key: cord-271602-f14wccj3 authors: Kim, Hyun Cheol; Kim, Soontae; Kim, Byeong-Uk; Jin, Chun-Sil; Hong, Songyou; Park, Rokjin; Son, Seok-Woo; Bae, Changhan; Bae, MinAh; Song, Chang-Keun; Stein, Ariel title: Recent increase of surface particulate matter concentrations in the Seoul Metropolitan Area, Korea date: 2017-07-05 journal: Sci Rep DOI: 10.1038/s41598-017-05092-8 sha: doc_id: 271602 cord_uid: f14wccj3 Recent changes of surface particulate matter (PM) concentration in the Seoul Metropolitan Area (SMA), South Korea, are puzzling. The long-term trend of surface PM concentration in the SMA declined in the 2000s, but since 2012 its concentrations have tended to incline, which is coincident with frequent severe hazes in South Korea. This increase puts the Korean government’s emission reduction efforts in jeopardy. This study reports that interannual variation of surface PM concentration in South Korea is closely linked with the interannual variations of wind speed. A 12-year (2004–2015) regional air quality simulation was conducted over East Asia (27-km) and over South Korea (9-km) to assess the impact of meteorology under constant anthropogenic emissions. Simulated PM concentrations show a strong negative correlation (i.e. R = −0.86) with regional wind speed, implying that reduced regional ventilation is likely associated with more stagnant conditions that cause severe pollutant episodes in South Korea. We conclude that the current PM concentration trend in South Korea is a combination of long-term decline by emission control efforts and short-term fluctuation of regional wind speed interannual variability. When the meteorology-driven variations are removed, PM concentrations in South Korea have declined continuously even after 2012. particulate matter (PM; PM 10 , particles with a diameter of 10 µm or less, is used in the study unless specified) concentration is its interannual trend, which recently showed a signal of increase after a long downward trend. Since the early 2000s, annual mean surface PM concentrations have decreased continuously over the SMA and South Korea (Fig. 1 ). The annual mean concentration was 71.9 μg/m 3 in 2002, and continued to decrease until it reached its minimum (46.9 μg/m 3 ) in 2012, but since then it has begun to rise (51.2 μg/m 3 , 51.9 μg/m 3 , 51.0 μg/ m 3 in 2013, 2014, and 2015, respectively). Although several reasons (e.g., regional transport changes in China 12, 13 or increased use of diesel-engine passenger cars) 14 have been suggested for the recent changes, the potential of meteorological variance has not been actively discussed until now. In addition to the changes in the amount of released emissions of pollutants and their precursors, meteorological conditions are also primary contributors to regional haze and air pollutions 15, 16 . Wind, in its velocity and transport pattern, is one of the major meteorological components that affects surface air pollution 17, 18 . Winds can control vertical mixing and regional ventilation 19 . Persistent stagnant conditions (i.e. calm wind) provide critical conditions that lead to the development of local pollution episodes. Wind direction controls the source-receptor relationship 20 , and it also directly initiates local emissions such as dust storms or sea salt emissions 21, 22 . Large scale transport, both regional or inter-continental, can affect the atmospheric composition by altering their lifetime (e.g. longer lifetime at higher altitudes), and influences global meteorology and climate 23, 24 . Identifying the major drivers in the long-term trend of surface PM concentration can be limited due to the intricacy of chemical and meteorological processes. A 12-year simulation with a constant anthropogenic emission inventory was designed to isolate the interannual variation of surface PM concentrations solely due to variations of meteorology (See method). After testing multiple meteorological variables, we found a significant association of wind speed and surface PM concentrations. Normalized anomalies (annual averages subtracted and then divided by the 2004-2015 average) of surface PM concentration and 10-m wind speed averaged over three geographical coverages (i.e. 9-km domain-wide, Korea, and the SMA) are compared (Fig. 2) . Noticeably, interannual variations of modeled surface PM concentrations show very similar patterns in all regions. The observed PM, averaged over 247 sites in South Korea, shows similar but slightly different interannual variability. Its year-to-year variation pattern is very similar to that of modeled variation; a small drop in 2004, strong positives in 2005-2007, strong negative in 2012, and an increase in 2013-2014. Its general pattern, however, is mostly positive in early years and negative in later years. It thus seems reasonable to say that the observations show mixed signals; likely a combination of a short-term year-to-year fluctuations (caused by meteorological changes) and a long-term decline (caused by changes in anthropogenic emissions) 25 . Normalized anomalies of 10-m wind speed, on the other hand, show good agreement between observations (averaged over 79 sites in South Korea) and modeled wind speed over three regions, demonstrating that the model reasonably simulated interannual variations of the wind field. The key feature from the interannual variations of surface PM concentrations and surface wind is that they show totally opposite phases in their interannual variabilities. It should be noted that the interannual variations of PM concentrations, in all three regions, are best explained by the changes of domain-wide averaged wind speed compared with smaller scale averages. This might imply that the controlling mechanism is not on a local scale (e.g. sea breeze or local circulation) but rather synoptic scale activities (e.g. frontal passage). Considering the efficiency of the wipe-out mechanism by cold frontal activities, the annual wind speed variability seems to represent the efficiency of regional ventilation 26 . Northeastern Asia is a routine passage of mid-latitude cyclogenesis, initiated on the lee side of the Altai-Sayan Mountains 27 . The role of Altai-Sayan cyclogenesis, as an efficient cleaning mechanism of pollutants over northern China and Korea during cool season, is demonstrated and discussed in Kim et al. 28 . The change of summertime circulation (i.e. East Asia summer monsoon) on regional pollution is also discussed in Zhu et al. 29 . Scatter plots of normalized anomalies between wind speed and surface PM concentration from the model ( −10.9% and 7.7%, while the wind speed anomaly varied between −3.5% and 4.8%. The range of variability in PM concentration is impressive since it is almost compatible to total long-term changes; that is ~−2% per year or ~−24% during 2004-2015. Remarkably, even with excellent agreement between modeled and observed interannual anomalies in surface PM concentration and wind speed, a direct comparison between observed PM concentration and observed wind speed variability shows poor correlation (i.e. R = 0.07). It clearly shows the advantage of the emission-isolating modeling approach, otherwise the meteorology-driven interannual variability of PM concentration seems to be overshadowed by emission reduction in the long-term trends. Finally, we tried to remove the meteorology-driven variances from the original PM concentrations. Fractional interannual anomalies from Fig. 2A are multiplied to an overall mean of PM concentrations from each region, and then subtracted from the original annual mean concentrations. Considering the uncertainties from the original model bias, using the relative variances instead of absolute concentrations from the model is a more robust approach 30 . Adjusted annual PM concentrations, with meteorology-driven variances removed, are shown in Fig. 3 in relation to Seoul, the SMA, and South Korea. The most surprising change of the annual PM concentrations, compared with those of Fig. 1 , is their trend in recent years. The increase since 2012, which had frustrated Apparently, the decreasing rate in Seoul is faster than the national average, implying the effectiveness of special emission control efforts in the locale, including replacing diesel buses and diesel garbage trucks with natural gas vehicles 31, 32 . The local minimum of 2008 in the adjusted PM concentrations is noteworthy. The most likely explanation for this decrease is the impact of the global recession during 2008-2009, as indicated by several previous researches 33, 34 . At this point, we are not able to distinguish if the impact is due to changes caused by the South Korean economy's experience of the global recession, or due to changes of transported pollutants and precursors by Chinese economic changes. The case of 2015 is also interesting. PM concentration was higher in early 2015 (64.1 μg/m 3 in 2014 Jan-Mar and 72.5 μg/m 3 in 2015 Jan-Mar), but the annual mean is lower than that of 2014 because its Spring-to-Fall concentration is very low, even with the weakest wind speed observed. This is likely an extraordinary case by the impact of the Middle East Respiratory Syndrome (MERS), which has seriously impacted the South Korean economy and society 35 . Although further investigations are necessary, these cases in 2008 and 2015 may provide interesting examples of the socio-economic impact on the environment, in addition to emission control policy-driven impact. Three factors, long-term trend by emission control, short-term variation by meteorology, and sporadic offsets by unexpected social or economic episodes, seem to be affecting PM concentrations in South Korea. To conclude, we found that there is a strong correlation between variations of surface PM concentrations in South Korea and variations of wind speed. This study addresses several important implications. First, interannual variation of meteorological conditions has the strong potential to affect long-term trends in surface PM pollution events. At this point, we do not have any evidence that the reduced wind speed is part of long-term change or just short-term fluctuation in general circulations or ventilation patterns. However, it is worth further investigation since the frequency of mid-latitude cyclones can be reduced in warmer weather conditions, as suggested through observations [36] [37] [38] [39] and models studies 19, 40-42 . Mickley et al. 19 used the Goddard Institute of Space Studies model to demonstrate that the severity and duration of summertime regional pollution can increase due to a decline in the frequency of mid-latitude cyclones under future warmer climates. This may provide a key feature to understand the frequent occurrence of severe haze episodes not only in South Korea but also in East Asia, especially in northern China. Yang et al. 43 also demonstrated haze days over eastern China has increased between 1980 and 2014, and argued that the weakening of winds is the dominant factor leading the decadal increase. Second, accurate simulation of surface wind field, especially in terms of wind speed, seems to be critical for an accurate regional air quality model and forecast. Analysis suggests a 5% change in annual wind speed is associated with 10% annual surface PM concentration. In many air forecast simulations in South Korea, including the current long-term simulation used in this study, modeled surface winds tend to be overestimated (2. Third, most importantly, emission control efforts from the South Korean government and community, as well as neighboring countries, seem to be effective. However, changes in meteorological conditions seem to offset those efforts. Recent space-borne observations also confirm a considerable decrease of NO 2 and SO 2 vertical column densities over the SMA, South Korea, and northeastern China (i.e. the Beijing-Tianjin-Hebei region) 44, 45 . In order to investigate the interannual variation of surface PM concentration in the SMA, we conducted a 12-year simulation using a regional air quality modeling system. The Weather Research and Forecasting Model (WRF), the Sparse Matrix Operator Kernel Emission (SMOKE), and the Community Multiscale Air Quality (CMAQ) models were utilized to simulate meteorology and chemistry over East Asia (27- 52 was used to prepare biogenic emissions. MCIP-processed ground reaching solar radiation and 2-m temperature were used to adjust hourly biogenic emissions such as isoprene and monoterpenes. Other vegetation data like leaf area index, plant function type, and emission factors were also used as released with MEGAN v2.04. Biomass burning and dust emissions were not included. The modeling configuration was tested for multiple years as part of the Integrated Multi-Scale Air Quality Study for Korea (IMAQS-K) system, which was initially developed as a prototype of the official national air quality forecast system in South Korea. The IMAQS-K system has been operational since 2012. Descriptions on physical options are provided in Kim et al. 53 , and basic model performance evaluations for 12-year period are provided in the supplementary materials. Model performed well in reproducing spatial and temporal variations of meteorology and chemical components. While total particulate matter concentration generally underestimates in the model, current analysis, using percent change of inter-annual variation, is designed to be less affected by the model bias. Surface monitoring data. Hourly observations of surface PM concentrations and wind speed were obtained from NIER and KMA, respectively. PM measurements, based on the beta-ray absorption method 54 , were collected from 247 urban air monitoring network sites, and wind data were collected from 79 meteorological monitoring sites. Locations of monitors are shown in the supplementary material. Note on other controlling factors. (1) A recent study on the regional emission attribution to the South Korea suggests that the relative attribution of foreign emission sources is not sensitive to inter-annual variation of meteorology 55 , contributing around 60% of the SMA PM concentration. Combined with the current declining trend of anthropogenic emissions from China, we can rule out the possibility of the increased efficiency of Chinese emissions transport. (2) The number of diesel vehicles in South Korea has increased 13% from 2012 to 2014. Since non-truck diesel vehicles emit 5.6% of total NO x emission in South Korea 5 , resultant impact is small (<1% of total PM). However, careful and continuous monitoring of diesel vehicle emissions is necessary due to their high uncertainty in emissions factors as we have learned from the Volkswagen emission scandal 56 . Exploring the severe winter haze in Beijing: the impact of synoptic weather, regional transport and heterogeneous reactions Characterization of submicron aerosols during a month of serious pollution in Beijing Possible influence of atmospheric circulations on winter haze pollution in the Beijing-Tianjin-Hebei region, northern China Air quality modeling in East Asia: present issues and future directions Korean National Emissions Inventory System and Effects of the Asian Dust Events on Daily Mortality in Seoul A review on East Asian dust storm climate, modelling and monitoring The chemical composition of fine and coarse particles in relation with the Asian Dust events Impact of Siberian forest fires on the atmosphere over the Korean Peninsula during summer Effects of Siberian forest fires on air quality in East Asia during May 2003 and its climate implication National reporting guidelines for CSD-14/15 Thematic areas -Atmosphere/Air Pollution, Republic of Korea Long-range transport of air pollutants originating in China: A possible major cause of multi-day high-PM10 episodes during cold season in Seoul Influence of transboundary air pollutants from China on the high-PM10 episode in Seoul, Korea for the period Demand forecasting of diesel passenger car considering consumer preference and government regulation in South Korea Synoptic and mesoscale weather conditions during air pollution episodes in Does weather confound or modify the association of particulate air pollution with mortality? Review: Untangling the influence of air-mass history in interpreting observed atmospheric composition Simulation of the interannual variations of aerosols in China: role of variations in meteorological parameters Effects of future climate change on regional air pollution episodes in the United States Seasonal Source-Receptor Relationships in Asia Oceanographic Sciences Library Modeling the atmospheric dust cycle: 1. Design of a soil-derived dust emission scheme Assessing the effects of anthropogenic aerosols on Pacific storm track using a multiscale global climate model The role of mineral-dust aerosols in polar temperature amplification Decadal trend and interannual variation of outflow of aerosols from East Asia: Roles of variations in meteorological parameters and emissions Effect of climate change on air quality Synoptic Climatology of Cyclogenesis over East Asia Synoptic perspectives on pollutant transport patterns observed by satellites over East Asia: Case studies with a conceptual model Increases in aerosol concentrations over eastern China due to the decadal-scale weakening of the East Asian summer monsoon Verification, validation, and confirmation of numerical models in the Earth sciences Co-benefit analysis of an air quality management plan and greenhouse gas reduction strategies in the Seoul metropolitan area Natural Gas Vehicles Promotion Program in Urban Areas Trends in OMI NO2 observations over the US: effects of emission control technology and the economic recession Reductions in nitrogen oxides over Europe driven by environmental policy and economic recession Middle East Respiratory Syndrome-advancing the public health and research agenda on MERS-lessons from the South Korea outbreak Long-term trend and spatiotemporal variations of haze over China by satellite observations from 1979 to 2013 Trends in Cyclone and Anticyclone Frequency and Comparison with Periods of Warming and Cooling over the Northern Hemisphere Multidecadal global and regional trends in 1000 mb and 500 mb cyclone frequencies Trends in Northern Hemisphere Surface Cyclone Frequency and Intensity Changes in mid-latitude variability due to increasing greenhouse gases and sulphate aerosols Possible Change of Extratropical Cyclone Activity due to Enhanced Greenhouse Gases and Sulfate Aerosols-Study with a High-Resolution AGCM Model-Simulated Northern Winter Cyclone and Anticyclone Activity under a Greenhouse Warming Scenario Increase in winter haze over eastern China in recent decades: Roles of variations in meteorological parameters and anthropogenic emissions Aura OMI observations of regional SO 2 and NO 2 pollution changes from A space-based, high-resolution view of notable changes in urban NO x pollution around the world A time-split nonhydrostatic atmospheric model for weather research and forecasting applications Research Data Archive at the National Center for Atmospheric Research Review of the Governing Equations, Computational Algorithms, and Other Components of the Models-3 Community Multiscale Air Quality (CMAQ) Modeling System Model representation of secondary organic aerosol in CMAQv4 Documentation of the SAPRC-99 Chemical Mechanism for Asian emissions in 2006 for the NASA INTEX-B mission Estimates of global terrestrial isoprene emissions using MEGAN (Model of Emissions of Gases and Aerosols from Nature) Influence of fossil-fuel power plant emissions on the surface fine particulate matter in the Seoul Capital Area Measurement Methods to Determine Compliance with Ambient Air Quality Standards for Suspended Particles Regional Contributions to Particulate Matter Concentration in the Seoul Metropolitan Area, Korea: Seasonal Variation and Sensitivity to Meteorology and Emissions Inventory Impact of the Volkswagen emissions control defeat device on US public health H.C.K. and S.K. planned the research, S.K., C.B., M.B. performed simulations. H.C.K. wrote the paper. B.U.K., R.P. and A.S. provided discussion and comments on chemistry. C.S.K., S.H., S.W.S., and C.K.S. provided discussions and comments on meteorological aspect. Supplementary information accompanies this paper at doi:10.1038/s41598-017-05092-8Competing Interests: The authors declare that they have no competing interests.Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.