Juseon Bak

Satellite remote sensing of ozone has progressed from daily global observations by polar‐orbiting platforms to hourly regional monitoring by geostationary platforms. However, retrieving detailed vertical profiles remains challenging because satellite measurements have inherently limited vertical sensitivity. Within the optimal estimation (OE) framework, a priori information constrains the retrieval, preventing unphysical deviations driven by measurement uncertainties while supplementing the limited information content of the observations. In this study, we evaluate ozone profiles from the Goddard Earth Observing System Composition Forecasting (GEOS‐CF) system as an alternative a priori to the tropopause‐based (TB) ozone profile climatology. To ensure a robust assessment, retrieval experiments are performed across multiple satellite platforms, including OMI, TROPOMI, and GEMS. The GEOS‐CF ozone analysis systematically exhibits a systematic positive bias in the troposphere relative to ozonesonde soundings, but agreeing with limb observations within a few percent throughout most of the stratosphere, particularly between 50 and 1 hPa. Larger discrepancies are found in the tropopause region (>100 hPa) and in the upper atmosphere (<1 hPa). These characteristics are reflected in the satellite retrievals: the positive tropospheric bias in the GEOS‐CF a priori propagates into the retrieved ozone profiles, whereas the improved representation of the stratosphere in GEOS‐CF reduces the retrieval biases associated with the TB a priori. These results support a hybrid a priori strategy that combines the tropospheric stability of the TB climatology with the improved stratospheric representation of GEOS‐CF.

This study developed the GEMS Level‐3 GT‐O3T (GEMS TROPOMI‐referenced Total Column Ozone) algorithm to construct a consistent long‐term ozone dataset by correcting GEMS biases using TROPOMI as a reference. The algorithm generates daily 0.25° gridded TROPOMI and GEMS total ozone datasets and applies monthly ratio‐based correction factors to minimise spatiotemporal discrepancies. Compared with TROPOMI, the RMSE decreased from 14.29 DU to 3.83 DU in 2023 and from 15.63 DU to 4.35 DU in 2024. The systematic underestimation observed in the original GEMS product, particularly over mid‐ and high‐latitude regions and across seasonal cycles, was significantly mitigated. Validation with Pandora measurements further demonstrated the effectiveness of the correction, with the mean bias error (MBE) decreasing from 12.21 DU to −2.32 DU. OMI, TROPOMI, and corrected GEMS data based on the GT‐O3T algorithm were then combined to create a merged record covering 2005–2024. The merged dataset closely reproduced Dobson observations with enhanced temporal continuity and reduced regional bias, particularly after 2023 when GEMS data were included.

Ozone (O ₃ ) over South Korea has risen in recent years, underscoring the need to accurately quantify emissions of nitrogen oxides (NO _x ) and volatile organic compounds (VOC). We develop a hybrid inverse modeling framework that couples the Finite Difference Mass Balance (FDMB) method with four-dimensional variational data assimilation (4D-Var) using the Community Multiscale Air Quality (CMAQ) model to jointly constrain spatiotemporal NO _x and VOC emissions over South Korea. The inversion is constrained by Tropospheric Monitoring Instrument (TROPOMI) NO ₂ and HCHO columns and by surface NO ₂ and O ₃ concentrations from the Air Quality Monitoring Station (AQMS) network. The analysis covers 1–14 May 2022, during a month that exhibited the highest mean O ₃ over the past decade. Optimized NO _x emissions exhibit strong diurnal adjustments relative to the prior (nighttime reductions up to 51 % and daytime increases up to 14 %). The joint NO _x –VOC inversion produced the best consistency with assimilated AQMS O ₃ observations (Index of Agreement > 0.8). Optimized emissions shift O ₃ sensitivity from VOC-sensitive to NO _x -sensitive across much of the domain, improving spatial consistency with TROPOMI-derived formaldehyde-to-NO ₂ ratio diagnostics. Adjoint-based hourly ΔO ₃ responses reveal distinct temporal characteristics: O ₃ titration by NO _x is immediate, whereas photochemical O ₃ production by VOCs requires a 1–2 h reaction time. Furthermore, reactive biogenic VOCs contribute to a slight nighttime O ₃ sink under NO _x -sensitive conditions. These findings motivate hour-specific, regime-specific controls rather than uniform daily reductions. Overall, the hybrid framework improves O ₃ simulations and sensitivity-regime diagnosis, enabling spatiotemporally resolved precursor emission reduction guidance for effective O ₃ mitigation.

Ozone (O₃) over South Korea has risen in recent years, underscoring the need to accurately quantify emissions of nitrogen oxides (NO_x) and volatile organic compounds (VOC). We develop a hybrid inverse modeling framework that couples the Finite Difference Mass Balance (FDMB) method with four-dimensional variational (4D-Var) assimilation using the Community Multiscale Air Quality (CMAQ) model to jointly constrain spatiotemporal NO_x and VOC emissions. The inversion is constrained by Tropospheric Monitoring Instrument (TROPOMI) NO₂ and HCHO columns and by surface NO₂ and O₃ from the air quality monitoring station network. The analysis covers 1–14 May 2022, a period of climatologically high O₃. Optimized NO_x emissions exhibit strong diurnal adjustments relative to the prior (nighttime reductions up to 51 % and daytime increases up to 14 %). The joint inversion of NO_x and VOC delivers the largest improvement in O₃ simulations, achieving the best agreement with observations (IOA > 0.8). Constrained emissions shift O₃ sensitivity from VOC-sensitive to NO_x-sensitive across much of the domain, improving spatial consistency with TROPOMI-derived formaldehyde-to-NO₂ ratio (FNR) diagnostics. Adjoint-based hourly ΔO₃ responses reveal regime- and hour-dependent behavior: VOC controls are most effective under VOC-sensitive conditions, whereas NO_x controls are more direct under NO_x-sensitive conditions. Importantly, because O₃ titration is immediate while photochemical production requires finite reaction time, emissions released approximately 1–2 hours earlier have the greatest influence on current O₃, motivating hour-specific, regime-specific controls. Overall, the hybrid framework improves O₃ simulations and sensitivity-regime diagnosis, enabling spatiotemporally resolved precursor emission reduction guidance for effective O₃ mitigation.

Ozone (O₃) over South Korea has risen in recent years, underscoring the need to accurately quantify emissions of nitrogen oxides (NO_x) and volatile organic compounds (VOC). We develop a hybrid inverse modeling framework that couples the Finite Difference Mass Balance (FDMB) method with four-dimensional variational (4D-Var) assimilation using the Community Multiscale Air Quality (CMAQ) model to jointly constrain spatiotemporal NO_x and VOC emissions. The inversion is constrained by Tropospheric Monitoring Instrument (TROPOMI) NO₂ and HCHO columns and by surface NO₂ and O₃ from the air quality monitoring station network. The analysis covers 1–14 May 2022, a period of climatologically high O₃. Optimized NO_x emissions exhibit strong diurnal adjustments relative to the prior (nighttime reductions up to 51 % and daytime increases up to 14 %). The joint inversion of NO_x and VOC delivers the largest improvement in O₃ simulations, achieving the best agreement with observations (IOA > 0.8). Constrained emissions shift O₃ sensitivity from VOC-sensitive to NO_x-sensitive across much of the domain, improving spatial consistency with TROPOMI-derived formaldehyde-to-NO₂ ratio (FNR) diagnostics. Adjoint-based hourly ΔO₃ responses reveal regime- and hour-dependent behavior: VOC controls are most effective under VOC-sensitive conditions, whereas NO_x controls are more direct under NO_x-sensitive conditions. Importantly, because O₃ titration is immediate while photochemical production requires finite reaction time, emissions released approximately 1–2 hours earlier have the greatest influence on current O₃, motivating hour-specific, regime-specific controls. Overall, the hybrid framework improves O₃ simulations and sensitivity-regime diagnosis, enabling spatiotemporally resolved precursor emission reduction guidance for effective O₃ mitigation.

Ozone (O₃) over South Korea has risen in recent years, underscoring the need to accurately quantify emissions of nitrogen oxides (NO_x) and volatile organic compounds (VOC). We develop a hybrid inverse modeling framework that couples the Finite Difference Mass Balance (FDMB) method with four-dimensional variational (4D-Var) assimilation using the Community Multiscale Air Quality (CMAQ) model to jointly constrain spatiotemporal NO_x and VOC emissions. The inversion is constrained by Tropospheric Monitoring Instrument (TROPOMI) NO₂ and HCHO columns and by surface NO₂ and O₃ from the air quality monitoring station network. The analysis covers 1–14 May 2022, a period of climatologically high O₃. Optimized NO_x emissions exhibit strong diurnal adjustments relative to the prior (nighttime reductions up to 51 % and daytime increases up to 14 %). The joint inversion of NO_x and VOC delivers the largest improvement in O₃ simulations, achieving the best agreement with observations (IOA > 0.8). Constrained emissions shift O₃ sensitivity from VOC-sensitive to NO_x-sensitive across much of the domain, improving spatial consistency with TROPOMI-derived formaldehyde-to-NO₂ ratio (FNR) diagnostics. Adjoint-based hourly ΔO₃ responses reveal regime- and hour-dependent behavior: VOC controls are most effective under VOC-sensitive conditions, whereas NO_x controls are more direct under NO_x-sensitive conditions. Importantly, because O₃ titration is immediate while photochemical production requires finite reaction time, emissions released approximately 1–2 hours earlier have the greatest influence on current O₃, motivating hour-specific, regime-specific controls. Overall, the hybrid framework improves O₃ simulations and sensitivity-regime diagnosis, enabling spatiotemporally resolved precursor emission reduction guidance for effective O₃ mitigation.

The Geostationary Environment Monitoring Spectrometer (GEMS) onboard the GEO-KOMPSAT-2B satellite provides high spatiotemporal resolution observations of various atmospheric constituents over East Asia, including ozone, its precursors (NO2, HCHO), SO2, and aerosols. The GEMS total ozone retrieval algorithm (O3T), based on the TOMS Version 9 look-up table approach, was recently updated to Version 2.2 in December 2024. However, the current GEMS total ozone product has shown a systematic underestimation compared to other satellite observations, such as TROPOMI and OMI, as well as ground-based measurements, mainly due to limitations in the Level 1C (L1C) irradiance data.To address these issues, we evaluate the O3T algorithm using two types of updated L1C datasets: one applying only the bidirectional transmittance distribution function (BTDF) correction (A1), and another applying both the BTDF and degradation corrections (A2). The results show that the use of both A1 and A2 substantially reduces the underestimation of total ozone, with particularly pronounced improvements at high ozone levels. Furthermore, when using A2, the retrieved total ozone exhibits improved long-term stability. Using these improved GEMS total ozone products, we analyze the characteristics and variability of total ozone over East Asia. In addition, we investigate the impact of cloud information, another critical input to the O3T algorithm, on the total ozone retrieval. These results enhance the reliability of GEMS ozone retrievals and provide a foundation for further algorithm optimization.

The choice of a priori ozone profile is a key factor in IASI ozone profile retrievals. In the operational FORLI algorithm, a single fixed a priori profile from the McPeters/Labow/Logan climatology (McPeters et al., 2007) is used globally (Hurtmans et al., 2012). Previous work has shown that a tropopause-based ozone profile climatology can provide a more appropriate constraint near the tropopause and improve retrieval performance (Bak et al., 2013). However, the use of dynamic-dependent priors complicates pixel-to-pixel global intercomparisons. Here, we quantify the sensitivity of IASI ozone profile retrievals to the a priori choice over East Asia, leveraging intensive ozonesonde observations in Korea. Using the research retrieval framework Atmosphit, we conduct controlled sensitivity experiments in which only the a priori ozone climatology is changed. We first reproduce the FORLI-type configuration by adopting the McPeters/Labow/Logan climatology within Atmosphit and then construct and apply an alternative tropopause-based (TB) ozone climatology following Bak et al. (2013). Retrieved profiles are evaluated against continuous ozonesonde profiles from Anmyeondo during Pre-ACCLIP (2021) and ACCLIP (2022). Compared with the fixed climatological prior, the TB a priori improves retrieval consistency in the UTLS. Improvements are also found for tropospheric ozone consistency and total column ozone (TCO). These results support the use of tropopause-based ozone climatologies to enhance IASI ozone profile retrieval quality and provide practical guidance for research retrieval configurations targeting UTLS processes.Financial supportThis work was funded by the Korea Meteorological Administration Research and Development Program under Grant (RS-2025-02219688). References1. Bak, J., Liu, X., Wei, J. C., Pan, L. L., Chance, K., & Kim, J. H. (2013). Improvement of OMI ozone profile retrievals in the upper troposphere and lower stratosphere by the use of a tropopause-based ozone profile climatology. Atmospheric Measurement Techniques, 6(9), 2239-2254. https://doi.org/10.5194/amt-6-2239-20132. Hurtmans, D., Coheur, P. F., Wespes, C., Clarisse, L., Scharf, O., Clerbaux, C., Hadji-Lazaro, J., George, M., & Turquety, S. (2012). FORLI radiative transfer and retrieval code for IASI. Journal of Quantitative Spectroscopy and Radiative Transfer, 113(11), 1391-1408. https://doi.org/https://doi.org/10.1016/j.jqsrt.2012.02.0363. McPeters, R. D., Labow, G. J., & Logan, J. A. (2007). Ozone climatological profiles for satellite retrieval algorithms. Journal of Geophysical Research: Atmospheres, 112(D5). https://doi.org/https://doi.org/10.1029/2005JD006823

The Tropospheric Emissions: Monitoring of Pollution (TEMPO) mission is part of a global constellation of geostationary satellites, along with GEMS and Sentinel-4, dedicated to monitoring air quality across the Northern Hemisphere. TEMPO is the first geostationary satellite instrument to monitor air pollutants over North America on an hourly basis at nearly neighborhood-scale resolution, covering an area from Mexico City to the Canadian oil sands and from the Atlantic to the Pacific Ocean.TEMPO measures backscattered ultraviolet and visible radiation to observe several trace gases important to air quality, including ozone, nitrogen dioxide, and formaldehyde, with observations every 40–60 minutes and at a high spatial resolution of approximately 2 × 4.75 km². TEMPO was successfully launched in April 2023 and began nominal operations in October 2023. Since then, it has been continuously monitoring atmospheric pollutants across its observation domain.This presentation summarizes the Version 4 (V04) updates and improvements to the TEMPO total-ozone (O3TOT) and ozone-profile (O3PROF) retrieval algorithms. This presentation also presents the evaluation of the upcoming V04 TEMPO O3TOT product through comparisons of total ozone columns (TOCs) with measurements from other satellite instruments (e.g., OMPS and TROPOMI) and ground-based instruments, including Pandora, Brewer, and Dobson spectrometers. The V04 TEMPO O3PROF algorithm, which is UV-only, is validated through comparisons of ozone profiles, tropospheric ozone, and 0–2 km ozone columns with those from the Tropospheric Ozone Lidar Network (TOLNet) and aircraft observations, as well as through validation with MLS, EPIC, TROPOMI, and OMI observations.