Emerging evidence witnesses the association of air pollution exposure with sleep disorders or the risk of obstructive sleep apnea (OSA); however, the results are not consistent. OSA patients with or without a low arousal threshold (LAT) have different pathology and therapeutic schemes. No study has evaluated the potential diverse effects of air pollution on the phenotypes of OSA. The current study aimed to evaluate the associations of short-term and long-term exposure to air pollution with sleep-disordered measures and OSA phenotypes. This cross-sectional study consisted of 4634 participants from a sleep center in Taipei from January 2015 to April 2019. The personal exposure to ambient PM₂.₅ and NO₂ was assessed by a spatial-temporal model. Overnight polysomnography was used to measure the sleep parameters. According to a developed clinical tool, we defined the low arousal threshold (LAT) and identified the OSA patients with or without LAT. We applied a generalized linear model and multinomial logistic regression model to estimate the change of sleep measures and risk of the OSA phenotypes, respectively, associated with an interquartile range (IQR) increment of personal pollution exposure after adjusting for the essential confounders. In the single-pollutant model, we observed the associations of NO₂ with sleep-disordered measures by decreasing the total sleep time, sleep efficiency, extending the time of wake after sleep onset, and the association of NO₂ with the increased risk of LAT OSA by around 15%. The two-pollutant model with both long-term and short-term exposures confirmed the most robust associations of long-term NO₂ exposure with sleep measures. An IQR increment of NO₂ averaged over the past year (6.0 ppb) decreased 3.32 min of total sleep time and 0.85% of sleep efficiency. Mitigating exposure to air pollution may improve sleep quality and reduce the risk of LAT OSA.

Reliable long-horizon PM₂.₅ forecasts are crucial and beneficial for health protection through early warning against air pollution. However, the dynamic nature of air quality makes PM₂.₅ 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 PM₂.₅ 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 PM₂.₅ 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 PM₂.₅ 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 PM₂.₅ 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.

Seawater, sediments, and three genera of wild scleractinian corals were collected from four coral reef areas nearshore Liuqiu Island, southwestern Taiwan. Abundance, characteristics (sizes, colors, shapes, and polymer types), and enrichment of microplastics (MPs) in the corals, and their impacts on coral cover were determined. The average MPs abundances were 0.95, 0.77, and 0.36 item/g for Galaxea sp, Acropora spp, and Pocillopora sp, respectively. The MPs abundance was relatively higher on the coral surfaces than inside the skeletons, dominated by blue rayon-fibers, correspondingly observed in seawater and sediments. Large-size colorless MPs tended to be mis-ingested by Galaxea sp. (71%) compared with Pocillopora sp. (43%) and Acropora spp. (31%). The low hard coral cover (12.5%) observed at Yufu (L1) on the northeastern coastal zone nearby tourism center of Liuqiu Island where correspondingly associated with high MPs abundance in seawater (10 item/L), sediments (260 item/kg), and corals (0.60 item/g). Tourism induced sewage discharges and sailing activities significantly contributed to the MPs pollution, probably contributing to the loss of coral cover. High MPs enrichment in corals (EFMP = 25–283) shows that the marine MPs pollution can critically threaten coral reef ecosystems. Fibrous MPs present inside the coral skeleton serve as potential indicator of MPs’ impact on corals—with the dominance of textile-related rayon and polyester/PET microfibers in the coral reef zones. This study provided valuable information for coral conservation and coastal management.

The levels and characteristics of atmospheric metals vary in time and location, can result in various health impacts, which increases the challenge of air quality management. We aimed to investigate PM₂.₅-bound metals in multiple locations and propose a methodology for comparing metal elements across study regions and prioritizing source contributions through integrated health risk assessments. PM₂.₅-bound metals were collected in the urban, suburban, rural, and industrial regions of Taiwan between 2016 and 2018. We incorporated the positive matrix factorization (PMF) with health risk assessments (considering estimates of the margin of exposure (MOE) and excess cancer risk (ECR)) to prioritize sources for control. We found that the concentrations of Fe, Zn, V, Cu, and Mn (industry-related metals) were higher at the industrial site (Kaohsiung) and Ba, Cr, Ni, Mo, and Co (traffic-related metals) were higher at the urban site (Taipei). The rural site (Hualian) had good air quality, with low PM₂.₅ and metal concentrations. Most metal concentrations were higher during the cold season for all study sites, except for the rural. Ambient concentrations of Mn, Cr, and Pb obtained from all study sites presents a higher health risk of concern. In Kaohsiung, south Taiwan, PM₂.₅-bound metals from the iron ore and steel factory is suggested as the first target for control based on the calculated health risks (MOE < 1 and ECR > 10⁻⁶). Overall, we proposed an integrated strategy for initiating the source management prioritization of PM₂.₅-bound metals, which can aid an effort for policymaking.

Ambient fine particulate matter (PM₂.₅) has been ranked as the sixth leading risk factor globally for death and disability. Modelling methods based on having access to a limited number of monitor stations are required for capturing PM₂.₅ spatial and temporal continuous variations with a sufficient resolution. This study utilized a land use regression (LUR) model with machine learning to assess the spatial-temporal variability of PM₂.₅. Daily average PM₂.₅ data was collected from 73 fixed air quality monitoring stations that belonged to the Taiwan EPA on the main island of Taiwan. Nearly 280,000 observations from 2006 to 2016 were used for the analysis. Several datasets were collected to determine spatial predictor variables, including the EPA environmental resources dataset, a meteorological dataset, a land-use inventory, a landmark dataset, a digital road network map, a digital terrain model, MODIS Normalized Difference Vegetation Index (NDVI) database, and a power plant distribution dataset. First, conventional LUR and Hybrid Kriging-LUR were utilized to identify the important predictor variables. Then, deep neural network, random forest, and XGBoost algorithms were used to fit the prediction model based on the variables selected by the LUR models. Data splitting, 10-fold cross validation, external data verification, and seasonal-based and county-based validation methods were used to verify the robustness of the developed models. The results demonstrated that the proposed conventional LUR and Hybrid Kriging-LUR models captured 58% and 89% of PM₂.₅ variations, respectively. When XGBoost algorithm was incorporated, the explanatory power of the models increased to 73% and 94%, respectively. The Hybrid Kriging-LUR with XGBoost algorithm outperformed the other integrated methods. This study demonstrates the value of combining Hybrid Kriging-LUR model and an XGBoost algorithm for estimating the spatial-temporal variability of PM₂.₅ exposures.

The occurrence of emergent contaminants, 24 polybrominated diphenyl ethers (PBDEs), di(2-ethylhexyl)phthalate (DEHP), dibutyl phthalate (DBP), butyl benzyl phthalate (BBP), diethyl phthalate (DEP), dimethyl phthalate (DMP), di-n-octyl phthalate (DnOP), bisphenol A (BPA) and nonylphenol (NP), was investigated in sediments and fishes collected from the Tamsui River system to determine the factors that influence their distribution and their risk to aquatic ecosystems and human health. The concentrations of total PBDEs, DEHP, DBP, BBP, DEP, DMP, DnOP, BPA and NP in sediments were 1–955, ND-23570, <50–411, <50–430, ND-80, ND-<50, ND-<50, 1–144, 3–19624 μg/kg dw, respectively. The spatial-temporal distribution trends of these compounds in sediments could be attributed to urbanization, industrial discharge and effluents from wastewater treatment plants. The PBDE congener distribution patterns (BDE-209 was the dominant congener) in sediments reflected the occurrence of debromination of BDE-209 and the elution of penta-BDE from the treated products. The concentrations of total PBDEs, DEHP, DBP, BBP, DEP, DMP, DnOP, BPA and NP in fish muscles were 2–66, 17–1046, <10–231, <10–66, <30, ND-<30, ND-<30, 0.4–7 and 3–440 μg/kg ww, respectively. The species-specific bioaccumulation of these compounds by fish was found and four species particularly showed high bioaccumulation potential. BDE-47 was the predominant BDE congener in fish muscles, suggesting high bioavailability and bioaccumulation of this compound. The results of biota–sediment accumulation factors showed that BDE-47, 99, 100, 153 and 154 had relatively high bioavailability and bioaccumulation potential for some fish species. The ecological risk assessment showed that the concentrations of BPA and NP in sediments were likely to have adverse effects on aquatic organisms (risk quotients > 1). The human health risk assessment according to hazard quotients (HQs) and carcinogenic risks (CRs) revealed no remarkable risk to human health through consumption of fish contaminated with BDE-47, 99, 100, 154, 209, DEHP, BPA and NP.

The understanding for the impact of petrochemical pollutants exposure on renal functions is limited.Our study examined the associations between renal functions and pollutants exposure in adult residents living in the vicinity of a petrochemical industry.We recruited 2069 adult residents near a big petrochemical complex in Taiwan in 2009–2012, and they were categorized into high exposure (HE) and low exposure (LE) groups based on their address to source by 10 km radius. Study subjects were measured the urinary levels of arsenic, cadmium, mercury, thallium, and 1-hydroxypyrene (1-OHP). The estimated glomerular filtration rate (eGFR) was calculated using the Taiwanese Chronic Kidney Disease Epidemiology Collaboration equation, and the chronic kidney disease (CKD) prevalence and risks were defined according to KDIGO 2012 guidelines. Adjusted generalized linear and logistic regression models were applied to evaluate the associations between petrochemical exposure and renal functions.Subjects in the HE areas had significantly lower eGFR, higher CKD prevalence, and higher levels of urinary arsenic, cadmium, mercury, thallium and 1-OHP. The closer to complex and high exposure group of study subjects were significantly associated with the decrease in eGFR, higher ORs for CKD and high-intermediate risk of CKD. In addition, the study subjects who had two-fold urinary arsenic and 1-OHP levels were significantly with decreased 0.68 and 0.49 ml/min/1.73 m2 of eGFR, respectively.Residing closer and higher arsenic and polycyclic aromatic hydrocarbon exposure were associated with the renal impairment and risks of CKD among the residential population near the petrochemical industry.

This study examined the long-term trends in chemical components in PM₂.₅ (particulate matter with aerodynamic diameter ≤2.5 μm) samples collected at Lulin Atmospheric Background Station (LABS) located on the summit of Mt. Lulin (2862 m above mean sea level) in Taiwan in the western North Pacific during 2003–2018. High ambient concentrations of PM₂.₅ and its chemical components were observed during March and April every year. This enhancement was primarily associated with the long-range transport of biomass burning (BB) smoke emissions from Indochina, as revealed from cluster analysis of backward air mass trajectories. The decreasing trends in ambient concentrations of organic carbon (−0.67% yr⁻¹; p = 0.01), elemental carbon (−0.48% yr⁻¹; p = 0.18), and non–sea-salt (nss) K⁺ (−0.71% yr⁻¹; p = 0.04) during 2003–2018 indicated a declining effect of transported BB aerosol over the western North Pacific. These findings were supported by the decreasing trend in levoglucosan (−0.26% yr⁻¹; p = 0.20) during the period affected by the long-range transport of BB aerosol. However, NO₃⁻ displayed an increasing trend (0.71% yr⁻¹; p = 0.003) with considerable enhancement resulting from the air masses transported from the Asian continent. Given that the decreasing trends were for the majority of the chemical components, the columnar aerosol optical depth (AOD) also demonstrated a decreasing trend (−1.04% yr⁻¹; p = 0.0001) during 2006–2018. Overall decreasing trends in ambient (carbonaceous aerosol and nss-K⁺) as well as columnar (e.g., AOD) aerosol loadings at the LABS may influence the regional climate, which warrants further investigations. This study provides an improved understanding of the long-term trends in PM₂.₅ chemical components over the western North Pacific, and the results would be highly useful in model simulations for evaluating the effects of BB transport on an area.

Air quality in the megacity Delhi is affected not only by local emissions but also by pollutants from crop residue burning in the surrounding areas of the city, particularly the rice straw burning in the post monsoon season. As a major burning product, gaseous CO₂, which is rather inert in the polluted atmosphere, provides an alternative solution to characterize the impact of biomass burning from a new perspective that other common tracers such as particulate matters are limited because of their physical and chemical reactiveness. Here, we report conventional ([CO₂], δ¹³C, and δ¹⁸O) and unconventional (Δ¹⁷O) isotope data for CO₂ collected at Connaught Place (CP), a core area in the megacity Delhi, and two surrounding remote regions during a field campaign in October 18–20, 2017. We also measured the isotopic ratios near a rice straw burning site in Taiwan to constrain their end member isotopic compositions. Rice straw burning produces CO₂ with δ¹³C, δ¹⁸O, and Δ¹⁷O values of −29.02 ± 0.65, 19.63 ± 1.16, and 0.05 ± 0.02‰, respectively. The first two isotopic tracers are less distinguishable from those emitted by fossil fuel combustion but the last one is significantly different. We then utilize these end member isotopic ratios, with emphasis on Δ¹⁷O for the reason given above, for partitioning sources that affect the CO₂ level in Delhi. Anthropogenic fraction of CO₂ at CP ranges from 4 to 40%. Further analysis done by employing a three-component (background, rice straw burning, and fuel combustion) mixing model with constraints from the Δ¹⁷O values yields that rice straw burning contributes as much as ∼70% of the total anthropogenic CO₂, which is more than double of the fossil fuel contribution (∼30%), during the study days.