Search

Since the 24-hr PM 2.5 (particle aerodynamic diameter less than 2.5 μm) concentration standard was regulated in Taiwan in 2012, the PM 2.5 concentration has been decreasing year by year, but the ozone (O 3 ) concentration remains almost the same. In particular, the daily maximum 8-hr average O 3 (MDA8 O 3 ) concentration frequently exceeds the standard. The goal of this study is to find a solution for reducing PM 2.5 and O 3 simultaneously by numerical modeling. After the Volatile Organic Compounds (VOC S )-limited and nitrogen oxides (NO X )-limited areas were defined in Taiwan, then, in total, 50 scenarios are simulated in this study. In terms of the average in Taiwan, the effect of VOC S emission reduction is better than that of NO X on the decrease in PM 2.5 concentration, when the same reduction proportion (20%, 40%) is implemented. While the effect of further NO X emission reduction (60%) will exceed that of VOC S . The decrease in PM 2.5 is proportional to the reduction in precursor emissions such as NO X , VOC S , sulfur dioxides (SO 2 ), and ammonia (NH 3 ). The lower reduction of NO X emission for whole Taiwan caused O 3 increases on average but higher reduction can ease the increase, which suggests the implement of NO X emission reductions must be cautious. When comparing administrative jurisdictions in terms of grids, districts/towns, and cities/counties, it was found that controlling NO X and VOC S at a finer spatial resolution of control units did not benefit the decrease in PM 2.5 but did benefit the decrease in O 3 . The enhanced O 3 control strategies obviously cause a higher decrease of O 3 throughout Taiwan due to NO X and VOC S emission changes when they are implemented in the right places. Finally, three sets of short-term and long-term goals of controlling PM 2.5 and O 3 simultaneously are drawn from the comprehensive rankings for all simulated scenarios, depending on whether PM 2.5 or O 3 control is more urgent. In principle, the short-term scenarios could be ordinary or enhanced version of O 3 decrease with lower NO X /VOC S emissions, while the long-term scenario is enhanced version of O 3 decrease plus high emission reductions for all precursors., (Copyright © 2022 Elsevier Ltd. All rights reserved.)

Reliable long-horizon PM 2.5 forecasts are crucial and beneficial for health protection through early warning against air pollution. However, the dynamic nature of air quality makes PM 2.5 forecasts at long horizons very challenging. This study proposed a novel machine learning-based model (MCNN-BP) that fused multiple convolutional neural networks (MCNN) with a back-propagation neural network (BPNN) for making spatiotemporal PM 2.5 forecasts for the next 72 h at 74 stations covering the whole Taiwan simultaneously. Model configuration involved an ensemble of massive hourly air quality and meteorological monitoring datasets and the existing publicly-available PM 2.5 simulated (forecasted) datasets from an atmospheric chemical transport (ACT) model. The proposed methodology collaboratively constructed two CNNs to mine the observed data (the past) and the forecasted data from ACT (the future) separately. The results showed that the MCNN-BP model could significantly improve the accuracy of spatiotemporal PM 2.5 forecasts and substantially reduce the forecast biases of the ACT model. We demonstrated that the proposed MCNN-BP model with effective feature extraction and good denoising ability could overcome the curse of dimensionality and offer satisfactory regional long-horizon PM 2.5 forecasts. Moreover, the MCNN-BP model has considerably shorter computational time (5 min) and lower computational load than the compute-intensive ACT model. The proposed approach hits a milestone in multi-site and multi-horizon forecasting, which significantly contributes to early warning against regional air pollution., (Copyright © 2022 Elsevier Ltd. All rights reserved.)

Long-range transport (LRT) reportedly carries air pollutants and microorganisms to downwind areas. LRT can be of various types, such as dust storm (DS) and frontal pollution (FP); however, studies comparing their effects on bioaerosols are lacking. This study evaluated the effect of LRT on viral and bacterial concentrations in Northern Taiwan. When LRT occurred and possibly affected Taiwan from August 2013 to April 2014, air samples (before, during, and after LRT) were collected in Cape Fugui (CF, Taiwan's northernmost point) and National Taiwan University (NTU). Reverse-transcription quantitative polymerase chain reaction (RT-qPCR) was applied to quantify influenza A virus. qPCR and qPCR coupled with propidium monoazide were, respectively, used to quantify total and viable bacteria. Types and occurrence of LRT were confirmed according to the changing patterns of meteorological factors and air pollution, air mass sources (HYSPLIT model), and satellite images. Two Asian DS and three FP cases were included in this study. Influenza A virus was detected only on days before and during FP occurred on January 3-5, 2014, with concentrations of 0.87 and 10.19 copies/m 3 , respectively. For bacteria, the increase in concentrations of total and viable cells during Asian DSs (17-19 and 25-29 November 2013) was found at CF only (from 3.13 to 3.40 and from 2.62 to 2.85 log copies/m 3 , respectively). However, bacterial levels at NTU and CF both increased during FP and lasted for 2 days after FP. In conclusion, LRT increased the levels of influenza A virus and bacteria in the ambient air of Northern Taiwan, particularly at CF. During and 2 days (at least) after LRT, people should avoid outdoor activities, especially in case of FP., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020. Published by Elsevier Ltd.)

Background: Evidence of the effects of long-term exposure to fine particulate matter (PM 2.5 ) air pollution on the development of dyslipidaemia is limited. This study aimed to investigate the association between long-term exposure to ambient PM 2.5 and incident dyslipidaemia in a large cohort., Methods: We studied 66,702 participants aged ≥18 years belonging to a cohort from a standard medical examination programme conducted in Taiwan between 2001 and 2014. The PM 2.5 concentration at each participant's address was estimated using a satellite-based spatiotemporal model at a high resolution (1 km 2 ). A time-varying Cox regression model was used to examine the association between long-term exposure to ambient PM 2.5 and the development of dyslipidaemia. Additionally, sensitivity analyses were conducted to examine the stability of these associations., Results: Compared with participants exposed to the 1st tertile of PM 2.5 , participants exposed to the 2nd and 3rd tertiles of PM 2.5 had respective hazard ratios of 1.02 [95% confidence interval (CI): 0.98-1.06] and 1.08 (95%CI: 1.04-1.13) for incident dyslipidaemia. Sensitivity analyses generally yielded similar results., Conclusion: Long-term exposure to ambient PM 2.5 is associated with a higher risk of dyslipidaemia. Global strategies for reducing air pollution are needed to prevent the development of dyslipidaemia., (Copyright © 2019 Elsevier Inc. All rights reserved.)

Daily and hourly average data from nine air-quality monitoring stations distributed across central Taiwan, which include ten items (i.e., PM₁₀, PM₂.₅, wind direction, wind speed, temperature, relative humidity, SO₂, NO₂, NO, and CO), were collected from 2005 to 2009. Four episode types: long-range transport with dust storms (DS), long-range transport with frontal pollution (FP), river dust (RD), and stagnant weather (SW), and one mixed type of episode were identified. Of these four episode types, the SW was the dominant type, averaging about 70%. The mean ratio of PM₂.₅/PM₁₀ was the lowest during the RD episodes (0.42), while the mean ratio of PM₂.₅/PM₁₀ was the highest during the SW episodes (0.64). Fine aerosol (PM₂.₅) and coarse aerosol (PM₁₀-₂.₅) samples were collected by high-volume samplers for chemical composition analysis, from only three stations (Douliou, Lunbei, and Siansi) during the days of SW, RD, DS, and FP. The concentrations of PM₂.₅ and three ionic species (NH₄⁺, NO₃⁻, and SO₄²⁻) all showed significant differences among the four episode types. The highest levels of NO₃⁻ (12.1 μg/m(3)) and SO₄²⁻ (20.5 μg/m(3)) were found during the SW and FP episodes, respectively. A comparison on the spatial similarity of aerosol compositions among the episodes and/or non-episodes (control) was characterized by the coefficient of divergence (CD). The results showed higher CD values in PM₁₀-₂.₅ than in PM₂.₅, and the CD values between RD episodes and the other three episodes were higher than those between two types of episode for the other three episodes. The ratios of SOR (sulfur oxidation ratio), SO₄²⁻/EC (elemental carbon), NOR (nitrogen oxidation ratio), and NO₃⁻/EC showed that sulfate formation was most rapid during the FP, while nitrate formation was most rapid during the SW.

Continuous measurements of ozone (O(3)) and carbon monoxide (CO) were carried out at Mei-Feng (24.05°N, 120.10°E, 2269 m above sea level), a remote mountain site in central Taiwan, to investigate the influence of long-range transported air pollution on O(3) and CO variations in the subtropical Pacific region. Data collected from March 2009 to September 2010 revealed average mixing ratios of 37±14 ppb for O(3) and 188±82 ppb for CO at this remote site. Diurnal variations for both O(3) and CO were observed as well in all seasons. The higher levels for O(3) and CO in the afternoon were attributed to transport of boundary layer pollution to the site during daytime upslope flow. Monthly means of both O(3) and CO showed maxima in spring and in the continental air masses from Southeast Asia, coastal China, and Korea/Japan. On the contrary, the lower O(3) and CO levels found in summer were due to the marine air masses originating from the Philippine Sea and Pacific Ocean. The relationship between O(3) and CO was analyzed, using nighttime data to minimize any local influence. The results showed a fairly good correlation between O(3) and CO from March to September. The contribution of CO from the Asian outflow reached a maximum in spring (88 ppb) and had a minimum in summer (27 ppb). The photochemical buildup of O(3) resulting from anthropogenic emissions in continental Asia was estimated to be 15 ppb in spring, while its production was insignificant, with an average of 4 ppb, in summer. A positive correlation between O(3) and CO plus high ozone levels in springtime suggested that the enhancements of O(3) were likely due to O(3) which was photochemically produced over this region., (Copyright © 2011 Elsevier B.V. All rights reserved.)

A catastrophic earthquake, namely the 921-earthquake, occurred with a magnitude of M(L)=7.3 in Taiwan on September 21, 1999, causing severe disaster. The evaluation of real-time air-quality data, obtained by the Taiwan Environmental Protection Administration (EPA), revealed a staggering increase in ambient SO(2) concentrations by more than one order of magnitude across the island several hours prior to the earthquake, particularly at background stations. The abrupt increase in SO(2) concentrations likely resulted from seismic-triggered degassing instead of air pollution. An additional case of a large earthquake (M(L)=6.8), occurring on March 31, 2002, was examined to confirm our observations of significantly enhanced SO(2) concentrations in ambient air prior to large earthquakes. The coincidence between large earthquakes and increases in trace gases during the pre-quake period (several hours) indicates the potential of employing air-quality monitoring data to forecast catastrophic earthquakes., (Copyright 2010 Elsevier B.V. All rights reserved.)

Understanding temporal coherence in particulate matter (PM) pollution across East Asian outflow regions is crucial for collaborative efforts in addressing air pollution challenges. In this study, we analyze daily PM exceedance patterns using ground-based measurements over the past two decades, focusing on the influence of El Niño-Southern Oscillation (ENSO) and Asian dust activity. ENSO-related coherent PM exceedance shows lower occurrence during El Niño compared to La Niña in spring, due to weakened continental outflows and increased precipitation. Temporal distribution of dust-related coherent PM exceedance closely mirrors springtime dust activity in source regions, indicating significant dust emission impact on outflow regions. Our findings have implications for climate variability and the assessment of multi-decade model simulations. [ABSTRACT FROM AUTHOR]

In addition to the Taiwan Strait, the geographical channel effect (GCE) has been observed in the Taipei Basin during winters. This study explored the potential links amongst GCE phenomena in the basin, fine particulate matter (PM2.5) events, and position of the continental high-pressure system (CHPS). Principal component analysis was applied to classify the CHPS distribution patterns influencing PM2.5 concentrations and meteorological parameters in the Taipei metropolis. In addition, non-parametric Kruskal–Wallis and Dwass–Steel–Critchlow–Fligner tests were applied to compare the parameters amongst distinct groups. Moreover, track analysis was utilised to trace the trajectories of air masses. The merged Dark Target Aerosol Angstrom Exponent (Ocean), Deep Blue Aerosol Angstrom Exponent (Land), and Aerosol Optical Depth (AOD) 3 km (Land and Ocean) of Terra/MODIS and Aqua/MODIS were used to confirm the connection of long-range PM2.5 transport. Two features in the atmospheric environment induce double GCEs: (1) wind direction below 90° in most of the upwind area of the Taisumi and Xindian River Basins; (2) CHPS centres moved across 110° E. Double GCEs can reduce the magnitude of PM2.5 air pollution and shape the distribution of PM2.5; however, they cannot prevent PM2.5 events. [ABSTRACT FROM AUTHOR]

Highlights: [•] Winter monsoon not only brings cold air but also transports air pollutants to Taiwan. [•] Biomass burnings are important dioxin emission sources in Southeast Asia. [•] Typhoon events have a documented impact on the remobilization of dioxins. [Copyright &y& Elsevier]

Abstract: Surface ozone (O3) and carbon monoxide (CO) mixing ratios were measured at Mei-Feng (24.05 °N, 120.10 °E, 2269 m above sea level) remote mountain site between March 2009 and September 2010 to investigate the impact of regional pollution on O3 and CO. The results showed that the maximum values of both O3 and CO were found in the springtime. Backward trajectory analysis, combined with MODIS fire spots suggested that the enhanced O3 and CO in springtime could be attributed to biomass burning (BB) activities over Southeast (SE) Asia. Thirteen BB events were identified by backward trajectory analysis, MODIS fires, NCEP weather data sets and CO concentrations. Good correlation between O3 and CO was found during the BB plumes. Using the linear regression, the slope (ΔO3/ΔCO) was calculated to be 0.18 ± 0.08 (mean ± 1σ). This value was in agreement with that of 0.2 observed over the west Pacific region during the TRACE-P campaign, but was higher than those (0.11–0.14) of Canadian and Siberian fires. Moreover, significant enhanced O3 productivity was also found in aged BB plumes and that mixed with urban emissions from SE coastal China. To assess the net influence of SE Asian BB, the air masses from SE Asia and SE China were divided in two groups: those that passed over the fire regions (PF) and those that did not (NP). The result showed that the maximum differences between PF and NP were estimated in March with 8 ppb for O3 and 45 ppb for CO, respectively, accounting for 23% of both CO and O3 levels at Mei-Feng. Although uncertainties existed in the estimations, the significant discrepancies of O3 and CO in the two air groups suggested the air pollutants emitted by SE Asian BB could be transported and influence the air quality over subtropical Taiwan in springtime. [Copyright &y& Elsevier]

Abstract: Carbon monoxide (CO) mixing ratios were observed from 30 January to 7 April 2008 at Mt. Lulin (23.51°N, 120.92°E, 2862masl) in central Taiwan to investigate characteristics of CO during biomass burning periods. During the sampling campaign, the average mixing ratio of CO was 234±63ppb with higher levels observed in March. The elevated CO in March can, on the basis of backward trajectories and satellite fire spots analyses, possibly be attributed to biomass burning activities in the Asian continent. Significant diurnal variations of CO mixing ratios were observed at the remote site. The higher CO levels in the afternoon were influenced by the transport of boundary layer pollution to the site during daytime upslope flow. Backward trajectory analysis showed that air masses mainly originated from India (ID), the Indochina Peninsula (IP) and South Coastal China (SC), which together accounted for 85% of the total trajectories. Higher mixing ratios of CO were found in the ID, IP, and SC categories, indicating significant impacts of anthropogenic emissions on the Pacific region. Furthermore, the air parcels were divided into two categories, those that passed over the fire regions and those that did not. The result showed that the average difference of CO levels between the two categories was approximately 79ppb, suggesting that Asian biomass burning plays an important role in CO levels at this remote site during the springtime. [Copyright &y& Elsevier]

Abstract: The impact of long-range transport on the air quality of Taiwan was assessed by using measurements at stations of the air quality monitoring network of the Taiwan Environment Protection Administration over the winter monsoon periods of 2000 and 2001. The primary long-range transport process can be clearly identified to be the prevailing northeasterly following the cold front of winter monsoon that originates in the Asian continent. Based on pollutants concentration differences among different meteorological conditions, we estimate that the long-range transport of particulate pollutants contributes to about 30μgm−3 to the PM10 concentrations in northern and eastern Taiwan. A smaller contribution is estimated for the western plain of Taiwan. Contributions of the long-range transport to CO and SO2 are about 230 and 0.5ppb, respectively. The identified, dust events have the highest average inflow concentration (about 71±34μgm−3) of PM10. The dust impacts on PM10 have been estimated to be nearly 100% at windward background stations along the north and northeast coast. In the northern cities of Taiwan, such as Taipei and Hsinchu, the impacts decrease to around 60–80% as local emission increase. Due to geographic blocking, a smaller impact is estimated for the western plain of Taiwan. However, the identified frontal pollution cases have significant impact on CO and SO2, which is more evident than the impact of dust events. In general, impacts of long-range transport on air pollutants at coastal stations are greater than those in urban areas. [Copyright &y& Elsevier]

Abstract: In recent years, the high levels of surface ozone have become a major air pollutant in Taiwan. Over the western plain of Taiwan, where most of the 23 million inhabitants live, the concentration of ozone increased substantially during the last ten years (1994–2003). This increase is puzzling as the emissions of ozone precursors, oxides of nitrogen (NO x ) and volatile organic compounds (VOCs) or non-methane hydrocarbon (NMHC) decreased significantly in the same period. An in-depth analysis focusing on “total oxidant” (estimated by O3+NO2) showed that its concentration remained essentially constant in Taipei, Taiwan, during that same ten year time period. Given that the regional tropospheric ozone is increasing, the constant oxidant level suggests that the oxidant production from photochemical reactions in Taipei should have been reduced to some extent, due to the reduction in emissions of ozone precursors. Consequently, the increasing trend in ozone was a result of the reduced titration by NO. These results are important to the formulation of ozone control strategies, especially for newly developed countries like Taiwan. [Copyright &y& Elsevier]

Abstract: Ambient air samples were collected at 13 air quality monitoring stations in Kaohsiung city, Kaohsiung county, and Pingtung county (KKP) to investigate the composition and spatial distribution of C2–C10 non-methane hydrocarbons (NMHCs) in southern Taiwan. Ozone formation potentials (OFPs) of NMHCs were estimated using maximum incremental reactivity (MIR) and k OH method (reactivity of NMHC with OH radical) to assess the relative effects of hydrocarbons on ozone formation. The measurements showed that mixing ratios of toluene, ethene, ethyne, ethane, isopantane and propane were the highest among all measured species at most of the sampling sites. Nevertheless, considering both the photochemical reactivities and mixing ratios of all the measured species, toluene, xylene, ethene and propene were calculated to have the highest OFPs and reactivities. The OFPs and reactivities assessed by the MIR and k OH methods for the four compounds accounted for 54.5% and 39.3% of all the measured species. Larger benefit margin of ozone abatement may be obtained by reducing emissions of a group of key species with high OFPs. 2,2-dimethylbutane (22DMC4) was used as an indicator of traffic emissions to distinguish traffic from non-traffic contributions of key species in Kaohsiung metropolitan area. It revealed that the contribution of non-traffic source was significant for toluene, whereas xylene was found to be primarily from the traffic source in Kaohsiung metropolitan area during the sampling periods. [Copyright &y& Elsevier]

Abstract Anthropogenic climate change has been increasingly confirmed by weather observation and scientific literature in recent decades. Atmospheric stability, which has strong effects on vertical mixing of air pollutants and thus air quality, may be affected under climate change. This study aims to statistically assess the impacts of climate change alone on the future air quality in the Pearl River Delta region in the near future (2030–2039) and the far future (2090–2099) under two Representative Concentration Pathways (RCP) scenarios, i.e., RCP 4.5 and RCP 8.5, based on the future surface and upper level meteorological data projected by one regional climate model (RCM): WRF, and by four general circulation models (GCMs): CanESM2, MIROC, MRI-CGCM3 and MPI-ESM-LR. The arithmetic means of projections reveal an increase in the levels of air pollutants [ozone (O 3), respirable suspended particulates (RSP) and sulphur dioxide (SO 2)] in various seasons, even though a decrease is projected to occur in June-July-August. These changes in projected mean concentration are more significant in the far future, and under the RCP8.5 scenario. Among difference meteorological variables, surface temperature is most associated with the projected change in the three pollutants, with a range from 56.9% to 65.2% in all seasons and for all pollutants, relative to all contributions in RCP8.5 for example. Other notable associations include positive effects of vertical temperature gradient and the temperature–dew point difference on pollutant concentration. We found an increase in frequency of high pollution levels in December–January–February and March–April–May, as the occurrence proportion of pollutant concentration greater than the recent 95th percentile is 9.5%–9.6% and 6.4%–9.2%, respectively. We conclude that climate change alone is projected to have significant effect on air quality in the Pearl River Delta region in future, implying the necessity of more stringent air pollutant emission control policies to mitigate air pollution in the future. Highlights • We estimated the sole impact of climate change on air quality through vertical atmospheric conditions over the PRD region. • We projected a rise of mean pollutant concentration in MAM and DJF for all scenarios, but a tendency to decline in JJA. • High pollution conditions are projected a more frequent occurrence in DJF for all scenarios and species. • Besides surface temperature, other meteorological variables contribute evenly along heights to the future air quality change. [ABSTRACT FROM AUTHOR]

Abstract: To investigate the size resolved physico-chemical properties of aerosols transported from the eastern Asian Continent following the winter monsoons, an intensive field experiment was conducted in Taipei, Taiwan in January 2003. Mass concentration and the abundance of major functional groups in size-segregated aerosol samples were measured. The size distribution parameters of the aerosol mass and the associated functional groups were analyzed. The results of the infrared spectroscopic analysis showed that the contents of sulfate and ammonium in Taipei''s sub-micron aerosols were significantly increased due to continental pollution. In addition, the Asian outflow aerosols contained higher levels of silicate and nitrate in the super-micron size range. The size-resolved functionalities of the aerosol samples collected during Asian outflow episodes were compared with those of local pollution samples. This comparison revealed that ammonium and sulfate generally coexisted in sub-micron aerosols, suggesting that there are substantial emissions of ammonia to neutralize the acidity of sulfate in the region from eastern China to northern Taiwan. The nitrate had a bimodal distribution over the size range between 0.1 and 10 μm. A significant enhancement was found in coarse nitrate during the Asian outflow episodes, implying heterogeneous reactions of nitric acid on the surface of dust and sea salt particles during the transport processes. High concentrations of sub-micron ammonium nitrate aerosols were observed prior to the frontal passage and served as an important indicator of local photochemical pollution. The aliphatic and carbonyl carbon in the aerosols were suggested as being mostly due to local pollution in Taipei. Nevertheless, our analysis found that the aerosols of Asian outflows had a higher infrared absorbance ratio of carbonyl to aliphatic carbon than those of the local pollution. The enrichment of carbonyl carbon in the Asian outflow aerosols is most likely due to the emissions of biomass burning aerosols in the eastern China. [Copyright &y& Elsevier]