et al., 2020). The GEMS NO2 VCD accuracy is estimated to be 1 × 1015 molecules cm–2. The GEMS NO2 VCD also showed reasonable consistency with the ozone monitoring instrument (OMI) (the correlation coefficient and root-mean-squared-error [RMSE] were 0.87 and 2.66 × 1015 molecules cm–2, respectively; Kim et al., 2020). In this study, we utilized version 2.0, level 2 products of the GEMS NO2 in 2022. For a consistent comparison with Pandora, we analyzed the total column VCD retrievals in this study.

2.2. Pandora

Pandora is a ground-basedUV-visible spectrometerthat measures direct sunlight, moon, and sky radiances over the 280–525 nm wavelength range with a full-width-half-maximum (FWHM) of about 0.6 nm. Pandora providestrace gas column amounts(e.g., O3, NO2, SO2, HCHO) using spectral fitting based on the BeerBouguer-Lambert(B-B-L)law(Cede, 2021).Herman et al.(2009) estimated that NO2 VCD from the direct-sun measurement of Pandora has high precision with a random uncertainty of 0.01 DU and a nominal accuracy of about 0.1 DU. The Pandora measurement data in the global area are available from the Pandonia Global Network (PGN; https://www.pandonia-globalnetwork.org/) and have been adopted in many satellite comparison and validation studies (Herman et al., 2009, 2018; Choi et al., 2020; Judd et al., 2020; Verhoelst et al., 2021; Kim et al., 2023). The L2 data measured in Busan (35.24°N, 129.08°E) during 2022 was used in this study with quality flags of 0 and 10 (good data quality).

2.3. KME Network

The Korea Ministry of Environment (KME) has measured standard air pollutant species (i.e., NO2, SO2, CO, O3, and particulate matter), and the supersite additionally measures ion and metal species based on various methods. Urban ambient monitoring stations of the KME network are typically located on the rooftops of public buildings (fewer than five stories), whereas roadside monitoring stations are located near major roads about 2.5 m above the ground. In 2022, 24 urban ambient monitoring stations and four roadside monitoring stations provided NO2 surface mixing ratios over Busan (National Institute of Environmental Research, 2022).

At each station, oxides of nitrogen analyzers (model 2108, Dasibi Environmental Corp.; United States Environmental ProtectionAgency reference method RFNA-1192-089) measure the NO2 mixing ratio based on the chemiluminescence method with a lower detection limit of 2 ppb. The KME inspects all the instruments monthly following a two-step quality assurance process; abnormal data are screened in the first step based on information on the instrument (e.g., calibration, inspection, or malfunction). In the second step, the KME screens outliers that exceed the normal range or rate of change (National Institute of Environmental Research, 2022;Korea Ministry of Environment, 2024). For the distribution version of the data, five minutes of temporal resolution samples are averaged hourly and then uploaded to the Air Korea homepage (https://www.airkorea.or. kr/web/last_amb_hour_data). In this study, we utilized all the KME measurement data in Busan (i.e., 24 urban ambient monitoring stations and four roadside monitoring stations).

3. Results

3.1. Spatiotemporal Variations

Fig. 1 shows the annual mean values of NO2 in 2022 measured from the GEMS, binned to a comparable resolution of the GEMS (0.1° × 0.1° of the horizontal grid); panel (a) presents those over the Korean peninsula, and (b) shows over Busan. In panel (b), surface NO2 mixing ratiosfrom theKME stations are also shown, where the circles indicate ambient urban monitoring stations, and the squares denote the roadside monitoring stations. The red starin Fig. 1(b)indicatesthe location of the Pandora instrument.

Fig. 1(a) showed a similar spatial distribution of the NO2 with the TROPOMI reported by Jeong et al. (2020), where the highest burden of NO2 was observed over the SeoulMetropolitan areas. Busan, the second-largest city in South Korea, also has high NOx emissions from vehicles, ships, and industrial complexes (Statistics Korea, 2024), resulting in large amounts of NO2, as shown in Figs. 1(a, b). In Busan, higher NO2 values were observed in the western part of the city by both the GEMS and KME stations, where industrial complexes are distributed. The KME stations are densely distributed within the GEMS pixels over Busan (Fig. 1b). In general, the GEMS and KME annual mean values showed good agreement as compared to Fig. 1(b). However, the KME measurements showed significant spatial variabilities even within each GEMS pixel; surface NO2 mixing ratio measurements from the roadside monitoring stations showed significantly higher values than the urban ambient monitoring stations. The high values near the roadside stations are attributable to the dense transportation in Busan, and large spatial variabilities are due to the short lifetime of NO2 in the atmosphere (i.e., a few hours, depending on the condition).

OGCSBN_2024_v40n1_1_f0001.png 이미지

Fig. 1. Annual mean NO2 vertical column densities (VCD) from the Geostationary Environment Monitoring Spectrometer (GEMS) in 2022 over (a) South Korea and (b) Busan. Colors in circles and squares present surface in-situ measurements of the Korean Ministry Environment (KME) groundbased network in 2022, where the circle depicts ambient urban monitoring stations, and the square denotes roadside monitoring stations. The red star in panel (b) indicates the location of the Pandora instrument.

Fig. 2 shows the temporal variation of the NO2 over Busan in 2022, where faint colors show all the samples and thicker circles with lines present monthly mean values. Fig. 2(a) demonstrates the KME mean surface mixing ratio (i.e., the hourly mean value of the 28 KME stations in Fig. 1(b), and panel (b) shows the vertical column density of NO2 measured by Pandora (green) and the GEMS (red). NO2 retrievals from the GEMS were collected with longitudes from 128.8 to 129.3°E and latitudes from 35.0 to 35.3°N, which covers most of Busan city (Fig. 1b), to calculate the red data in Fig 2(b). A quality flag of 10 (nearreal-time, good quality) of the Pandora retrievalswas utilized for the analysis as reprocessing data has not been applied yet.

The KME, Pandora, and GEMS data showed generally consistent seasonal variabilities as shown in Fig. 2, which are higher in winter and lower in summer. Error bars in Fig. 2(a) show the monthly standard deviation, which is higher in winter and lower in summer. It can be attributed to a longer lifetime of NO2 during thewinter; a larger burden of NO2 over Busan results in higher spatiotemporal variabilities compared to the summer. Jeong and Hong (2021a) also reported a similar trend in the Korean peninsula when they utilized the TROPOMI data, and the correlation coefficient of monthly mean values between the TROPOMI and KME data was about 0.73. Note that the KME measuressurface concentrations of the NO2 and the TROPOMI provides column-integrated burden of NO2, which can differ due to their vertical distributions. The correlation coefficient between the monthly mean values of the KME measurements and Pandora retrievals was 0.82, and that between the KME and GEMS was 0.76, comparable to or higher than that using the TROPOMI. Note that Jeong and Hong (2021a, 2021b) calculated all the KME stations in South Korea, and we utilized data from Busan in this study. Notably, the Pandora data showed higher consistency with the KME data, although their measurement parameter is not identical; KME measures surface mixing ratio, and Pandora measures VCD like the GEMS. The results demonstrate the capability of the satellite NO2 retrievals to monitorsurface air quality over Busan. The differences between the Pandora and GEMS NO2 VCD are attributable to the errors in the AMF calculations in the GEMS algorithm and different measurement geometries. More discussions are in the following Section 3.2.

OGCSBN_2024_v40n1_1_f0002.png 이미지

Fig. 2. Temporal variations of the (a) surface mixing ratio and (b) vertical column density (VCD) of NO2 in Busan in 2022. Faint colors show all the samples, and the symbol with a line denotes monthly mean values. Error bars in panel (a) depict the standard deviation in each month.VMR:volume mixing ratio.

3.2. Comparison between the GEMS, Pandora, and KME Measurements

Fig. 3 compares the NO2 VCDs from the GEMS and Pandora in Busan in 2022. The GEMS data were spatially sampled by collecting retrievalswithin ±0.05°from the Pandora site, and the Pandora datawere temporally averagedwithin ±30 minutesfrom the GEMS scan time. Pandora retrievals are often used to validate satellite data (Jeong et al., 2018; 2022; Kim et al., 2023). Kim et al. (2023) compared the initial GEMS data (version 1.0) with the Pandora during the 2020 GEMS Map of Air Pollution (GMAP) campaign for about three months (from November 2020 to January 2021) in Seosan, in the middle of the Korean peninsula. The correlation coefficients ranged from 0.69 to 0.81, and the mean-bias-error(MBE)ranged from –0.43 × 1016 to –0.16 × 1016 molecules cm–2 during the campaign (Kim et al., 2023). The correlation between the GEMS and Pandora NO2 VCD (r=0.53) was slightly lowerthan the previous study for a year-long record in Busan, asshown in Fig. 3.The GEMS retrievalsshowed slightly higher values compared to the Pandora products(MBE of about 0.06 × 1016 molecules cm–2), whereas Kim et al. (2023) reported negative biases. It can be attributed to the different locations, times, and versions of the GEMS products. Longer periods of comparison in thisstudy also caused different comparison results as the biases between the GEMS and Pandora showed seasonal variations (Fig. 2b). The GEMS NO2 VCD also showed higher values compared to the Pandora in winter, which is a consistent result with the Kim et al. (2023), but lower values in summer as shown in Fig. 2(b) and Fig. 3. The seasonality of the biases can be attributed to the air massfactor calculationswith atmospheric profiles and surface albedo, which needs to be analyzed using multiple locations and longer periods of data.

OGCSBN_2024_v40n1_1_f0003.png 이미지

Fig. 3. Comparison of NO2 vertical column densities (VCDs) from ground-based Pandora instrument (x-axis) and the GEMS (y-axis). The dotted line shows a one-to-one line, and the solid line shows the regression line. Different colors denote data during different seasons.

Estimating the surface mixing ratio of NO2 using the satellite product is important to estimate long-term exposure to the public, particularly in remote areas or developing countries (Jeong and Hong, 2021a; 2021b). Fig. 4 shows the relationship between the NO2 surface mixing ratio measured from the KME network and VCDsfrom the GEMS and Pandora; panel(a) used the Pandora data, and panel(b) employed the GEMS data.Jeong and Hong (2021a, 2021b) compared the KME NO2 and CO measurements with the TROPOMI VCD data in 2019 and showed a good correlation (r=0.67). The GEMS and Pandora NO2 products also showed a moderate correlation with the KME measurements (r=0.49 and 0.42 for Pandora and GEMS, respectively) as shown in Fig. 4. For the comparison, the GEMS andKME datawithin ±0.05°from the Pandora sitewere collected for consistency.

Retrieval of NO2 VCD from direct sun measurements of the Pandora is simple and straightforward as it utilizes geometric AMF to convert SCDsto VCDs. In contrast, AMF calculation of the GEMS requires atmospheric profiles(temperature, pressure, aerosols, trace gases, and clouds)from chemical transport models or other retrievals, which results in additional uncertainties. Surface reflectance is also one of the majorsources of errorin the GEMS NO2 products (Lin et al., 2015). The differences in AMF calculations ofthe Pandora and GEMS algorithms are attributable to the differences in Figs. 4(a, b).

Temporal variationsin the ratio of NO2 surface concentrations to VCDs are shown in Fig. 5. Due to the shallow mixing layer height by lower temperatures in winter, NO2 burden is more likely distributed near the surface, resulting in a higher ratio in Fig. 5. The ratios of the KME to the Pandora measurements showed such seasonal tendency, which are higher in winter and lower in summer, whereas those of the KME to the GEMS measurements showed opposite seasonal trend. As discussed above, the AMF calculations for satellite data are affected by various parameters, including meteorology, vertical profiles of trace gases and aerosols, clouds, and surface reflectance. Further studies to analyze the effects of each parameter are required to assess the accuracy of the GEMS data to provide reliable information for various applications.

OGCSBN_2024_v40n1_1_f0004.png 이미지

Fig. 4. Comparison of NO2 vertical column densities (VCDs) and surface mixing ratio measurements from the Korea Ministry of Environment (KME) stations. Panel (a)shows a comparison using the ground-based Pandora data,whereas panel (b) utilizesthe satellite-based Geostationary Environment Monitoring Spectrometer (GEMS) data. The solid lines in both panels show the regression lines.

OGCSBN_2024_v40n1_1_f0005.png 이미지

Fig. 5. Temporal variations of the ratio of NO2 surface mixing ratio to the VCD of NO2 in Busan in 2022. Faint colors show all the samples, and the symbol with a line denotes monthly mean values.

4. Discussion and Conclusions

Thisstudy assessed the GEMSNO2 data over Busan using various platforms, including surface in-situ network and ground-based remote sensing instruments (KME network and Pandora). Considering the differences in surface concentration (KME network) and vertical column amount measurements (Pandora and GEMS), their spatiotemporal variability showed reasonable consistency over Busan, meaning that the GEMS data can complement ground-based air quality monitoring networks over Asia. However, due to the short lifetime of NO2, the GEMS footprint size cannot resolve small-scale variabilities such as the emissions from the road and point sources. Airborne-based measurement can complementsuch limitationsin future studies (Choo et al., 2023). NO2 data from the KME network, GEMS, and Pandora showed consistent seasonal trends in 2022, which are higher in winter and lower in summer. However, the GEMS NO2 product showed seasonal biases compared to the Pandora products,which canbe attributed to theGEMS’sAMF calculations. Future studies using more measurementsitesfor more extended periods of data can help understand the GEMS’s uncertainty, thereby improving the algorithm. Improvements in the GEMS algorithm can further contribute to understanding air quality over Asia by providing reliable diurnal cycles.

Acknowledgments

This work was supported by a Research Grant of Pukyong National University (2022). We also thank the Pandonia Global Network for processing and providing the NO2 data and the National Institute of Environmental Research for providing the GEMS and KME network data.

Conflict of Interest

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

References

  1. Boersma, K. F., Eskes, H. J., Richter, A., De Smedt, I., Lorente, A., Beirle, S. et al., 2018. Improving algorithms and uncertainty estimates for satellite NO2 retrievals: results from the quality assurance for the essential climate variables (QA4ECV) project. Atmospheric Measurement Technique, 11(12), 6651-6678. https://doi.org/10.5194/amt-11-6651-2018
  2. Burrows, J. P., Weber, M., Buchwitz, M., Rozanov, V., Ladstatter Weissenmayer, A., Richter, A. et al., 1999. The global ozone monitoring experiment(GOME): Mission concept and first scientific results. Journal of the Atmospheric Sciences, 56(2), 151-175. https://doi.org/10.1175/1520-0469(1999)056<0151:TGOMEG>2.0.CO;2
  3. Cede A., 2021. Manual for Blick Software Suite 1.8. Available online: https://www.pandonia-global-network.org/wpcontent/uploads/2021/09/BlickSoftwareSuite_Manual_v1-8-4.pdf (accessed on Jan. 3, 2024).
  4. Chen, R., Samoli, E., Wong, C. M., Huang, W., Wang, Z., Chen, B., and Kan, H., 2012. Associations between short-term exposure to nitrogen dioxide and mortality in 17Chinese cities: The China Air Pollution and Health Effects Study (CAPES). Environment International, 45, 32-38. https://doi.org/10.1016/j.envint.2012.04.008
  5. Choi, S., Lamsal, L. N., Follette-Cook, M., Joiner, J., Krotkov, N. A., Swartz, W. H. et al., 2020. Assessment of NO2 observations during DISCOVER-AQ and KORUS-AQ field campaigns. Atmospheric Measurement Technique, 13(5), 2523-2546. https://doi.org/10.5194/amt-13-2523-2020
  6. Choo, G.-H., Lee, K., Hong, H., Jeong, U., Choi, W., and Janz, S., 2023. Highly resolved mapping of NO22 vertical column densities from GeoTASO measurements over a megacity and industrial area during the KORUS-AQ campaign. Atmospheric Measurement Technique, 16(3), 625-644. https://doi.org/10.5194/amt-16-625-2023
  7. Herman, J., Cede, A., Spinei, E., Mount, G., Tzortziou, M., and Abuhassan,N., 2009. NO2 column amounts from ground based Pandora and MFDOAS spectrometers using the direct-sun DOAS technique: Intercomparisons and application to OMI validation. Journal of Geophysical Research, 114(D13). https://doi.org/10.1029/2009JD011848
  8. Herman, J., Spinei, E., Fried, A., Kim, J., Kim, J., Kim, W. et al., 2018. NO2 and HCHO measurements in Korea from 2012 to 2016 from Pandora spectrometer instruments compared with OMI retrievals and with aircraft measurements during the KORUS-AQ campaign. Atmospheric Measurement Technique, 11(8), 4583-4603. https://doi.org/10.5194/amt-11-4583-2018
  9. Jeong, U., and Hong, H., 2021a. Assessment of tropospheric concentrations of NO2 from the TROPOMI/Sentinel-5 Precursor for the estimation of long-term exposure to surface NO2 over South Korea. Remote Sensing, 13(1), 1877. https://doi.org/10.3390/rs13101877
  10. Jeong, U., and Hong, H., 2021b. Comparison of total column and surface mixing ratio of carbon monoxide derived from the TROPOMI/Sentinel-5 Precursor with in-situ measurements from extensive ground-based network over South Korea. Remote Sensing, 13(19), 3987. https://doi.org/10.3390/rs13193987
  11. Jeong, U., Tsay, S.-C., Hsu, N. C., Gile, D. M., Cooper, J. W., Lee, J. et al., 2022. Simultaneous retrievals of biomass burning aerosols and trace gases from the ultraviolet to nearinfrared over northern Thailand during the 2019 premonsoon season. Atmospheric Chemistry and Physics, 22(8), 11957-11986.https://doi.org/10.5194/acp-22-11957-2022
  12. Jeong, U., Tsay, S.-C., Pantina, P., Butler, J. J., Loftus, A. M., Abuhassan, N. et al., 2018. Langley calibration analysis of solar spectroradiometric measurements: Spectral aerosol optical thickness retrievals. Journal of Geophysical Research, 123(8), 4221-4238. https://doi.org/10.1002/2017JD028262
  13. Judd, L. M., Al-Saadi, J. A., Szykman, J. J., Valin, L. C., Janz, S. J., Kowalewski, M. G. et al., 2020. Evaluating Sentinel-5P TROPOMI tropospheric NO2 column densities with airborne and Pandora spectrometers near New York City and Long Island Sound. Atmospheric Measurement Technique, 13(11), 6113-6140. https://doi.org/10.5194/amt-13-6113-2020
  14. Kim, J., Jeong, U., Ahn, M.-H., Kim, J. H., Park, R. J., Lee, H. et al., 2020. New era of air quality monitoring from space, Geostationary Environment Monitoring Spectrometer (GEMS). Bulletin of the American Meteorological Society, 101(1), E1-E22. https://doi.org/10.1175/BAMS-D-18-0013.1
  15. Kim, S., Kim, D., Hong, H., Chang, L.-S., Lee, H., Kim, D.-R. et al., 2023. First-time comparison between NO2 vertical columns from Geostationary Environment Monitoring Spectrometer (GEMS) and Pandora measurements. Atmospheric Measurement Technique, 16(16), 3959-3972. https://doi.org/10.5194/amt-16-3959-2023
  16. Korea Ministry of Environment, 2024. Air Korea. Available online: https://www.airkorea.or.kr/web/last_amb_hour_data (accessed on Jan. 1, 2024).
  17. Levelt, P. F., van den Oord, G. H. J., Dobber, M. R., Malkki, A., Visser, H., de Vries, J. et al., 2006. The ozone monitoring instrument.IEEE Transactions onGeoscience and Remote Sensing, 44(5), 1093-1101. https://doi.org/10.1109/TGRS.2006.872333
  18. Lin, J.-T., Liu, M.-Y., Xin, J.-Y., Boersma, K. F., Spurr, R., and Zhang, Q., 2015. Influence of aerosols and surface reflectance on satellite NO2 retrieval: Seasonal and spatial characteristics and implications for NOx emission constraints. Atmospheric Chemistry and Physics, 15(19), 11217-11241. https://doi.org/10.5194/acp-15-11217-2015
  19. Liu, F., Beirle, S., Zhang, Q., Dorner, S., He, K., and Wagner, T., 2016. NOx lifetimes and emissions of cities and power plants in polluted background estimated by satellite observations. Atmospheric Chemistry and Physics, 16(8), 5283-5298. https://doi.org/10.5194/acp-16-5283-2016
  20. National Institute of Environmental Research, 2022.Annualreport of air quality in Korea (NIER-GP2023-046). Ministry of Environment.
  21. Park, J., Lee, H., and Hong, H., 2020. Geostationary Environment Monitoring Spectrometer(GEMS) algorithm theoretical basis document NO2 retrieval algorithm. Available online: https://nesc.nier.go.kr/ko/html/satellite/doc/doc.do (accessed on Jan. 3, 2023).
  22. Statistics Korea, 2024. Korean Statistical Information Service. Available online: https://kosis.kr/index/index.do (accessed on Jan. 1, 2024).
  23. Spurr, R., and Christi, M., 2014. On the generation of atmospheric property Jacobians from the (V)LIDORT linearized radiative transfer models.Journal of Quantitative Spectroscopy and Radiative Transfer, 142, 109-115. https://doi.org/10.1016/j.jqsrt.2014.03.011
  24. Squizzato, S., Masiol, M., Brunelli, A., Pistollato, S., Tarabotti, E., Rampazzo, G., and Pavoni, B., 2013. Factors determining the formation of secondary inorganic aerosol: A case study in the Po Valley (Italy).Atmospheric Chemistry and Physics, 13(4), 1927-1939. https://doi.org/10.5194/acp13-1927-2013
  25. United States Environmental Protection Agency, 2024. Nitrogen dioxide (NO2) pollution. Available online: https://www.epa.gov/NO2-pollution (accessed on Jan. 3, 2024).
  26. Verhoelst, T., Compernolle, S., Pinardi, G., Lambert, J.-C., Eskes, H. J., Eichmann, K.-U. et al., 2021. Ground-based validation of the Copernicus Sentinel-5P TROPOMI NO2 measurements with the NDACCZSL-DOAS, MAXDOAS and Pandonia global networks. Atmospheric Measurement Technique, 14(1), 481-510. https://doi.org/10.5194/amt-14-481-2021
  27. WHO, 2013. Review of evidence on health aspects of air pollution - REVIHAAP project (Technical report). WHO Regional Office for Europe. https://iris.who.int/bitstream/handle/10665/341712/WHO-EURO-2013-4101-43860-61757-eng.pdf?sequence=1
  28. Yoo, E.-C., and Park, O.-H., 2004. A study on allocation of air pollution monitoring network by spatial distribution analysis of ozone and nitrogen dioxide concentrations in Busan. Korean Journal of Chemical Engineering, 27, 494-503. https://doi.org/10.1007/s11814-010-0076-z