Search

In the winter and spring times, the cold continental airs not only often outflow to Taiwan, but also bring air pollutants and dusts. Moreover, springtime happens to be the biomass burning season in Indochina. Under favor weather conditions, the products of biomass burning pollutants could be transported easily to Taiwan and even East Asia. Due to the unique geographic location of Taiwan, that allows us to receive more than just one source in spring. Actually, the complex interactions of these air pollutants and aerosols features in the boundary layer and aloft have resulted in complex characteristics of air pollutants and aerosols distributions in the lower troposphere over Taiwan and East Asia. The impacts are also essential and complex. The project "Effect of Megacities on the transport and transformation of pollutants on the Regional and Global scales (EMeRGe)" aims to improve our knowledge and prediction of the transport and transformation patterns of European and Asian megacities pollutant outflows. In EMeRGe Asia, the composition of the plumes of pollution entering and leaving Asia measured by the new High Altitude and LOng Range (HALO) aircraft research platform. The HALO aircraft performing optimized transects and vertical profiling in Asia during 12 March and 7 April in 2018. To design the measurement of aircraft flight paths and elevations, a high resolution, 9 km, numerical prediction by Weather Research Forecast (WRF) and WRF-Chem models were joined and performed during the campaigns. The EMeRGe Asia has successfully finished more than 10 missions during study period. Model performances and preliminary results will discuss in this meeting. Overall, this series of studies significantly fill the gap of our understanding on air pollutants transformation and transport to Taiwan and East Asia, and show the potential directions of future studies. [ABSTRACT FROM AUTHOR]

Particulate lead (Pb) is a primary air pollutant that affects society because of its health impacts. This study investigates the source sectors of Pb associated with ambient fine particulate matter (PM2.5) over central-western Taiwan (CWT) with new constraints on the Pb-isotopic composition. We demonstrate that the contribution of coal-fired facilities is overwhelming, which is estimated to reach 35 ± 16% in the summertime and is enhanced to 57 ± 24% during the winter monsoon seasons. Moreover, fossil-fuel vehicles remain a major source of atmospheric Pb, which accounts for 12 ± 5%, despite the current absence of a leaded gasoline supply. Significant seasonal and geographical variations in the Pb-isotopic composition are revealed, which suggest that the impact of East Asian (EA) pollution outflows is important in north CWT and drastically declines toward the south. We estimate the average contribution of EA outflows as accounting for 35 ± 15% (3.6 ± 1.5 ng/m3) of the atmospheric Pb loading in CWT during the winter monsoon seasons. [ABSTRACT FROM AUTHOR]

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. 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 km2). 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. 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. 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. • This is the largest longitudinal cohort study consisting of 66,702 adults. • The mean PM 2.5 was 27.0 μg/m3 and the interquartile range was 9.6 μg/m3 with a range of 5.8–50.3 μg/m3 • Ambient PM 2.5 is associated with a higher risk of dyslipidaemia. [ABSTRACT FROM AUTHOR]

Abstract Large-eddy simulation (LES) is used to investigate the effect of the spatial arrangement of a utility-scale wind turbine array on the power outputs. Eight turbine-array layouts are considered, where each has 120 turbines installed in 30 rows with aligned or staggered configurations along the wakewise direction. We perform the LESs of neutrally-stratified atmospheric boundary layer over the eight large wind farms with the turbines arranged with a perfectly-aligned configuration, four laterally-staggered configurations, and three vertically-staggered configurations. Unlike the alignment of the turbine micro-siting in the aligned wind farm, both the laterally-staggered and vertically-staggered configurations lead to the misalignment of the turbines with staggered arrangement in the lateral and vertical directions. Simulation results show that the power outputs in the wind farms have obvious decreases to 45–65% within the first 12 turbine rows and retain within that range in the rest. In general, the staggered wind farms produce more power than the aligned wind farm. The laterally-staggered wind farm configurations due to the better adaptability in spatial configuration provide higher power production in the first 6th turbine rows. Moreover, the vertically-staggered configuration causes clear reduction in the maxima of the velocity deficit and the turbulence intensity inside the farm wake flow. Highlights • A validated LES framework is used to quantify the power output in a large wind farm with 120 turbines installed. • Three types of turbine arrays are considered: aligned, laterally-staggered, and vertically-staggered configurations. • The laterally-staggered configuration can promote more power generation in the first 6 turbine rows. • The vertically-staggered configuration leads to clear reduction in the maxima of the velocity deficit and the turbulence intensity. [ABSTRACT FROM AUTHOR]

[Display omitted] • Quantify the effect of local flows on AC cooling demand in a hilly, coastal city. • Downstream UHI could accelerate mountain-induced channel flows by 50.26 % • Channel winds cause heat advection that soar downstream AC load by 2.62 W m−2. • UHI-induced local flows interact with mountains, stagnating the sea-breeze on leeside. • Stagnant sea breeze surges downstream 2-m temperature (AC load) by 0.9 °C (6.41 W m−2) Extreme heat aggravates thermal stress and electricity shortage in urban areas. This study investigates the (circulating) winds in Hong Kong during a heatwave. Unprecedentedly, the collective effect of coastal winds, complex terrain, and local flows on urban temperatures and air-conditioning load intensity (ACLI) is examined using the mesoscale Weather Research and Forecasting (WRF) model. Three representative wind patterns, including urban-accelerated channel wind, channel-wind-induced heat advection, and urban-mountain-stagnated sea-breeze, are analyzed. Our results show that the mountain blockage in foothill areas would increase 2-m temperatures (T 2) and ACLI by 1 °C to 2 °C and 5 W/m2, respectively. ACLI in compact high-rise areas (LCZ 1) is most sensitive to extreme heat. Moreover, the urban heat island (UHI) downstream is crucial that would accelerate channel flows by 1.66 m/sec (50.26 %). On the other hand, terrain-induced channel winds augment heat advection, increasing downstream T 2 (0.7 °C) and ACLI (2.62 W/m2). UHI-induced local flows interact with hilly slopes, stagnating the sea breeze on mountain leeward side. Subsequently, the winds would be slowed down by 0.81 m/sec while the temperature T 2 would be increased by 0.9 °C in downstream urban areas. Eventually, the daytime ACLI could be raised as much as 6.41 W/m2. [ABSTRACT FROM AUTHOR]

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]

Air pollution is an increasingly concerning problem in many metropolitan areas due to its adverse public health and environmental impacts. Vertical atmospheric conditions have strong effects on vertical mixing of air pollutants, which directly affects surface air quality. The characteristics and magnitude of how vertical atmospheric conditions affect surface air quality, which are critical to future air quality projections, have not yet been fully understood. This study aims to enhance understanding of the annual and seasonal sensitivities of air pollution to both surface and vertical atmospheric conditions. Based on both surface and vertical meteorological characteristics provided by 1994–2003 monthly dynamic downscaling data from the Weather and Research Forecast Model, we develop generalized linear models (GLMs) to study the relationships between surface air pollutants (ozone, respirable suspended particulates, and sulfur dioxide) and atmospheric conditions in the Pearl River Delta (PRD) region. Applying Principal Component Regression (PCR) to address multi-collinearity, we study the contributions of various meteorological variables to pollutants’ concentration levels based on the loading and model coefficient of major principal components. Our results show that relatively high pollutant concentration occurs under relatively low mid-level troposphere temperature gradients, low relative humidity, weak southerly wind (or strong northerly wind) and weak westerly wind (or strong easterly wind). Moreover, the correlations vary among pollutant species, seasons, and meteorological variables at various altitudes. In general, pollutant sensitivity to meteorological variables is found to be greater in winter than in other seasons, and the sensitivity of ozone to meteorology differs from that of the other two pollutants. Applying our GLMs to anomalous air pollution episodes, we find that meteorological variables up to mid troposphere (∼700 mb) play an important role in influencing surface air quality, pinpointing the significant and unique associations between meteorological variables at higher altitudes and surface air quality. [ABSTRACT FROM AUTHOR]

A long term assessment of atmospheric polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) at Mt. Lulin, located in center of Taiwan was carried out from 2008 to 2013 (n = 81) assuming Mt. Lulin to be background area. During monitoring processes, PCDD/F samples collected in the field occasionally reached high concentration. To investigate this situation, simultaneous sample collection was carried out in Southeast Asia countries (i.e., Vietnam and Thailand) and Taiwan in 2013. The average concentration of atmospheric PCDD/Fs in biomass-burning source regions, namely Son La and Doi Ang Khang were 19.8 ± 12.1 fg I-TEQ m −3 (n = 19) and 17.8 ± 12.4 fg I-TEQ m −3 (n = 20), respectively. In the downwind area of Mt. Lulin, the average concentration of PCDD/Fs was found to be 4.64 ± 3.77 fg I-TEQ m −3 (n = 18). PCDD/F concentration in the source region was much higher than that in the downwind region. On March 19, 2013, the atmospheric PCDD/F concentrations increased dramatically from 7.71 to 484 fg I-TEQ m −3 at Mt. Lulin, which many times exceeded that of assumed source region in Southeast Asia. Moreover, mainland Southeast Asia and the southeast coast of China was suspected to be the main contributors of atmospheric PCDD/Fs and biomass markers, such as nonsea-salt K + and NH 4 + , during the spring. WRF-Chem and Potential Source Contribution Function (PSCF) simulations have confirmed this correlation. It can be concluded that atmospheric PCDD/Fs observed at Mt. Lulin during spring mostly derived from the air mass transport not only from Southeast Asia but also the southeast coast of China. [ABSTRACT FROM AUTHOR]

Reliable energy models are required to predict urban-scale air-conditioning (AC) load under extreme outdoor temperatures which is crucial to power security. In this study, we proposed a new landuse/landcover (LULC) scheme for these models by combining the local climate zones (LCZs) from World Urban Database Access Portal Tools (WUDAPT) and building category (BC) map. Apart from evaluating against the conventional LULC of Moderate-resolution Imaging Spectrometer (MODIS) and WUDAPT, the impact of urban canopy parameters and BC on the simulated meteorological conditions and AC load during the 2016 heatwave in Hong Kong was examined by the Weather Research and Forecasting (WRF) model. Incorporating LCZs (BC) reduced the bias of 2-m temperature (T 2), 2-m relative humidity (RH 2), and AC load intensity by 0.4 °C (2.0 °C), 1.98% (0.68%) and 6.78 W m−2 (13.09 W m−2), respectively. It captured the heterogeneous, intra-urban T 2 and turbulence kinetic energy (TKE). Further sub-classifying the BC in LCZs led to WUDAPT/BC which performed even better and significantly improved WRF output. It simulated well the daytime wind speed, mitigated the overprediction (underprediction) of T 2 (RH 2), and rectified the underpredicted AC load in WUDAPT. Its benefit was prominent in daytime as well as in compact high-rise/mid-rise and open high-rise areas. (200 words). • A new LULC scheme is proposed by combining WUDAPT and building category (BC) data. • WUDAPT/BC is unprecedentedly used to improve WRF reliability. • WUDAPT/BC outperforms WUDAPT and MODIS in urban-climate simulation. • WUDAPT/BC outperforms WUDAPT and MODIS representing intra-urban heterogeneity. • Including BC data reduces the bias of urban-scale air-conditioning load by 54.23%. [ABSTRACT FROM AUTHOR]

[Display omitted] • The Budyko framework was applied to 75 catchments covering a large area of Taiwan. • Varied landscape effects determined water resources availability among catchments. • Considerable spatial variability in the Budyko-related variables were found. • Three clusters were identified based on catchments' movements in the Budyko space. • 64 catchments will face the reduced availability of future water resources. Assessing water resources availability in a changing climate is a determinant of promoting sustainable development in a region. Such assessment should better be inter-basin in order to unveil a region's overall risk, so a time-varying Budyko framework is developed and adopted in this study. Gauge-based gridded precipitation and temperature, in company with runoff data, were used to derive the Budyko curves for 75 catchments over Taiwan. Downscaled precipitation and temperature data from selected global climate models and Representative Concentration Pathways (RCPs) were then used to calculate the shifts in the Budyko space of each catchment. It was found that the Budyko-related variables (e.g., evaporative and aridity indices, and catchment-specific parameter) exhibited considerable variability island-wide, with a few outliers in certain catchments being modulated by distinct landscape features and/or human activities. Under all RCPs analyzed, the majority of catchments showed the magnified movement vectors pointing at the 1st quadrant in the Budyko space in the late-21st century, suggesting a predominant and aggravating drying trend over Taiwan. Cluster analysis of the movement vectors under RCP8.5, which entailed the least inter-model differences, identified three major clusters, two of which (64 catchments included) indicated a drying trend of varied significancy. According to our findings, without effective mitigation of climate change, an increased likelihood of the worst-case scenario will substantialize the reduced availability of Taiwan's water resources. [ABSTRACT FROM AUTHOR]

Abstract: Increased mortality, linked to events of extreme high temperatures, is recognized as one critical challenge to the public health sector. Therefore, this ecological study was conducted to assess whether this association is also significant in Taiwan and the characteristics of the relationship. Daily mean heat indices, from 1994 through 2008, were used as the predictor for the risk of increased mortality in populations from 6 major Taiwanese cities. Daily mortality data from 1994 through 2008 were retrieved from the Taiwan Death Registry, Department of Health, Taiwan, and meteorological data were acquired from the Central Weather Bureau. Poisson regression analyses using generalized linear models were applied to estimate the temperature–mortality relationship. Daily mean heat indices were calculated and used as the temperature metric. Overall, increased risk ratios in mortality were associated with increased daily mean heat indices. Significantly increased risk ratios of daily mortality were evident when daily mean heat indices were at and above the 95th percentile, when compared to the lowest percentile, in all cities. These risks tended to increase similarly among those aged 65years and older; a phenomenon seen in the cities of Keelung, Taipei, Taichung, Tainan, and Kaohsiung, but not Chiayi. Being more vulnerable to heat stress is likely restricted to a short-term effect, as suggested by lag models which showed that there was dominantly an association during the period of 0 to 3days. In Taiwan, predicting city-specific daily mean heat indices may provide a useful early warning system for increased mortality risk, especially for the elderly. Regional differences in health vulnerabilities should be further examined in relation to the differential social–ecological systems that affect them. [Copyright &y& Elsevier]

Abstract: In this study, the synoptic weather types that have high probability and low probability of producing PM10 episode are referred to as HPE and LPE, respectively. Multiple linear regressions analysis showed that NO2 was the most important contributor (35.61%) to the concentrations of PM10 for HPE weather. For LPE weather, the season factor had the greatest contribution (48.11%) to the concentrations of PM10. Using the correlation coefficients between the concentrations of PM10 and SO2 or NO2 on HPE and LPE to calculate the increase of PM10 from LPE to HPE, we found that the increase of PM10 owing to the increase of SO2 and NO2 from LPE to HPE was 12.93μg/m3 which was about 51% of the total amount of PM10 increased from LPE to HPE. Results of factor analysis showed that the first component could be attributed to the result of local pollution especially for the weather patterns of types P3 and P6, while the secondary component for the weather patterns of types P1 and P4 can be attributed to the long-range transport of SO2 pollutants from China. [Copyright &y& Elsevier]

Riverine dissolved organic carbon (DOC), responsible for riverine productivity, is rarely documented in subtropical small mountainous rivers (SMRs) where high rainfall and steep slopes are the main features. This study investigated the DOC export at eight sites in three Taiwan SMRs to characterize the dynamics and controlling factors of DOC transport. Results showed that the mean DOC concentration of ~0.78 mg L−1 is much lower than the global average of ~5.29 mg L−1. However, the mean DOC yield, ~22.51 kg-C ha−1 yr−1, is higher than the global average of 14.4-19.3 kg-C ha−1 yr−1. Comparing with worldwide rivers from literature, the annual discharge, slope, and SOC (soil organic carbon) are controlling factors as expected, though they influence in different ways. SOC stock likely regulated by elevation-dependent biomes dominate the DOC supply, while slope restrains the DOC generation due to shallow soil depth and fast runoff velocity. However, the abundant discharge flushing this persistent low supply leads to a large DOC export in the SMRs. Furthermore, the DOC dynamics during typhoon periods showed a clockwise hysteresis, suggesting that the DOC is mainly from the riparian zone or downslope area during the rising limb of the hydrograph. This study elucidates the DOC transport in SMRs and provides an atypical yet significant piece of understanding on DOC transport in a global context. [ABSTRACT FROM AUTHOR]

Objective: To measure the association between ambient heat and hypoglycemia-related emergency department visit or hospitalization in insulin users., Research Design and Methods: We identified cases of serious hypoglycemia among adults using insulin aged ≥65 in the U.S. (via Medicare Part A/B/D-eligible beneficiaries) and Taiwan (via National Health Insurance Database) from June to September, 2016-2019. We then estimated odds of hypoglycemia by heat index (HI) percentile categories using conditional logistic regression with a time-stratified case-crossover design., Results: Among ∼2 million insulin users in the U.S. (32,461 hypoglycemia case subjects), odds ratios of hypoglycemia for HI >99th, 95-98th, 85-94th, and 75-84th percentiles compared with the 25-74th percentile were 1.38 (95% CI, 1.28-1.48), 1.14 (1.08-1.20), 1.12 (1.08-1.17), and 1.09 (1.04-1.13) respectively. Overall patterns of associations were similar for insulin users in the Taiwan sample (∼283,000 insulin users, 10,162 hypoglycemia case subjects)., Conclusions: In two national samples of older insulin users, higher ambient temperature was associated with increased hypoglycemia risk., (© 2024 by the American Diabetes Association.)

Autism spectrum disorder (ASD) is a group of neurodevelopmental disorders, and its incidence is increasing over time. Although several environmental factors have been suspected to be risk factors for ASD, studies on the effects of airborne heavy metals on newly developed ASD are still limited. We conducted a large birth cohort study of 168,062 live term births in Taichung during 2004-2011 to assess the association of heavy metals in particulate matter with an aerodynamic diameter less than 2.5 μm (PM 2.5 ) with ASD, and identify sensitive time windows during prenatal and postnatal periods. Heavy metals, including arsenic (As), cadmium (Cd), mercury (Hg), and lead (Pb) in PM 2.5 , were estimated using the Weather Research and Forecasting/Chem (WRF/Chem), inserted from the top 75 emission sources for the module. The association between childhood ASD and 4 metals were analyzed from pregnancy to 9 months after birth. The Cox proportional hazard model with a distributed lag nonlinear model (DLNM) was used to estimate the association between heavy metals in PM 2.5 and ASD. We identified 666 incident ASD cases in 168,062 participants. A positive association between Hg and ASD was found at 9 months after birth (Hazard Ratio: 1.63; 95% CI: 1.13-2.36). According to the DLNM, there was an increased risk of exposure to Hg during 10-25 weeks after birth, and decreased risk of exposure to Hg during gestational weeks 4-6. Exposure to As and Hg on the risk of ASD were significantly stronger in low birth weight infants (<2500 g) than in those of birth weight ≥2500 g during postnatal period. Postnatal exposure to Hg in PM 2.5 may associate with increased ASD incidence. Infants with low birth weight and exposure to As and Hg in PM 2.5 are more likely to develop ASD., 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 © 2023. Published by Elsevier Inc.)

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.)

Particulate lead (Pb) is a primary air pollutant that affects society because of its health impacts. This study investigates the source sectors of Pb associated with ambient fine particulate matter (PM 2.5 ) over central-western Taiwan (CWT) with new constraints on the Pb-isotopic composition. We demonstrate that the contribution of coal-fired facilities is overwhelming, which is estimated to reach 35 ± 16% in the summertime and is enhanced to 57 ± 24% during the winter monsoon seasons. Moreover, fossil-fuel vehicles remain a major source of atmospheric Pb, which accounts for 12 ± 5%, despite the current absence of a leaded gasoline supply. Significant seasonal and geographical variations in the Pb-isotopic composition are revealed, which suggest that the impact of East Asian (EA) pollution outflows is important in north CWT and drastically declines toward the south. We estimate the average contribution of EA outflows as accounting for 35 ± 15% (3.6 ± 1.5 ng/m 3 ) of the atmospheric Pb loading in CWT during the winter monsoon seasons., (© 2022. The Author(s).)

Wind is one of the cleanest renewable energy resources. Through the "Thousand Wind Turbines Project", Taiwan is planning to increase the proportion of power generation from renewable energy and has set a target of 5.7 GW for offshore wind by 2025. The effects of future offshore wind farms (OWFs) over the Taiwan Strait on the atmospheric environment have not been evaluated. This study examined the potential effects of proposed OWFs on the atmospheric environment if the OWFs had existed during Tropical Storm Haitang (2017) by using Weather Research and Forecasting (WRF) model. A small set of ensemble simulations was conducted for studying the sensitivity of the ambient conditions in the region to the wind farm locations, the number and density of the turbines, and the initial time of simulations. Following the landfall and northward movement of Tropical Storm Haitang, a series of complex interactions between the typhoon circulation and the wind farm emerged, including small time slots of wake effect and mountain blocking effect. The combination of these rapidly changing OWFs-related effects contributed to a weak reduction in precipitation (- 1.08 mm) and hub-height wind speed (- 0.25 m s -1 ), as well as minimal warming near the surface (+ 0.13 °C) over southern Taiwan., (© 2022. The Author(s).)

Background: Asthma is the most common chronic allergic disease in children; it affects more than 300 million people worldwide. Information on the association between exposure to ambient heavy metals and incidence of pediatric asthma is limited., Objective: We sought to evaluate the effects of heavy metals during pregnancy and infancy periods with asthma and identify a sensitive time window, clarifying the effect of ambient heavy metals on lung development., Methods: A total of 171,281 children, who were born from 2004 to 2011 in Taichung City, were followed until 2014. Concentrations of ambient heavy metals such as arsenic (As), cadmium (Cd), mercury (Hg), and lead (Pb) were obtained from the Weather Research and Forecasting/Chem model, considering the top 75 emission sources in Taiwan. The distributed lag nonlinear model was used to investigate the relationship between combined exposure to heavy metals in 2.5 μm particulate matter and asthma in pregnant women and 1-year-old infants., Results: We identified 31,277 new asthma cases from the birth cohort. After adjustment for socioeconomic status, maternal age, maternal atopy, maternal anemia, and maternal kidney disease, distributed lag nonlinear model results revealed positive associations of asthma with exposure to Pb during gestational weeks 1 to 14 and 21 to 40, and 1 to 3 weeks after birth. Regarding the sensitivity analyses, coexposure to Pb and As, coexposure to Pb and Cd, and coexposure to Pb and Hg were positively associated with asthma onset as well., Conclusions: Our study suggested that combined exposure to Pb with As, Cd, and Hg during early and late gestational weeks was associated with the incidence of pediatric asthma., (Copyright © 2020 American Academy of Allergy, Asthma & Immunology. Published by Elsevier Inc. 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.)

This study investigated the drivers and extended-range prediction of the mid-July to early August 2018 heatwaves over northwestern Europe, focusing on regional heatwave events over Scandinavia (SC) and Western Europe (WE). The persistent blocking regime (BL) was the most influential contributor for the 2018 heatwave over SC, and both the Atlantic Low regime (AL) and North Atlantic Oscillation (NAO) were secondary contributors for the heatwave, but with different effect directions. The major contributor to the heatwave over WE was AL. These causal relationships remained valid when the evolution of warm spells was considered. A multi-model ensemble of real-time forecasts from the subseasonal to seasonal (S2S) database captured the evolution of the warm spells over SC and WE up to 3 weeks in advance. However, the predictions of heatwave occurrence and significance for the two regions are unsatisfactory. BL and AL can be predicted 2 weeks in advance, resulting in the successful predictions of warm spells over SC and WE. Although variations in Azores High and NAO were captured in the forecasts, their contribution to the warm spells remains unclear.

The consistent, sporadic transmission of shigellosis in Taiwan necessitates an exploration of risk factors for the occurrence of shigellosis. The purpose of this study was to study the epidemiologic characteristics and the relationship between climatic factors and the incidence of shigellosis in Taiwan. We collected data from cases of shigellosis reported to the Taiwan Centers for Disease Control (Taiwan CDC) from 2001 to 2016. Climatic data were obtained from the Taiwan Central Weather Bureau. The relationships between weather variability and the incidence of shigellosis in Taiwan were determined via Poisson regression analyses. During the 16-year study period, a total of 4171 clinical cases of shigellosis were reported to the Taiwan CDC. Among them, 1926 (46.2%) were classified as confirmed cases. The incidence of shigellosis showed significant seasonality, with the majority of cases occurring in summertime (for oscillation, P < .001). The number of shigellosis cases started to increase when temperatures reached 21°C (r = 0.88, P < .001). Similarly, the number of shigellosis cases began to increase at a relative humidity of 70-74% (r = 0.75, P < .005). The number of shigellosis cases was positively associated with the mean temperature and relative humidity in the period preceding the infection. In conclusion, the occurrence of shigellosis is significantly associated with increasing temperature and relative humidity in Taiwan. Therefore, these factors could be regarded as warning signals indicating the need to implement preventive measures.

Aims/hypothesis: Information on the associations of long-term exposure to fine particulate matter (with an aerodynamic diameter less than 2.5 μm; PM 2.5 ) with the development of type 2 diabetes is scarce, especially for south-east Asia, where most countries are experiencing serious air pollution. This study aimed to investigate the long-term effects of exposure to ambient PM 2.5 on the incidence of type 2 diabetes in a population of Taiwanese adults., Methods: A total of 147,908 participants without diabetes, at least 18 years of age, were recruited in a standard medical examination programme between 2001 and 2014. They were encouraged to take medical examinations periodically and underwent at least two measurements of fasting plasma glucose (FPG). Incident type 2 diabetes was identified as FPG ≥7 mmol/l or self-reported physician-diagnosed diabetes in the subsequent medical visits. The PM 2.5 concentration at each participant's address was estimated using a satellite-based spatiotemporal model with a resolution of 1 × 1 km 2 . The 2 year average of PM 2.5 concentrations (i.e. the year of and the year before the medical examination) was treated as an indicator of long-term exposure to ambient PM 2.5 air pollution. We performed Cox regression models with time-dependent covariates to analyse the long-term effects of exposure to PM 2.5 on the incidence of type 2 diabetes. A wide range of covariates were introduced in the models to control for potential effects, including age, sex, education, season, year, smoking status, alcohol drinking, physical activity, vegetable intake, fruit intake, occupational exposure, BMI, hypertension and dyslipidaemia (all were treated as time-dependent covariates except for sex)., Results: Compared with the participants exposed to the first quartile of ambient PM 2.5 , participants exposed to the second, third and fourth quartiles of ambient PM 2.5 had HRs of 1.28 (95% CI 1.18, 1.39), 1.27 (95% CI 1.17, 1.38) and 1.16 (95% CI 1.07, 1.26), respectively, for the incidence of type 2 diabetes. Participants who drank occasionally or regularly (more than once per week) or who had a lower BMI (<23 kg/m 2 ) were more sensitive to the long-term effects of exposure to ambient PM 2.5 ., Conclusions/interpretation: Long-term exposure to ambient PM 2.5 appears to be associated with a higher risk of developing type 2 diabetes in this Asian population experiencing high levels of air pollution.

High energy consumption in the urban environment impacts the urban surface energy budget and causes the emission of anthropogenic heat fluxes (AHFs) into the atmosphere. AHFs vary over time and space. Thus, a reliable estimation of AHF is needed for mesoscale meteorological modeling. This study used a statistical regression method to estimate the annual mean gridded AHF with high spatial (1-km) resolution. Compared with current methods for AHF estimation, the statistical regression method is straightforward and can be easily incorporated with meteorological modeling. AHF of the highly populated urban areas in Taiwan were estimated using data from the anthropogenic pollutant emission inventory of CO and NOx for year 2010. Over 40% of the total AHF values in Taiwan main island fell within the range of 10-40 Wm(-2). When the study domain was confined to urban land, the percentage contributions from AHF values were increased, with over 68% of the total AHF values within the range of 10-40 Wm(-2). AHF values > 40 Wm(-2) were more abundant in the Southern region, followed by the Central and Northern regions. An assessment of the heat emissions by the large scale urban consumption of energy (LUCY) model revealed that the mean AHFs are reasonably close to those produced while the maximum AHFs are underestimated. The results obtained evidence the impact of spatial distribution of land use types, particularly population densities, main highways and industries on AHF generation in Taiwan., (Copyright © 2016 Elsevier Ltd. All rights reserved.)