Low-cost particulate matter(PM) sensors require demonstrated long-term stability and representativeness to be reliably applied in complex urban environments. This study evaluates a climate-controlled light-scattering sensor (Gonggam Sensor GGS727) by analyzing four years (2021 similar to 2024) of continuous PM2.5 and PM1 measurements collected at Korea University and comparing them with Beta Attenuation Method (BAM) observations from 25 regulatory monitoring stations across Seoul. The GGS727 achieved high data completeness (>97%) and maintained stable sensitivity throughout the observation period. Hourly-averaged sensor data showed strong spatial correlations with BAM measurements, with an average R-2 of 0.81 and RMSE of similar to 6.5 mu g m(-3) across all districts. Performance decreased gradually with distance from the sensor site, indicating that airflow similarity and spatial proximity largely govern coherence between local sensor observations and city-scale PM2.5 variability. Long-term regression parameters exhibited minimal interannual drift, demonstrating stable temporal consistency of the sensor. Seasonal comparison with the Jongno monitoring station revealed distinct relative response behaviors. In summer, the sensor tended to report slightly lower concentrations, consistent with organic-rich fine aerosol characterized by lower refractive index, enhanced volatility, and a shift toward smaller particle sizes with reduced scattering efficiency. In winter, nitrate-rich inorganic aerosol enhanced hygroscopic growth and increased optical scattering, leading the sensor to report relatively higher PM2.5 than the BAM instrument. These discrepancies reflect inherent differences in measurement principles and aerosol optical/thermodynamic properties rather than a loss of accuracy or stability. The sensor further provided robust PM1 measurements, with PM1 accounting for similar to 69% of PM2.5 on average and exhibiting strong seasonal variability (higher in summer, lower in winter). These patterns align with transitions between secondary organic aerosol formation and wintertime accumulation of inorganic salts, underscoring the value of submicron measurements for interpreting aerosol processes. Overall, the results demonstrate that a miniature climate-controlled sensor can reproduce urban PM2.5 variability with FEM-comparable fidelity while capturing additional information on ultrafine particles. Such performance highlights the utility of low-cost sensors for dense urban monitoring networks and for enhancing population exposure assessment in environments with strong seasonal and spatial heterogeneity.
Organic compounds, such as polycyclic aromatic hydrocarbons (PAHs) and fatty acids (FAs), are emitted from specific sources and can serve as molecular markers for PM2.5 source apportionment. In this study, the characteristics and sources of PM2.5 and 57 associated organic species (OS) were investigated in Ulsan, Korea. The OS exhibited significant correlations with PM2.5 and organic carbon, indicating that they can serve as effective tracers of PM2.5 high-pollution episodes. FAs and n-alkanes were dominant in spring, reflecting mixed influences from industrial activities and biogenic emissions, whereas levoglucosan concentrations were elevated in fall, highlighting a strong impact of biomass burning. Positive Matrix Factorization (PMF) was applied under two scenarios: one using major PM2.5 components and the other incorporating both major components and organic species. The PMF results identified secondary inorganic aerosol as the largest contributor to PM2.5, followed by biomass burning, industrial activities, fossil fuel combustion, and natural sources. When 21 organic tracers were included, the PMF analysis further resolved secondary and primary organic aerosol sources and distinguished biomass-burning contributions into local and long-range transported sources. This source separation was further supported by diagnostic ratios of PAHs. Overall, these results demonstrate that integrating organic molecular markers into receptor modeling substantially improves PM2.5 source resolution and provides quantitative evidence for the combined effects of primary emissions, secondary formation, and regional transport in an industrialized coastal city.
This study aims to analyze the chemical composition of particulate and gaseous substances in real time during the burning of agricultural crop residues (i.e., rice straw, barley stalk, corn stalk, sesame stalk, perilla stalk, peach branch, apple branch, grape branch, pear branch), and to quantitatively evaluate the impact of biomass burning by calculating emission factors by combustion fuel. An open chamber system for crop residue burning was constructed in compliance with EPA5G regulations and utilized mass flow controllers to provide precise flow rate control. The chemical composition and emission factors of PM1.0 sub-compounds (organics, nitrate (NO3-), sulfate (SO42-), ammonium (NH4+), chloride (Cl-), black carbon (BC)), gaseous substances(CO, CO2, CH4, O-3, NOx, SO2, NH3), and 12 types of volatile organic compounds(VOCs) were investigated for nine types of crop residues. In the case of agricultural residue incineration, the average PM1.0 emission factor was 0.01575 kg/kg, with the highest value observed in barley stalk. Rice straw and barley stalk recorded relatively high Cl- emission factors of 0.0029 kg/kg and 0.004 kg/kg, respectively, compared to other fuels. This is likely due to the accumulation of Cl- within the crops, attributed to the use of KCl fertilizer in paddy fields for cadmium contamination remediation. The average emission factors of CO, CO2, CH4, O-3, NOx, SO2, NH3 were 0.18, 4.54, 0.012, 0.0093, 0.0049, 0.00108, 0.0032 kg/kg, respectively while the average emission factor of VOCs was 0.001 kg/kg, with acetaldehyde, benzene, and acetone accounting for 64% of the total VOC emissions. The results provide fundamental data to support domestic air quality impact assessments, particularly in relation to emissions from agricultural residue combustion.
This study quantified the fraction of secondary organic carbon (SOC) within organic carbon (OC) in fine particles (PM2.5) and investigated the characteristics of SOC formation and spatial origins of both SOC and primary organic carbon (POC) in PM2.5 in Seoul using hourly resolved atmospheric monitoring data in 2023. While PM2.5 and its major components such as NO3- and elemental carbon(EC) exhibited typical winter-high and summer-low seasonal patterns, OC maintained elevated levels across all seasons. Using the minimum R-squared (MRS) method, the annual average POC and SOC concentrations were estimated at 2.3 and 1.9 mu g m(-3), respectively, with SOC accounting for 45% of the total OC and peaking at 3.7 mu g m(-3) in July. During summer, SOC showed strong correlations with O-3 and O-x (i.e., sum of O-3 and NO2) under high temperatures, indicating the dominant photochemical oxidation. In contrast, winter SOC was linked to elevated CO, NO2, and high relative humidity, which suggests the importance of aqueous-phase reactions and the accumulation of primary emissions. Spatial analyses, including bivariate polar plots and concentration weighted trajectory (CWT) modeling, confirmed that summer SOC was dominantly formed locally from domestic precursors under stagnant conditions. In contrast, elevated winter and spring carbonaceous aerosol concentrations were significantly influenced by the long-range transport of pollutants from northeastern China across the Yellow Sea, coupled with regional aqueous chemistry. These findings highlight that effective mitigation of OC in PM2.5 in Seoul requires season-specific strategies: controlling local VOCs and oxidant levels to reduce summer photochemical SOC while addressing both regional transport and local precursor emissions in winter.
Volatile organic compounds (VOCs) are key precursors in the formation of ozone and secondary organic aerosols. Emitted into the atmosphere from a variety of sources, VOCs exhibit a wide range of reactivity. Consequently, accurately determining VOC species is imperative for reducing VOC-related air pollution on local and regional scales. In this study, 56 VOC species were measured in real time using the TD-GC-FID method, which involves collecting airborne VOCs by thermal desorption (TD) tubes, separating them via gas chromatograph (GC), and determining their concentrations with a flame ion detector (FID). Experiments were conducted for the Satellite Integrated Joint Air Quality (SIJAQ) campaign at the Mediheal Earth Environmental Science Hall of Korea University in Seoul in May similar to June (early summer) and November (early winter) 2022. In this campaign, the proportion of alkene has increased compared to the past, with the proportion of alkene 25.3% in summer and 23.3% in winter, respectively. Alkenes such as propylene, trans-2-pentene, trans-2-butene were higher in summer than in winter and exhibited the highest ozone formation potential (OFP). In particular, butene and pentene isomers along with propylene, showed strong correlations with temperature, suggesting the contribution of fugitive emissions through evaporation.The temporal distribution of VOCs, correlations, and principal component analysis results all emphasize automobile emissions as the main source ofVOCs in Seoul.
This study conducted time-series and correlation analyses using data from the urban air quality and atmospheric heavy metal monitoring networks near a smelter. The results showed that the mean monthly concentration of SO2 in the smelter-affected area during the entire period (May 2019 similar to December 2024) was 0.0065 ppm, approximately 2.3 times higher than the national mean monthly average concentration (0.0028 ppm). Similarly, the concentrations of heavy metals such as Pb, As, and Cd were up to nine times higher in the vicinity of the smelter compared to the national average levels. Seasonal and spatial analyses across five sites(Site A, B, C, D, and E) revealed that PM10 and PM2.5 were highest in spring and winter, while NO2, SO2, and heavy metals(Pb and Cd) peaked in winter. Near the smelter, the seasonal variability of heavy metals was most pronounced, with Pb, Cd, and As concentrations 1.8 similar to 2.7 times higher in winter than in summer. Correlation analysis within the smelter region revealed a strong relationship between Pb and Cd(r(2)=0.81), suggesting a common emission source. In contrast, weak correlations between the smelter and reference sites (Site B, C, D, and E) indicate that the smelter area exhibits unique behavioral characteristics of heavy metals (Pb, Cd, and As) distinct from other regions. These findings imply that emissions from smelting processes play a dominant role in shaping local air quality and the distribution of heavy metals in the surrounding atmosphere.
Industrial emissions of volatile organic compounds (VOCs) are a significant environmental concern. Once released, VOCs undergo photochemical and oxidative reactions, generating secondary degradation products with structural and toxicological characteristics that differ from those of the parent compounds. However, understanding of the formation mechanisms and environmental behavior of these degradation products remain limited and current regulations primarily focus on VOC emission reduction rather than on the management of their transformation products. This study proposes an integrated analytical framework to predict and evaluate the potential hazards associated with VOC degradation products emitted from an industrial area. This framework was applied using VOC measurement data collected near an industrial complex in Incheon, South Korea. Two predictive models were applied to simulate atmospheric degradation processes: Zeneth, a rule-based system that identifies major reaction pathways under defined environmental conditions (pH, temperature, radicals, light, and humidity), and Reaction Mechanism Generator (RMG), a mechanistic model that automatically constructs reaction networks based on temperature, pressure, and concentration parameters. Applying both models enabled simultaneous structural prediction and kinetic interpretation, allowing realistic simulation of atmospheric VOC degradation processes. For toxicity assessment, in silico predictive tools such as VEGA QSAR, DEREK NEXUS, and the OECD QSAR Toolbox were used to analyze multiple toxicological endpoints. Some degradation products exhibited potential toxicity, and the prediction results were compared with GHS classifications to verify the reliability of the models. Overall, this integrated approach provides a scientific basis for understanding the degradation and toxicity characteristics of VOCs and can be utilized to identify and prioritize hazardous substances in environmental risk management.
Rice cultivation is the largest source of greenhouse gas emissions in the agricultural sector. Methane (CH4) emissions from rice paddies vary depending on cultivation practices such as water management and the use of organic amendments. To account for regional differences in these practices, this study applied the 2019 IPCC guidelines to estimate methane emissions from rice cultivation across the nations, covering every paddy field in each administrative districts in South Korea. Emission factors were calculated at the household level by incorporating variations in water management and organic amendment practices, and then aggregated by administrative district. The total national methane emissions from rice cultivation in 2020 were estimated at 244,911 tons of CH4, showing a consistency of 89.3%. A Monte Carlo simulation reflecting parameter uncertainties yielded a relative uncertainty of 61.6%. This bottom-up approach captures regional variations in cultivation management and complements the current national inventory methodology by providing a framework to enhance estimation accuracy and support region-specific methane mitigation strategies.
This study examines the distribution characteristics of volatile organic compounds (VOCs) over the Seoul metropolitan area in March 2020 based on aircraft measurements. A total of three research flights (RF1, RF2, RF3) were conducted under different meteorological conditions using a proton transfer reaction-time of flight-mass spectrometer (PTR-ToF-MS) mounted on a research aircraft. The measurements revealed that VOC concentrations were significantly elevated during stagnant atmospheric conditions (RF2), in conjunction with increased concentrations of ozone (O3), carbon monoxide (CO), and nitrogen dioxide (NO2), suggesting enhanced photochemical activity. Methanol was the most abundant VOCs across all RF, accounting for over 50% of the total VOCs. Acetaldehyde and acetone were also observed at high levels, indicating strong secondary production under photochemically favorable conditions. Anthropogenic VOCs such as toluene, benzene, and acetonitrile were notably higher in RF2, implying the influence of industrial and traffic-related emissions. Biogenic VOCs (BVOCs), including isoprene and pinene, showed increased levels in RF2, with isoprene oxidation products such as methyl vinyl ketone (MVK) providing additional evidence of active photochemistry. These findings highlight the complex interplay of anthropogenic, biogenic, and secondary sources in shaping the VOCs composition over the Seoul metropolitan region. This study demonstrates the utility of aircraft-based measurements for characterizing upper-level VOCs distributions and underscores the need for expanded airborne observations during different seasons to improve understanding of VOCs sources, transport pathways, and atmospheric transformation processes in urban environments.
This study investigated the transport characteristics and potential source regions of nitrogen dioxide (NO2) in Yeosu, a coastal industrial city in South Korea, using five years(2020 similar to 2024) of ground-based air quality observations. Uniform Manifold Approximation and Projection (UMAP) and K-Means clustering were applied to major air pollutant concentrations to identify four pollution clusters (Clusters 0 similar to 3), while associated meteorological conditions were examined separately. Backward trajectories (48 h, 500 m) were calculated using the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model, and cluster-specific Concentration Weighted Trajectory (CWT) and Potential Source Contribution Function (PSCF) analyses were conducted. The results revealed that the mean NO2 concentration in the high-pollution cluster (Cluster 3) was approximately 2.6 times higher than that in the low-pollution cluster(Cluster 2). Cluster 3 was primarily associated with wintertime northwesterly continental airflows and atmospheric stagnation, with potential source regions identified over the central Yellow Sea and eastern China. In contrast, Cluster 2 was dominated by summertime marine air masses and high precipitation, indicating limited long-range influence. Cluster 0 and 1 represented intermediate or transitional states characterized by particulate matter-dominant patterns and active photochemistry, respectively. This integrated approach demonstrates the critical importance of cluster-specific analysis in deciphering the complex transport dynamics of NO2 within coastal industrial regions.
Water-soluble organic matter (WSOM) is a key component of atmospheric aerosols, influencing their physicochemical properties and atmospheric behavior. In this study, the fluorescence characteristics and transport-related features of WSOM were investigated in Yeosu, a coastal industrial city in southern Korea affected by both continental and marine air masses. A total of 63 aerosol samples were collected at 8-h intervals from April to May 2025 and analyzed using excitation-emission matrix (EEM) fluorescence spectroscopy coupled with parallel factor analysis (PARAFAC). This analysis resolved three components, including two protein-like (C1 and C2) and one humic-like (C3) component. Additionally, fluorescence indices (FI, BIX, and HIX) and 48-h backward trajectory analysis using the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model were applied to examine general source characteristics and transport patterns. Notably, no statistically significant differences in fluorescence components or indices were observed among trajectory clusters. This points to a potential homogenization process in which active air-mass mixing overrides initial source-specific signals. Despite this homogenization, trajectory-based spatial analyses (CWT) suggested limited component-dependent transport tendencies, particularly for the more stable humic-like component. Overall, this study indicates that the observed homogeneity is a distinctive characteristic of WSOM processing in a complex coastal-industrial atmosphere. Furthermore, integrating fluorescence characteristics with air-mass transport may provide deeper insights into these source-processing relationships.
In this study, the performance of a Raman Lidar using an ultraviolet(UV) laser for remote measurement of methane VMRs was evaluated for the first time. Gas cell experiments were conducted to investigate the sensitivity of the methane raman signal to variations in methane partial pressure inside a chamber located at a remote distance from the Lidar. The results showed a very strong correlation between the Raman Lidar signal and the methane partial pressure in the chamber, with a correlation coefficient of 0.99, confirming that the methane raman signal accurately represents the methane VMR. To further verify the applicability of the system under real environmental conditions, field observations were carried out at a landfill site. Methane VMRs measured by the Raman Lidar at a distance of 336 m were compared with simultaneous measurements obtained from an methane in-situ instrument installed at the same location. The comparison showed a high correlation coefficient of 0.99 between the two datasets. The mean absolute error (MAE), root mean square error (RMSE), and percentile difference were calculated as 2.56 ppm, 4.43 ppm, and 6.76%, respectively. These results indicate that the Raman Lidar provides methane VMR measurements comparable to those of the in-situ instrument, even under remote measurement conditions. Overall, the results demonstrate that the Raman Lidar is capable of stably measuring methane VMRs over a wide range, from ppm levels to percent (%) levels, at remote distances. Therefore, the Raman Lidar is expected to be a useful remote sensing tool for measuring methane VMRs and detecting leaks from spatially distributed emission sources using a limited number of instruments, and for supporting methane emission estimation when combined with additional information.
