Evaluating the spatiotemporal ozone characteristics with high-resolution predictions in mainland China, 2013–2019

Environmental Pollution, Год журнала: 2022, Номер 299, С. 118865 - 118865

Опубликована: Янв. 18, 2022

Evaluating ozone levels at high resolutions and accuracy is crucial for understanding the spatiotemporal characteristics of distribution assessing exposure in epidemiological studies. The national models with to predict ground concentrations are limited China so far. In this study, we aimed develop a random forest model by combining measurements from fixed stations, simulations Community Multiscale Air Quality (CMAQ) modeling system, meteorological parameters, population density, road length, elevation maximum daily 8-h average (MDA8) level 1 km × spatial resolution. cross-validation R2 root mean squared error (RMSE) were 0.80 20.93 μg/m3 2013-2019, respectively. CMAQ near-surface temperature played vital roles predicting among all predictors. population-weighted median predicted MDA8 89.34 mainland 2013, reached 100.96 2019. However, long-term temporal variations regions heterogeneous. Central Eastern China, as well Southeast Coastal Area, suffered higher pollution increased rates 2013 seasonal pattern varied spatially. peak-season highest 6-month occurred different months regions, more than half domain April-September. predictions showed that not only annual but also percentages grid-days 100/160 have been increasing past few years China; meanwhile, majority areas air quality guidelines launched World Health Organization September 2021. proposed resolution full coverage could provide health studies flexible choices evaluate multiple scales future.

Язык: Английский

---

Prediction and evaluation of spatial distributions of ozone and urban heat island using a machine learning modified land use regression method

Sustainable Cities and Society, Год журнала: 2021, Номер 78, С. 103643 - 103643

Опубликована: Дек. 27, 2021

Язык: Английский

---

A comparison of machine learning methods for ozone pollution prediction

Journal Of Big Data, Год журнала: 2023, Номер 10(1)

Опубликована: Май 15, 2023

Abstract Precise and efficient ozone ( $$\hbox {O}_{3}$$ O3 ) concentration prediction is crucial for weather monitoring environmental policymaking due to the harmful effects of high pollution levels on human health ecosystems. However, complexity formation mechanisms in troposphere presents a significant challenge modeling accurately quickly, especially absence process model. Data-driven machine-learning techniques have demonstrated promising performance air pollution, mainly when model unavailable. This study evaluates predictive nineteen machine learning models prediction. Specifically, we assess how incorporating features using Random Forest affects investigate time-lagged measurements improve accuracy. Air meteorological data collected at King Abdullah University Science Technology are used. Results show that dynamic outperform static reduced models. Incorporating improves accuracy by 300% 200%, respectively, compared models, under RMSE metrics. And importantly, best with information only requires 0.01 s, indicating its practical use. The Diebold-Mariano Test, statistical test used compare forecasting also conducted.

Язык: Английский

---

The role of lifestyle in the association between long-term ambient air pollution exposure and cardiovascular disease: a national cohort study in China

BMC Medicine, Год журнала: 2024, Номер 22(1)

Опубликована: Март 5, 2024

Abstract Background Cardiovascular disease (CVD) caused by air pollution poses a considerable burden on public health. We aim to examine whether lifestyle factors mediate the associations of pollutant exposure with risk CVD and extent interaction between lifestyles regarding outcomes. Methods included 7000 participants in 2011–2012 followed up until 2018. The evaluation consists six as proxies, including blood pressure, glucose, lipids, body mass index, tobacco exposure, physical activity, were categorized into three groups according number ideal (unfavorable, 0–1; intermediate, 2–4; favorable, 5–6). Satellite-based spatiotemporal models used estimate ambient pollutants (including particles diameters ≤ 1.0 μm [PM 1 ], 2.5 10 nitrogen dioxide [NO 2 ozone [O 3 ]). Cox regression CVD. mediation modification effects categories association analyzed. Results After adjusting for covariates, per μg/m increase PM (HR: 1.09, 95% CI: 1.05–1.14), 1.04, 1.00–1.08), 1.05, 1.03–1.08), NO 1.11, 1.05–1.18) was associated an increased Adherence healthy reduced compared unfavorable 0.65, 0.56–0.76 intermediate HR: 0.41, 0.32–0.53 favorable lifestyle). Lifestyle played significant partial mediating role contribution CVD, proportion ranging from 7.4% 14.3% . Compared lifestyle, relative excess due healthier reduce effect − 0.98 (− 1.52 0.44) , 0.60 1.05 0.14) 1.84 2.59 1.09) 1.44 2.10 0.79) 2, 1.08, 0.12) O Conclusions partially mediated adherence could protect middle-aged elderly people adverse

Язык: Английский

---

Four years of National Clean Air Programme (NCAP) in Indian cities: Assessment of the impact on surface ozone during the period 2018–2022

Sustainable Cities and Society, Год журнала: 2024, Номер 101, С. 105207 - 105207

Опубликована: Янв. 18, 2024

Язык: Английский

---

Spatiotemporal estimation of hourly 2-km ground-level ozone over China based on Himawari-8 using a self-adaptive geospatially local model

Geoscience Frontiers, Год журнала: 2021, Номер 13(1), С. 101286 - 101286

Опубликована: Авг. 11, 2021

Ground-level ozone (O3) is a primary air pollutant, which can greatly harm human health and ecosystems. At present, data fusion frameworks only provided ground-level O3 concentrations at coarse spatial (e.g., 10 km) or temporal daily) resolutions. As photochemical pollution continues increasing over China in the last few years, high-spatial–temporal-resolution product required to enhance comprehension of formation mechanisms. To address this issue, our study creatively explores brand-new framework for estimating hourly 2-km across (except Xinjiang Tibet) using brightness temperature multiple thermal infrared bands. Considering heterogeneity O3, novel Self-adaptive Geospatially Local scheme based on Categorical boosting (SGLboost) developed train estimation models. Validation results show that SGLboost performs well area, with R2s/RMSEs 0.85/19.041 μg/m3 0.72/25.112 space-based cross-validation (CV) (2017–2019) historical CV (2019), respectively. Meanwhile, achieves distinctly better metrics than those some widely used machine learning methods, such as eXtreme Gradient Random Forest. Compared recent related works China, performance also more desired. Regarding distribution, estimated present continuous patterns without significantly partitioned boundary effect. In addition, accurate seasonal variations be observed area. It believed by will help further understand mechanisms China.

Язык: Английский

---

Air quality improvement and cognitive function benefit: Insight from clean air action in China

Environmental Research, Год журнала: 2022, Номер 214, С. 114200 - 114200

Опубликована: Авг. 27, 2022

Epidemiological evidence suggests associations between long-term exposure to air pollution and accelerated cognitive decline. China implemented a strict clean action plan in 2013; however, it is unclear whether the improvement of quality has alleviated impairment population. From Health Retirement Longitudinal Study, 8536 Chinese adults were enrolled 2011 followed up 2015. Satellite-based spatiotemporal models used estimate pollutants (including particles with diameters ≤1.0 μm [PM1], ≤2.5 [PM2.5], ≤10 [PM10], nitrogen dioxide [NO2], ozone [O3]). Cognitive function was evaluated using structured questionnaire three dimensions: episodic memory, orientation attention, visuoconstruction. The changes levels elucidated by logistic model. Bayesian Kernel Machine Regression (BKMR) model applied evaluate cumulative effect pollutants. mean (standard deviation) age all participants 58.6 (8.7) years. odds ratio (95% confidence interval) highest lowest quartile PM1 reduction for 0.46 (0.41, 0.53) after adjusting confounders. Similar protective effects observed decrease level PM2.5 (0.34 [0.30, 0.39]), PM10 (0.54 [0.48, 0.62]), NO2 (0.59 [0.51, 0.67]), while O3 appeared be less related (OR: 0.97 [0.85, 1.10]). association stronger males than females. Decreased dominate benefit relative PM1, PM10, NO2. implementation led significant PM2.5, NO2, which could slow decline function, may not.

Язык: Английский

---

Full-coverage spatiotemporal estimation of surface ozone over China based on a high-efficiency deep learning model

International Journal of Applied Earth Observation and Geoinformation, Год журнала: 2023, Номер 118, С. 103284 - 103284

Опубликована: Апрель 1, 2023

Ozone concentration Monitoring is essential to atmospheric pollution prevention and control. Against the background of severe ozone over China in recent years, a spatiotemporal contiguous mapping method was developed. We imputed significant data gaps Instrument's tropospheric NO2 content by using an efficient machine learning model named LightGBM. Then, we developed deep based on three-dimensional Convolutional Neural Network architecture for daily maximum 8 h average estimation China. With support satellite-retrieved precursor, meteorological other ancillary data, our achieved excellent performance with sample-based 10-fold cross-validation R2 = 0.88. Furthermore, generated datasets covering whole from 2016 2020. This study presents novel surface modeling, which can provide fundamental ecological changes caused pollution, such as crop loss, or harmful effects humans, increased incidence respiratory diseases.

Язык: Английский

---

An improved fusion of Landsat-7/8, Sentinel-2, and Sentinel-1 data for monitoring alfalfa: Implications for crop remote sensing

International Journal of Applied Earth Observation and Geoinformation, Год журнала: 2023, Номер 124, С. 103533 - 103533

Опубликована: Окт. 22, 2023

Precision agriculture management with remote sensing big data provides a promising solution to monitor crops. Alfalfa is an important forage crop for various livestock around the world. Unlike corn and soybean, alfalfa growth difficult describe using typical phenological curve since it characterized by monthly harvest rapid regrowth. Limited availability of high spatio-temporal resolution from single satellite sensor, there have been few studies reported on remotely monitoring alfalfa. Fusion Landsat/Sentinel-2 optical Sentinel-1 Synthetic Aperture Radar (SAR) can improve data. However, form unified SAR image multisource satellites, are still four issues during fusion. To overcome these challenges, this study proposes improved fusion Landsat-7/8, Sentinel-2, The framework includes three models, namely Landsat2Sentinel-2, SAR2Optical, Optical2SAR. results indicate that random forest (RF) algorithm known spectral bands, Landsat2Sentinel-2 model improves Landsat-based surface red edge reflectance root mean square error (RMSE) reduced 28.22–31.16 % compared multiple linear regression models. In addition, RMSE was further 22.61–23.58 if considering solar angles in RF algorithm. Compared traditional model, more accurately fused vegetation indices (VIs) derived Landsat Sentinel-2 46.81–51.16 %. SAR2Optical VIs 32.18–34.59 benefited involved climate Optical2SAR generated optical-based data, proving be reciprocally fused. Moreover, developed models present accuracy fusing at both top-of-atmosphere (TOA) levels. On other hand, proposed used build across different types. Overall, demonstrated has great potential crops satellites.

Язык: Английский

---

Distance adaptive graph convolutional gated network-based smart air quality monitoring and health risk prediction in sensor-devoid urban areas

Sustainable Cities and Society, Год журнала: 2023, Номер 91, С. 104445 - 104445

Опубликована: Фев. 11, 2023

Язык: Английский

---