Search

Tropospheric ozone (O 3 ) pollution is worsening in China, and an accurate forecast is a prerequisite to lower the O 3 peak level. In recent years, machine learning techniques have attracted increasing attention in O 3 prediction owing to their high efficiency and simple operation. However, the accuracy of predicting the daily O 3 level is low. This study proposed a novel model by coupling long short-term memory neural network with transfer learning (TL-LSTM), with meteorology and pollutant concentration information as the model input. L2 regularization was applied to reduce the risk of overfitting and to improve the accuracy and generalization ability of the model prediction. Our results indicated that by transferring the knowledge in the model configuration from the hourly LSTM module, TL-LSTM greatly improves the predictability of the daily maximum 8 h average (MDA8) of O 3 in Hong Kong. The coefficient of determination (R 2 ) increased from 0.684 to 0.783 and the mean square error (MSE) reduced from 1.36 × 10 -2 to 1.05 × 10 -2 . Furthermore, R 2 and MSE were the highest in summer, indicating an under-prediction of peak O 3 levels. This was a result of the limited number of high O 3 days, which did not provide sufficient knowledge for the model to make an accurate prediction. Sobol analysis indicated that wind speed was the most sensitive factor in O 3 prediction, largely due to the development of land-sea breeze circulation which effectively traps pollutants and expedites O 3 formation. The results clearly demonstrate the effectiveness of the TL-LSTM in predicting the daily O 3 concentration in Hong Kong. Thus, TL-LSTM can be promulgated into other photochemically active regions to assist in O 3 pollution forecasting and management., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have influenced the work reported in this paper., (Copyright © 2022 Elsevier B.V. All rights reserved.)

To curb the continuous deterioration of ozone (O 3 ) pollution in China, identifying the O 3 -precursor sensitivity (OPS) and its driving factors is a prerequisite for formulating effective O 3 pollution control measures. Traditional OPS identification methods have limitations in terms of spatiotemporal representation and timeliness; therefore, they are not appropriate for making OPS forecasts for O 3 contingency control. OPS is not only influenced by local precursor emissions but is also closely related to meteorological conditions governed by large-scale circulation (LSC). In this study, a localized three-dimensional numerical modeling system was used to investigate the relationship between LSC and OPS in the Pearl River Delta (PRD) of China during September 2017, a month with continuous O 3 pollution. Our results highlighted that there was a close relationship between LSC and OPS over the PRD, and the four dominant LSC patterns corresponded well to the NO x -limited, NO x -limited, VOC-limited, and transitional regimes, respectively. The clear linkage between LSC and OPS was mainly driven by the spatial heterogeneity of NO x and VOC emissions within and beyond the PRD along the prevailing winds under different LSC patterns. A conceptual model was developed to highlight the intrinsic causality between the LSC and OPS. Because current technology can accurately forecast LSC 48-72 h in advance, the LSC-based OPS forecast method provided us with a novel approach to guide contingency control and management measures to reduce peak O 3 at a regional scale., (Copyright © 2022 Elsevier Ltd. All rights reserved.)

Aims: To assess possible effect of air quality improvements, we investigated the temporal change in hospital admissions for acute exacerbations of chronic obstructive pulmonary disease (AECOPD) associated with pollutant concentrations., Methods: We collected daily concentrations of particulate matter (i.e., PM 2.5 , PM 10 and PM coarse ), sulfur dioxide (SO 2 ), nitrogen dioxide (NO 2 ), carbon monoxide (CO), ozone (O 3 ), and admissions for AECOPD for 21 cities in Guangdong from 2013 to 2017. We examined the association of air pollution with AECOPD admissions using two-stage time-series analysis, and estimated the annual attributable fractions, numbers, and direct hospitalization costs of AECOPD admissions with principal component analysis., Results: From 2013-2017, mean daily concentrations of SO 2 , PM 10 and PM 2.5 declined by nearly 40%, 30%, and 26% respectively. As the average daily 8 h O 3 concentration increased considerably, the number of days exceeding WHO target (i.e.,100 μg/m³) increased from 103 in 2015-152 in 2017. For each interquartile range increase in pollutant concentration, the relative risks of AECOPD admission at lag 0-3 were 1.093 (95% CI 1.06-1.13) for PM 2.5 , 1.092 (95% CI 1.08-1.11) for O 3 , and 1.092 (95% CI 1.05-1.14) for SO 2 . Attributable fractions of AECOPD admission advanced by air pollution declined from 9.5% in 2013 to 4.9% in 2016, then increased to 6.0% in 2017. A similar declining trend was observed for direct AECOPD hospitalization costs., Conclusion: Declined attributable hospital admissions for AECOPD may be associated with the reduction in concentrations of PM 2.5 , PM 10 and SO 2 in Guangdong, while O 3 has emerged as an important risk factor. Summarizes the main finding of the work： Reduction in PM may result in declined attributable hospitalizations for AECOPD, while O 3 has emerged as an important risk factor following an intervention., (Copyright © 2020 The Authors. Published by Elsevier Inc. All rights reserved.)

The relation of acute fluctuations of air pollution to hospital admission for bronchiectasis remained uncertain, and large-scale studies were needed. We collected daily concentrations of particulate matter (PM), sulfur dioxide (SO 2 ), nitrogen dioxide (NO 2 ), carbon monoxide (CO), ozone (O 3 ), and daily hospitalizations for bronchiectasis for 21 cities across Guangdong Province from 2013 through 2017. We examined their association using two-stage time-series analysis. Our analysis was stratified by specific sub-diagnosis, sex and age group to assess potential effect modifications. Relative risks of hospitalization for bronchiectasis were 1.060 (95%CI 1.014-1.108) for PM 10 at lag0-6, 1.067 (95%CI 1.020-1.116) for PM 2.5 at lag0-6, 1.038 (95%CI 1.005-1.073) for PM coarse at lag0-6, 1.058 (95%CI 1.015-1.103) for SO 2 at lag0-4, 1.057 (95%CI 1.030-1.084) for NO 2 at lag0 and 1.055 (95%CI 1.025-1.085) for CO at lag0-6 per interquartile range increase of air pollution. Specifically, acute fluctuations of air pollution might be a risk factor for bronchiectasis patients with lower respiratory infection but not with hemoptysis. Patients aged ≥65 years, and female patients appeared to be particularly susceptible to air pollution. Acute fluctuations of air pollution, particularly PM may increase the risk of hospital admission for bronchiectasis exacerbations, especially for the patients complicated with lower respiratory infection. This study strengthens the importance of reducing adverse impact on respiratory health of air pollution to protect vulnerable populations., (Copyright © 2020 Elsevier GmbH. All rights reserved.)

Serious particulate matter (PM) pollution problems in many polluted regions of China have been frequently reported in recent years. Long-term exposure to ambient PM pollution is significantly associated with adverse health effects. Characterizing the long-term trends and variation in PM pollution is a basic requirement for evaluating long-term exposure and for guiding future policies to reduce the effects of air pollution on health. However, long-term, ground-based PM measurements are only available at a few fixed stations. In this study, an algorithm is developed and validated to estimate PM concentrations based on the satellite atmospheric optical depth with 1 km spatial resolution. The long-term trends of PM10 concentrations in the entire Pearl River Delta (PRD) region and different cities are quantified and discussed. From 2001 to 2013, the PM10 pollution of the entire PRD region was dominated by a decreasing trend of -0.15 ± 0.23 μg/m(3)·yr. This decreasing PM10 trend was apparent over 75% of the PRD area, with the most significant decreases observed in the center of the region. However, the remaining 25%, mostly located in the outskirts of the region, showed an increasing PM10 trend. This overall decreasing trend indicates the effectiveness of the control measures applied in the past decade for the primary pollutants.

Background: The association between diurnal temperature range (DTR) and hospitalization for exacerbation of chronic respiratory diseases (CRD) was rarely reported.Objectives: To examine the association between DTR and daily hospital admissions for exacerbation of CRD and find out the potential effect of modifications on this association.Method: Data on daily hospitalization for exacerbation of chronic obstructive pulmonary disease (COPD), asthma and bronchiectasis and meteorology measures from 2013 through 2017 were obtained from 21 cities in South China. After controlling the effects of daily mean temperature, relative humidity (RH), particulate matter < 2.5 μm diameter (PM2.5) and other confounding factors, a standard generalized additive model (GAM) with a quasi-Poisson distribution was performed to evaluate the relationships between DTR and daily hospital admissions of CRD in a two-stage strategy. Subgroup analysis was performed to find potential modifications, including seasonality and population characteristics.Result: Elevated risk of hospitalization for exacerbation of CRD (RR = 1.09 [95%CI: 1.08 to 1.11]) was associated with the increase in DTR (the 75th percentile versus the 25th percentile of DTR at lag0-6). The effects of DTR on hospital admissions for CRD were strong at low DTR in the hot season and high DTR in the cold season. The RR (the 75th percentile versus the 25th percentile of DTR at lag0-6) of hospitalization was 1.11 (95%CI: 1.08 to 1.12) for exacerbations of COPD and 1.09 (95%CI: 1.05 to 1.13) for asthma. The adverse effect of DTR on hospitalization for bronchiectasis was only observed in female patients (RR = 1.06 [95%CI: 1.03 to 1.10]).Conclusion: Our study provided additional evidence for the association between DTR and daily hospitalization for exacerbation of CRD, and these associations are especially stronger in COPD patients and in the cold season than the hot season. Preventive measures to reduce the adverse impacts of DTR were needed for CRD patients. [ABSTRACT FROM AUTHOR]

Protecting human health is one of the fundamental goals of continuous air quality improvement. Considering the same sources of CO 2 and air pollutants emissions, their emission control measures often have a certain degree of synergistic emission reduction effects To balance the costs and policy efficiency, policy makers should assess the synergistic benefits of CO 2 emission reduction when formulating air quality improvement strategies. Aimed at the air quality improvement strategy targeting WHO-III level (i.e. 15 μg/m3) for PM2.5 in the Pearl River Delta (PRD) region, this study evaluates the synergistic CO2 reduction effects and the associated health benefits from this strategy which includes a range of air pollution prevention and control measures. The results of the study show that promoting the attainment of WHO-III for PM2.5 in the PRD region can bring about 74 Mt of synergistic CO2 emission reduction, resulting in an 18% reduction of CO2 emissions in the PRD compared to 2017. Among various measures considered, industrial restructuring, power supply transition and industrial energy consumption transformation exhibit the most pronounced synergistic effects. Therefore, these measures are recommended to be prioritized and promoted in the next stage of air pollution prevention and CO2 emission reduction. Furthermore, when the PM 2.5 concentration in the PRD region reaches the WHO-III level, the number of PM 2.5 -related deaths will be estimated to reduce approximately 5.5 thousand compared to that in the current policy scenario. Through continuous structural transformation and emission reduction efforts, it not only facilitates the decline in regional PM 2.5 concentration but also helps more regional residents to live in an environment with a relatively low PM 2.5 concentration. In addition, in order to reduce the health impacts of PM 2.5 , it is recommended that the government should guide people to change their production and living styles in order to reduce pollutant emissions from anthropogenic activities. • Measures promoting PRD's PM 2.5 to WHO-III level reduces 74 Mt CO 2 , resulting in an 18% reduction compared to 2017 • Structural adjustment measures have the most synergistic effect. • The PM 2.5 -related deaths decrease by 5.5 thousand compared to BAU scenario. • Consider air pollution and population structure for PM 2.5 -related deaths reduction. [ABSTRACT FROM AUTHOR]