Long-term historical trends in air pollutant emissions in South Korea (2000–2018)

This study aims to understand the impact of previous air quality improvement policies on historical emission changes by examining long-term emission trends in Korea. Annual emissions from 2000 to 2018 were estimated using Korea’s official emissions inventory, the Clean Air Policy Support System (CAPSS). To ensure a consistent comparison, standardization of the method for calculating emissions and unification of the reported emission sectors were conducted each year. Furthermore, Korea’s emissions history was compared with that of neighboring countries, such as China and Japan. The annual emissions of these countries were acquired from the HTAPv3 emissions inventory, an international long-term emission trend study. For comparison, the emission source classification of Korea was matched with that of HTAPv3. As a result of the analysis, NOx and SOx emissions in Korea have shown decreasing trends, whereas VOCs (volatile organic compounds) have indicated a gradual increasing trend since 2000. Compared to the previous period of implementing South Korea’s air quality improvement policy, changes in NOx and SOx emissions, which are combustion-related pollutants, showed a relationship with the policy’s timeline. However, non-combustion-related pollutants such as VOCs did not exhibit such a relationship. It was concluded that the related policies were not as effective in reducing VOCs as planned in the policy. By comparing the emission trends of Japan, Korea, and China, it was confirmed that Japan was the first country to experience a decrease in combustion-related pollutants emissions, followed by Korea and China. Additionally, combustion-related pollutants decreased in all three countries, whereas VOCs decreased only in Japan. VOC is a precursor material generating secondary PM2.5 and Ozone; considering that, if relevant policies are additionally implemented to control future PM2.5 concentrations, and to reduce emissions efficiently and effectively, Japan’s VOC reduction policies can be applied to Korea’s emission reduction policies. These results are expected to serve as important references when establishing future air quality improvement policies in Korea.


Introduction
With Korea's recent economic growth and increased interest in air pollution, arising attention has been paid to various issues regarding air quality.It is argued that the cause of the air quality problem is not only the effect of PM 2.5 emissions transported from China but also the increase in emissions from domestic sources (Bae and Kim, 2016;Kim et al., 2023).This issue has been addressed by the Korean government since 2000 while discussing better air quality and promoting diverse air quality policies.Air quality in Korea has improved since 2000 owing to efforts to improve the atmospheric environment; however, since 2013, the improvement has been stagnant, leading to a high level of PM 2.5 concentrations in the last 3 years (2015)(2016)(2017).The different types and scales of the previous air quality improvement policies yielded varying impacts.Therefore, it is essential to scrutinize the emission trends observed while implementing previous air pollution reduction policies.Such data can help predict the mitigating effects of future air quality policies and can assist in developing more effective policies.To accomplish this, a long-term emission inventory analysis is necessary.Since 1999, Korea has prepared the Clean Air Policy Support System (CAPSS) (Lee et al., 2011), a national air pollutant emission inventory.This inventory can be used to conduct long-term emission analyses.
In addition, it is essential to analyze the historical emission trends of adjacent countries such as China and Japan to develop more effective air quality improvement policies in Korea.Since the two countries are located closest to Korea among the Northeast Asian countries, share the atmosphere, and influence each other, understanding the emissions from these countries is quintessential.Furthermore, the sustained implementation of air quality improvement policies by these countries has a persistent influence on the air quality in Korea.Since 2000, there have been research projects on Long-range Transboundary Air Pollutants in Northeast Asia (LTP) (NIER, 2019) and annual Tripartite Environment Ministers' Meetings among Korea, China, and Japan (TEMM) since 1999 to understand and improve each country's air quality conditions.The Task Force on the Hemispheric Transport of Air Pollution (TF HTAP) analyzes long-term global emissions.The TF HTAP is an international scientific cooperation study to improve the understanding of the intercontinental transport of air pollution across the Northern Hemisphere and was organized with the support of the Convention on Long-range Transboundary Air Pollution (CLRTAP) of the United Nations Economic Commission for Europe (UNECE) in 2005 (Janssens-Maenhout et al., 2015).TF HTAP recently completed a study on calculating long-term emissions from 2000 to 2018 and provides emission data under HTAP version 3 (henceforth "HTAPv3") (Crippa et al., 2023).HTAPv3 contains emission data from China and Japan, enabling a comparison between these data and Korea's long-term emission trends.
In this study, Korea's annual emissions from 2000 to 2018 were estimated using CAPSS.However, CAPSS has different emission calculation methods for each year.In this regard, a unified annual emission calculation method is necessary to compare emission patterns from both long-term and continuous perspectives.This process also involved replenishing emissions from sectors unaccounted for in certain years.Therefore, to ensure an accurate comparison, standardization of the method for calculating emissions and unification of the reported emission sectors were conducted for each year.The most recent emissions calculation method, the 2018 method, was chosen as the standardized method.Subsequently, an analysis was conducted on the long-term trends in emissions in Korea, and the relationship between policies and emissions was investigated by comparing the variations in emissions with the timing of prior air pollution improvement policies.Second, the long-term emission trends in China, Japan, and Korea were compared.Emission data from 2000 to 2018 for China and Japan were acquired using HTAPv3.However, because the classification structures of the emission sources of HTAPv3 and Korea's emission inventory are different, it is difficult to compare these two datasets at the same level.For comparison, the emission source classification of Korea was matched with that of HTAPv3.The annual changes in emissions due to pollutants were then compared across countries.

Korea emission inventory, CAPSS
Through the CAPSS, the air pollutant emission inventory and emission calculation-related data from approximately 150 organizations and 300 activities were systematically collected and managed.CAPSS's classification system of emission sources is based on the European Environment Agency's (EEA) CORINAIR's (CORe INventory AIR emission) classification system of emission sources, which has been adapted for the Korean context (Lee et al., 2011;NIER, 2013a).Emission sources were divided into four levels (upper, middle, lower, and detailed), and emissions from 100 emission sources were estimated.The inventory provides information on nine designated pollutants (i.e., CO, NO x , SO x , TSP, PM 10 , PM 2.5 , NMVOC, NH 3 , and BC) and emissions data from 250 cities, counties, and districts throughout Korea since 1999.
The upper emission source level consists of 13 main categories of emission sources (i.e., energy industry combustion, non-industry combustion, manufacturing combustion, production processes, energy transport, storage, solvent use, roads, non-roads, waste management, agricultural activities, other area sources, fugitive dust, and biomass combustion), along with different calculation methods per emission source (NIER, 2013b).Point source emissions were calculated using a bottom-up method that relies on data from relevant agencies.However, fuel consumption from sources other than point sources, such as the energy industry, nonindustrial combustion, and manufacturing combustion, are classified as area sources.The emissions from these sources were calculated using a top-down approach that utilized statistical data from the Korea National Oil Corporation, Coal Association, and city gas companies (NIER, 2013a).Emissions from road mobile sources were estimated based on the average speed according to the methodology of the European CORINAIR COPERT (Computer Programme to calculate Emissions from Road Transport).
This study utilized emissions inventory data obtained from CAPSS from 2000 to 2018.During the analysis, changes were discovered in the methodology used to calculate emissions and classify emission sources by year.Previously unaccounted sources and pollutants are identified each year.Therefore, the corresponding activity data were collected and recalculated to supplement missing emissions.The most recent emissions calculation method, the 2018 one, was applied to recalculate emissions for each year.

National mosaic emission inventory, HTAP
The HTAP is a comprehensive global emission inventory based on official emission data and scientific evidence reported by each country to study the intercontinental transport of air pollutants across the Northern Hemisphere.HTAPv3 included data from CAMS-REG-v5.1 of the European Monitoring and Evaluation Programme (EMEP) for Europe (Kuenen et al., 2022), the US Environmental Protection Agency (EPA) for the US (https:// www.epa.gov/ cmaq/ equat es), the Environment and Climate Change Canada (ECCC) for Canada (APEI, 2018), and the Regional Emission Inventory in Asia (REAS) for Asian countries, except for Japan and Korea (https:// www.nies.go.jp/ REAS/).For Japan's emissions data, an inventory developed by Japan's Study for Reference Air Quality Modeling (J-STREAM) from the National Institute of Environmental Studies (NIES) and the PM 2.5 Emission Inventory (PM2.5EI)developed by the Ministry of the Environment, Japan (MOEJ) were used (https:// www.env.go.jp/ air/ osen/ pm/ info.html).In Korea, the modified CAPSS emissions resulting from this study were incorporated into the HTAPv3.EDGARv6.1 (Emissions Database for Global Atmospheric Research, https:// edgar.jrc.ec.europa.eu/ datas et_ ap61) was used to create emissions grid maps for the rest of the world.
The recently developed HTAPv3 has a resolution of 0.1 × 0.1°, encompassing a time series from 2000 to 2018.It consists of a global mosaic of emission grid maps of monthly air pollutants (CO, NO x , SO 2 , PM 10 , PM 2.5 , NMVOCs, NH 3 , BC, and OC), and provides information on 16 anthropogenic emission sectors, excluding Land Use, Land Use Change, and Forestry.Compared with previous HTAP mosaics of anthropogenic inventories (e.g., HTAPv1 and HTAPv2.2),HTAPv3 aims to expand the temporal range of air pollutant emissions, sectoral classification, and regional scope of official data.One of the key strengths of HTAPv3 is its temporal range, which enables the analysis of emission trends over the last two decades.This long-term global mosaic of emission inventories can be utilized as data contributing to global air quality modeling and each country's policies to solve the complexity of long-distance transport of air pollutants (Crippa et al., 2023).
The HTAPv3 dataset contains emissions data from 2000 to 2018.For this study, emissions from China and Japan were obtained from the dataset, and emissions from Korea were compared with those over the long term.To enable a proper comparison, Korea's emissions inventory was adjusted to have the same emission source classification structure as HTAPv3.The long-term emission trends of the three countries were analyzed and compared.

Reanalysis method
The CAPSS generates an emission inventory every year in which new emission sources are added and methodologies are improved as required.Thus, methods for estimating emissions from previously unaccounted sources are continuously being developed.However, this also implies annual differences in emission sources and calculation methods, making it difficult to directly compare past and current emissions on the same basis.To ensure a consistent comparison of Korea's emissions over time, we used the 2018 emissions calculation method (NAEIR, 2020a) and reanalyzed past emissions.This allowed us to fill the gaps in the data and update the methods to ensure consistency across the entire period.The reanalysis methods used in this study were divided into two categories: updating and gap-filling methods.We named the resulting emissions inventory from 2000 to 2018 as CAPSS-KU (CAPSS-Konkuk University).The CAPSS-KU was provided to HTAPv3 and has been shared as a dataset for the Korean region within HTAPv3.
Table 1 provides an overview of the updated emissions calculation methods and additional sectors included for each year.The updated method was applied to the following parts: update of activity data, update of emission factors at major and middle classifications, application of deterioration factors of mobile pollution sources, update of emission calculation methods from diesel vehicles and construction equipment, and update of VKT (vehicle kilometers traveled) for passenger vehicles.The gap-filling method was used to estimate emissions for sectors which were not previously included in the CAPSS inventory but included in the inventory from some years and onwards.These sectors include fugitive dust, biomass combustion, and livestock farming.The emissions of seven substances (CO, NO x , SO x , PM 10 , PM 2.5 , VOC, and NH 3 ) were estimated for these sectors.To calculate NO x emissions before 2015, emission factors based on the driving conditions applied since 2016 for the mobile pollutant sector were used.PM 10 emissions have been estimated based on imported anthracite use since 2007.To estimate PM 10 emissions during the period from 2000 to 2006, data on imported anthracite use were collected and used to calculate the corresponding emissions.PM 2.5 emissions were first calculated in 2011, whereas PM 10 emissions had already been calculated for previous periods.To estimate PM 2.5 emissions between 2000 and 2010, the ratio of PM 2.5 /PM 10 emissions since 2011 was calculated and then applied to the PM 10 emission data from an earlier period.

Mapping emission source categories
In this study, the emission inventory for Korea was corrected and labeled CAPSS-KU.When the CAPSS-KU was prepared for inclusion in the HTAPv3, it was reclassified into the emission source categories required by HTAPv3 and created.This procedure made it possible to compare Korea's emissions with China's and Japan's emissions, which are included within HTAPv3.As shown in Table 2, the CAPSS-KU, which had 13 major categories (level 1: (1) energy production, (2) non-industry combustion, (3) manufacturing industry, (4) industrial process, (5) energy transport and storage, (6) solvent use, (7) road transport, (8) non-road transport, (9) waste, (10) agriculture, (11) other surface-pollutant source, (12) fugitive dust, (13) biomass combustion), eventually had 14 reclassified categories based on the HTAPv3 classification system.After conducting a detailed comparison of CAPSS-KU's level 1 and 2 source categories with those of HTAPv3, they were mapped to the most closely matching source classification.The level 2 source category of CAPSS-KU is shown on the right side of the table, and the HTAPv3 source categories are shown on the left side of the table.The HTAPv3 categories consist of aviation (domestic aviation), energy, industry (industry, fugitive, solvents), ground transport (road transport, brake and tire wear, domestic shipping, other ground transport), residential, waste, agriculture (agricultural waste burning, agricultural livestock, agricultural crops).The original CAPSS inventory had a four-level classification system (upper, middle, lower, and detailed); however, for mapping to the HTAPv3 classification system, only the first two levels were used in this study.The emissions from the three countries were estimated using a standardized classification system for emission sources.

Air quality improvement policy
South Korea has implemented various air quality improvement policies to address air quality issues.117 businesses, which emitted air pollutants of more than 80 tons per year, and air pollutant emissions trading systems were implemented.In the second phase, total emission allowances were provided to more businesses than in the first.Air quality standards for PM 2.5 began to be applied in 2015 because of the continuously degrading air quality and more serious concerns about human damage caused by PM 2.5 .As the First Comprehensive Plan for Air Quality Improvement, the 10-year policy was completed in 2015, and the Second Comprehensive Plan for Air Quality Improvement was implemented in 2016.Under the Second Comprehensive Plan, there were additional standards for PM 2.5 .The stagnant PM 2.5 concentration drew higher public interest, leading to the announcement of Special Measures for Fine Dust Management.Comprehensive reinforced measures for fine dust management were announced in 2017.The implemented air pollution improvement policies indicate that Korea has been steadily promoting these policies.However, air quality improvement has been stagnant since 2013, and PM 2.5 concentrations have remained high since 2015, confirming that stricter air pollution improvement policies were implemented in 2016.In this regard, this study attempts to analyze the relationship between long-term emission patterns and the implementation of relevant policies.

Long-term emission trends in Korea
This section describes the long-term emission trends in Korea.In addition, this study investigated the relationship between these emission trends and past air pollutant policies in Korea by comparing changes in emissions during the period of policy implementation.Figures 2, 3, and 4 show NO x , SO x , and VOC emissions, respectively.In each graph, there are two solid lines to compare emission trends before and after the reanalysis (before: CAPSS, and after: CAPSS-KU), and the bar graph indicates the sectoral emissions of the CAPSS-KU emissions to examine emission changes per year and sector.At the top of the graph, we indicated the corresponding year to which the reanalysis method was applied and the application period.At the bottom of the graph, we present a table showing the year and period of policy implementation to identify emission changes according to the implementation time and policy period.
The graph of NO x emission trends (Fig. 2) shows the difference between CAPSS-KU and CAPSS up to 2016.NO x emissions mostly originated from road transport and energy sectors.The higher activity of the road transport sector from 2000 to 2004 led to higher NO x emissions, whereas the introduction of EURO-4 under the 2006 Comprehensive Plan for Air Quality Improvement resulted in lower emissions, confirming that the introduction of EURO-5 in 2009 and EURO-6 in 2014 had a direct impact on emission patterns.NO x emissions from the energy sector have decreased rapidly since 2007.This change can be attributed to the following.CAPSS's calculation method for emissions in the energy sector was based on emission factors and activity until 2006, and from 2007, CleanSYS was used to calculate measurement-based emissions.This confirms that the previous emissions calculation method resulted in an overestimation compared to the measurement-based emissions calculation.
The graph of SO x emission patterns (Fig. 3) shows the difference between CAPSS-KU and CAPSS up to 2015.The SO x emission trends are depicted in Fig. 3, which highlights the difference between the CAPSS-KU and CAPSS models until 2015.The graph shows that the biomass combustion emissions calculated since 2015 and the construction equipment sector emissions calculated since 2008, which were not considered in the original CAPSS model before those years, were included in the CAPSS-KU model emissions calculations for previous years.Specifically, biomass combustion emissions were included for the years 2000-2014, and construction equipment sector emissions were included for the years 2000-2007.This change in emissions calculations led to a difference in emissions data between the two inventories.
For the decrease in SO x emissions, the Special Act on the Improvement of Air Quality in Seoul Metropolitan Area was implemented in 2005 (ME, 2003), the First Comprehensive Plan for Air Quality Improvement (2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015) was implemented in 2006 (ME, 2005), and the Air Pollutant Emission-Cap Regulation in Seoul Metropolitan Area (First Phase) was implemented in 2007 (ME, 2018a, b); in particular, under the 2007 Clean Air Conservation Act (ME, 2007), clean fuels started to be more utilized through a notification on the use of clean fuel, and reinforced standards on limited SO x emissions by International Maritime Organization (IMO) in 2014 have contributed to gradual decreases in emissions since 2000 (EIA, 2018).SO x emissions mainly originate from the energy, industry, and domestic shipping sectors.Emissions from the energy sector decreased rapidly from 2004 to 2007.This decrease is attributable to the expansion of the lowsulfur fuel supply from 2004 to 2005.In 2007, similar to the aforementioned NO x emission trends, the emissions sharply decreased due to measurement-based emission calculations through the introduction of Clean-SYS.Emissions from the industry sector decreased from 2000 to 2004 and showed an increasing trend from 2004.Emissions from the domestic shipping sector steadily increased from 2000 and then rapidly decreased from 2014.This statement explains the change in how emissions were calculated in the domestic shipping sector after 2014.Specifically, the type of fuel used in ships was classified, and emissions were calculated based on the emission factors of the actual fuel used, such as diesel and gasoline, instead of using the emission factor of B-C oil, which has been used in the past.In addition, the impact of the strengthened sulfur oxide emission regulations of the IMO can be considered.
The graph of the VOC emission patterns (Fig. 4) shows the differences between CAPSS-KU and CAPSS up to 2015.The difference comes from the following conditions: emissions from the biomass burning sector, which started to be calculated in 2015, were applied to the period before 2014 and reanalyzed in CAPSS-KU.Furthermore, the emission factor of the solvent use sector from 2000 to 2006 was corrected after applying the emission factor change since 2007 in CAPSS-KU.
The solvent use sector accounted for 53% of the total VOC emissions in 2018.Additionally, emissions from the industrial sector have gradually increased since 2000.As Korea strived to reduce VOC emissions by 60% compared to the year 2000 emission level in the solvent use sector, during the period from 2000 to 2012, it implemented several regulations, such as the conversion of architectural paints to water paints, limited content of solvents, and setting standards for chimneys or fugitive emissions per facility feature (ME, 2005).However, it was found that CAPSS emission trends and CAPSS-KU emission trends increased slightly.In other words, although Korea implemented various policies to lower VOC emissions, no decreasing trend in emissions was achieved.
A comparison of the relationships between NO x and SO x emission trends and relevant policies showed that emissions decreased while implementing policies and measures.The NO x emission pattern was stagnant from 2008 to 2016; however, since 2016, it has decreased with the implementation of various air quality improvement policies.SO x emissions decreased continuously and showed a sharp decrease when the related policies were implemented.However, despite policies and measures to lower VOC emissions, there are no emission reduction patterns.Consequently, future air quality improvement policies should focus more on reducing VOC emissions than NO x and SO x emissions.
The long-term trends in emissions were compared with the long-term trends in the annual averages of observed concentrations in Korea.The observed concentration data were obtained from the urban air network data of the Korean Ministry of Environment (ME, 2018a, b).These data were collected from 523 measuring stations located in 161 provinces across Korea.NO 2 and SO 2 , which are primary pollutants, were selected for comparison because they are less influenced by transboundary precursor substances from neighboring countries.PM 2.5 and Ozone were excluded from the comparison due to their susceptibility to transboundary precursor substances from neighboring countries.As illustrated in Fig. 5, the concentration of NO 2 (represented by the dotted line) exhibits a gradual decrease from 2001 to 2018, closely resembling the decreasing trend in NO 2 emissions.Similar findings are presented for SO 2 in Fig. 5B.These results confirm that the long-term emissions data prepared in this study were effectively revised to accurately represent past emission conditions.

Inter-comparison of emission trends with neighborhood countries
The long-term emission trends in Korea from 2000 to 2018 and those in China and Japan were compared and analyzed in this study.Figures 6, 7, and 8 show graphs representing the HTAPv3 emission trends of  The period when air pollutants started to decrease was first found in Japan, followed by Korea and China.This tendency seemed to be attributable to differences in the economic growth of each country; Japan, which achieved the fastest economic growth among the three countries, saw a reduction in pollutants by first implementing air quality improvement policies.Korea and China followed these trends.However, China was the last country to show a reduction in pollutants among the three countries but a sharp fall in emissions, which seems to be due to the recently implemented stricter policies for air quality improvement.
Compared to combustion-related pollutants such as NO x and SO 2 , which are major precursors of secondary fine particles, the trend of VOC emissions shows a different pattern.In Japan, VOC emissions have been decreasing since 2000, whereas in Korea and China, they are still increasing.This trend suggests that Japan, which has experienced rapid economic growth, is actively working to reduce non-combustion-related pollutants, such as VOC, which are relatively more challenging to manage.Although Korea has policies to reduce VOC emissions, these reductions are not yet apparent.China has implemented strong policies to reduce combustion-related pollutants; however, the emission pattern of VOCs has not been reduced.
Figure 6 shows that NO x emissions have decreased recently in all three countries.However, it can be confirmed that different emission sources affect emission reductions in each country.For instance, in Japan, the road transport sector had the largest impact on emission reductions; as emissions sharply decreased after 2005, there were impacts on general trends.In Korea, the energy and road transport sectors impacted emission reductions, whereas in China, a sharp fall in emissions in the energy sector after 2011 led to a decrease in general NO x emissions.
The SO 2 emission trends in Fig. 7 show that all three countries exhibited decreasing tendencies.Japan and Korea showed a decreasing tendency before 2000, whereas China started to show a tendency after 2005.The figures indicate that the recent emissions reduction rate in China is much sharper than those in Japan and Korea.In Japan, the energy, industry, and domestic shipping sectors influenced SO 2 emission trends, whereas, in Korea, the energy and domestic shipping sectors had an impact.In China, emissions from the energy and industrial sectors have rapidly decreased, affecting general trends.
Finally, the VOC emission trends (Fig. 8) were confirmed to differ in the three countries.Japan experienced a decrease in VOC emissions, Korea experienced stagnant changes, and China experienced a sharp increase in emissions.Furthermore, in Japan, VOC emissions in the solvent sector decreased, whereas, in Korea, emissions in the solvent and industry sectors increased slightly.In China, emissions in the solvent sector have sharply increased, as have emissions in the industrial and fugitive sectors.According to Sugata et al. (2014), Japan has implemented regulations to mitigate the emission of VOCs across various sectors, including fuel combustion, waste incineration, outdoor burning in the agricultural sector, fuel evaporation, chemical manufacturing, and the use of solvents.Additionally, the Japanese government holds businesses accountable by enforcing regulations on VOC emissions if they fail to meet the stipulated criteria set by the Tokyo Metropolitan Government (2016).These criteria include (1) proper management of chemical substances, (2) distribution of information concerning chemical substances, and (3) reporting of the amount of properly managed chemical substances used.
When comparing the emission trends among the three countries, the emissions of NO x and SO 2 started to decrease at different periods but showed decreasing patterns in all three countries.However, Japan was the only country to experience a reduction in VOC emissions, while Korea exhibited a slight increase, and China exhibited a sharp increase.It is assumed that China and Korea should control VOC emissions, in addition to NO x and SO 2 emissions, which have been well controlled.

Conclusion
This study aimed to precisely analyze long-term emission trends in Korea.To this end, we utilized the 2018 emission calculation methodology of CAPSS and reanalyzed the emission data of CAPSS that were previously released from 2000 to 2018.The CAPSS generates an emissions inventory every year, with new emission sources being added and emission estimation methods being improved as needed.There are annual differences in the emission sources and calculation methods.To analyze long-term emission trends, emissions were recalculated using the same method to compare the emissions for each year.We utilized a bottom-up emission calculation formula to calculate the emissions, and the reanalyzed emission inventory was named CAPSS-KU.By recalculating emissions, it was possible to confirm the emission trends for pollutants from 2000 to 2018 in Korea, and there was a relationship between combustion-related pollutants, such as NO x and SO x , and air quality policies.It was also possible to confirm that the main emission source of NO x , which is road mobile sources, had fewer emissions, in line with stricter vehicle emission standards.The energy, industry, and shipping sectors, which are the main emission sources of SO x , had lower emissions owing to the increase in low-sulfur fuel supply and IMO's stricter SO x emission limit standards.In contrast, VOC, which was non-combustion-related pollutants, showed a steadily increasing trend in emissions from 2000 to 2018.It was also difficult to confirm the relationship between emission reductions and VOC reduction policies, such as the VOC emission reduction plan and implementation in the solvent use sector.
Furthermore, we converted the classification of the emission sources of CAPSS-KU into the HTAPv3 format and utilized the emission data of HTAPv3 to conduct a comparative analysis of emission trends among Korea and its neighboring countries, China and Japan.All three countries showed decreasing emissions of combustion-related pollutants, such as NO x and SO 2. However, Japan first showed a reduction tendency among the three countries, followed by Korea and then China.Even though China was the latest country to show emission reductions, it recently saw a sharp decrease in emissions, presumably due to the recently implemented stricter air quality improvement policies.VOC, a non-combustionrelated pollutant, showed a decreasing tendency only in Japan, whereas its emission change was stagnant in Korea, and China saw a rapid rise in emissions.In this sense, Japan's VOC reduction policies seem effective.Because VOC is a precursor material for generating secondary PM 2.5 , it is assumed that Japan's VOC reduction policy can be applied in Korea when implementing additional policies to efficiently and effectively reduce emissions for the future control of fine particulate matter concentrations.
The results of this study can provide useful information for future air quality improvement policies.However, certain aspects of this process require further investigation.In this study, a simple relationship between the reanalyzed emissions and air pollutant reduction policies in Korea was investigated.The reanalyzed emissions are data calculated independently from the air pollutant reduction policies.Although these data were independent of each other the emissions of NO x and SO x , which are combustion-related pollutants, showed a decreasing trend from the time the policy was implemented.However, VOC, a non-combustion-related pollutant, did not show a decrease in emissions, even though the policy had been implemented.When the connection between data that are not directly related is identified, it can be seen that NO x and SO x reduction policies affected emissions; however, policies implemented to reduce VOC had no effect or were insufficient in reducing VOC emissions.Hence, it is essential for future research to thoroughly examine past policies and their impact on decreasing combustion-related pollutants while identifying inadequacies in policies intended to reduce VOC.This study highlights the importance of conducting long-term emissions analyses to establish a connection between past policies and emissions trends, which is crucial for making informed decisions for the effective implementation and management of future air quality improvement policies.

Figure 1
Figure1shows the air quality improvement policies implemented since 2000 in chronological order(KEITI,  2020).In 2005, the Korean government implemented an air pollution control/management policy, the Special Act on the Improvement of Air Quality in the Seoul Metropolitan Area, targeting metropolitan areas with high pollution levels and severe damage from pollution due to dense populations.Under the Special Act, the Basic Plan on Metropolitan Area Air Quality Management and its Implementation Plan was established, and emission-cap regulations were introduced in places of businesses to control emitted pollutants from relevant industries.Furthermore, there was a greater supply of low-emission vehicles, reinforced control of vehicle emissions, and emission control from mobile sources.The First Comprehensive Plan for Air Quality Improvement, the 10-year air quality improvement policy, was implemented in 2006.The Plan was implemented only in five metropolitan cities with highly dense populations and in Gwangyang, where businesses were concentrated.Sectoral (e.g., industry and mobile pollution sources) countermeasures have been established and implemented to lower the emissions of targeted controlled pollutants, such as PM 10 , NO x , VOC, and SO x .The Air Pollutant Emission-Cap Regulation in Seoul Metropolitan Area was implemented in 2007 (phase 1) and 2009 (phase 2).In the first phase, total emission allowances for NO x and SO x were provided to first-class

Fig. 2
Fig. 2 NO x emission trends, reanalysis method, and timing of policy implementation

Fig. 3
Fig. 3 SO x emission trends, reanalysis method, and timing of policy implementation

Fig. 6 Fig. 7
Fig. 6 Comparison of NOx emission patterns via HTAPv3 (A China, B South Korea, C Japan)

Fig. 8
Fig. 8 Comparison of VOC emission patterns via HTAPv3 (A China, B South Korea, C Japan)

Table 2
Mapping with the HTAPv3 emission source classification system Fig. 1 Major Air quality improvement policies since 2000