Reconstructing 1-km-resolution high-quality PM2.5 data records from 2000 to 2018 in China: spatiotemporal variations and policy implications

Remote Sensing of Environment, Journal Year: 2020, Volume and Issue: 252, P. 112136 - 112136

Published: Oct. 30, 2020

Language: Английский

---

Improved 1 km resolution PM_2.5 estimates across China using enhanced space–time extremely randomized trees

Atmospheric chemistry and physics, Journal Year: 2020, Volume and Issue: 20(6), P. 3273 - 3289

Published: March 19, 2020

Abstract. Fine particulate matter with aerodynamic diameters ≤2.5 µm (PM2.5) has adverse effects on human health and the atmospheric environment. The estimation of surface PM2.5 concentrations made intensive use satellite-derived aerosol products. However, it been a great challenge to obtain high-quality high-resolution data from both ground satellite observations, which is essential monitor air pollution over small-scale areas such as metropolitan regions. Here, space–time extremely randomized trees (STET) model was enhanced by integrating updated spatiotemporal information additional auxiliary improve spatial resolution overall accuracy estimates across China. To this end, newly released Moderate Resolution Imaging Spectroradiometer Multi-Angle Implementation Atmospheric Correction AOD product, along meteorological, topographical land-use emissions, input STET model, daily 1 km maps for 2018 covering mainland China were produced. performed well, high out-of-sample (out-of-station) cross-validation coefficient determination (R2) 0.89 (0.88), low root-mean-square error 10.33 (10.93) µg m−3, small mean absolute 6.69 (7.15) m−3 relative 21.28 % (23.69 %). In particular, captured well at regional individual site scales. North Plain, Sichuan Basin Xinjiang Province always featured levels, especially in winter. outperformed most models presented previous related studies, strong predictive power (e.g., monthly R2=0.80), can be used estimate historical records. More importantly, study provides new approach obtaining dataset (i.e., ChinaHighPM2.5), important studies focused urban areas.

Language: Английский

---

The ChinaHighPM10 dataset: generation, validation, and spatiotemporal variations from 2015 to 2019 across China

Environment International, Journal Year: 2020, Volume and Issue: 146, P. 106290 - 106290

Published: Dec. 11, 2020

Respirable particles with aerodynamic diameters ≤ 10 µm (PM10) have important impacts on the atmospheric environment and human health. Available PM10 datasets coarse spatial resolutions, limiting their applications, especially at city level. A tree-based ensemble learning model, which accounts for spatiotemporal information (i.e., space-time extremely randomized trees, denoted as STET model), is designed to estimate near-surface concentrations. The 1-km resolution Multi-Angle Implementation of Atmospheric Correction (MAIAC) aerosol product auxiliary factors, including meteorology, land-use cover, surface elevation, population distribution, pollutant emissions, are used in model generate high-resolution (1 km) high-quality dataset China ChinaHighPM10) from 2015 2019. has an out-of-sample (out-of-station) cross-validation coefficient determination (CV-R2) 0.86 (0.82) a root-mean-square error (RMSE) 24.28 (27.07) μg/m3, outperforming most widely models previous related studies. High levels concentration occurred northwest (e.g., Tarim Basin) Northern Plain. Overall, concentrations had significant declining trend 5.81 μg/m3 per year (p < 0.001) over past five years China, three key urban agglomerations. ChinaHighPM10 potentially useful future small- medium-scale air pollution studies by virtue its higher overall accuracy.

Language: Английский

---

Machine Learning in Environmental Research: Common Pitfalls and Best Practices

Environmental Science & Technology, Journal Year: 2023, Volume and Issue: 57(46), P. 17671 - 17689

Published: June 29, 2023

Machine learning (ML) is increasingly used in environmental research to process large data sets and decipher complex relationships between system variables. However, due the lack of familiarity methodological rigor, inadequate ML studies may lead spurious conclusions. In this study, we synthesized literature analysis with our own experience provided a tutorial-like compilation common pitfalls along best practice guidelines for research. We identified more than 30 key items evidence-based based on 148 highly cited articles exhibit misconceptions terminologies, proper sample size feature size, enrichment selection, randomness assessment, leakage management, splitting, method selection comparison, model optimization evaluation, explainability causality. By analyzing good examples supervised reference modeling paradigms, hope help researchers adopt rigorous preprocessing development standards accurate, robust, practicable uses applications.

Language: Английский

---

Ground-level gaseous pollutants (NO₂, SO₂, and CO) in China: daily seamless mapping and spatiotemporal variations

Atmospheric chemistry and physics, Journal Year: 2023, Volume and Issue: 23(2), P. 1511 - 1532

Published: Jan. 26, 2023

Abstract. Gaseous pollutants at the ground level seriously threaten urban air quality environment and public health. There are few estimates of gaseous that spatially temporally resolved continuous across China. This study takes advantage big data artificial-intelligence technologies to generate seamless daily maps three major ambient pollutant gases, i.e., NO2, SO2, CO, China from 2013 2020 a uniform spatial resolution 10 km. Cross-validation between our observations illustrated high on basis for surface CO concentrations, with mean coefficients determination (root-mean-square errors) 0.84 (7.99 µg m−3), (10.7 0.80 (0.29 mg respectively. We found COVID-19 lockdown had sustained impacts pollutants, where recovered its normal in around 34th day after Lunar New Year, while SO2 NO2 rebounded more than 2 times slower due emissions residents' increased indoor cooking atmospheric oxidation capacity. Surface reached their peak annual concentrations 21.3 ± 8.8 m−3, 23.1 13.3 1.01 0.29 m−3 2013, then continuously declined over time by 12 %, 55 17 respectively, until 2020. The declining rates were prominent 2017 sharper reductions anthropogenic but have slowed down recent years. Nevertheless, people still suffer high-frequency risk exposure eastern China, almost World Health Organization (WHO) recommended short-term guidelines (AQG) since 2018, benefiting implemented stricter “ultra-low” emission standards. reconstructed dataset will benefit future (especially short-term) pollution environmental health-related studies.

Language: Английский

---

Himawari-8-derived diurnal variations in ground-level PM&lt;sub&gt;2.5&lt;/sub&gt; pollution across China using the fast space-time Light Gradient Boosting Machine (LightGBM)

Atmospheric chemistry and physics, Journal Year: 2021, Volume and Issue: 21(10), P. 7863 - 7880

Published: May 25, 2021

Abstract. Fine particulate matter with a diameter of less than 2.5 µm (PM2.5) has been used as an important atmospheric environmental parameter mainly because its impact on human health. PM2.5 is affected by both natural and anthropogenic factors that usually have strong diurnal variations. Such information helps toward understanding the causes air pollution, well our adaptation to it. Most existing products derived from polar-orbiting satellites. This study exploits use next-generation geostationary meteorological satellite Himawari-8/AHI (Advanced Himawari Imager) document variation in PM2.5. Given huge volume data, based idea gradient boosting, highly efficient tree-based Light Gradient Boosting Machine (LightGBM) method involving spatiotemporal characteristics namely space-time LightGBM (STLG) model, developed. An hourly dataset for China (i.e., ChinaHighPM2.5) at 5 km spatial resolution aerosol additional variables. Hourly estimates (number data samples = 1 415 188) are correlated ground measurements (cross-validation coefficient determination, CV-R2 0.85), root-mean-square error (RMSE) mean absolute (MAE) 13.62 8.49 µg m−3, respectively. Our model captures variations showing pollution increases gradually morning, reaching peak about 10:00 LT (GMT+8), then decreases steadily until sunset. The proposed approach outperforms most traditional statistical regression machine-learning models much lower computational burden terms speed memory, making it suitable routine monitoring.

Language: Английский

---

The significant impact of aerosol vertical structure on lower atmosphere stability and its critical role in aerosol–planetary boundary layer (PBL) interactions

Atmospheric chemistry and physics, Journal Year: 2020, Volume and Issue: 20(6), P. 3713 - 3724

Published: March 27, 2020

Abstract. The aerosol–planetary boundary layer (PBL) interaction was proposed as an important mechanism to stabilize the atmosphere and exacerbate surface air pollution. Despite tremendous progress made in understanding this process, its magnitude significance still have large uncertainties vary largely with aerosol distribution meteorological conditions. In study, we focus on role of vertical thermodynamic stability PBL development by jointly using micropulse lidar, sun photometer, radiosonde measurements taken Beijing. complexity distributions, cloud-free structures can be classified into three types: well-mixed, decreasing height, inverse structures. aerosol–PBL relationship diurnal cycles height PM2.5 associated these different show distinct characteristics. radiative forcing differs drastically among types, strong heating lower, middle, upper PBL, respectively. Such a discrepancy rate affects atmospheric buoyancy differently Absorbing aerosols weaker effect stabilizing lower under structure than structure. As result, strengthened potentially neutralized Moreover, both enhance suppress stability, leading positive negative feedback loops. This study attempts improve our interaction, showing importance observational constraint for simulating consequent feedbacks.

Language: Английский

---

Abnormally Shallow Boundary Layer Associated With Severe Air Pollution During the COVID‐19 Lockdown in China

Geophysical Research Letters, Journal Year: 2020, Volume and Issue: 47(20)

Published: Oct. 1, 2020

After the 2020 Lunar New Year, Chinese government implemented a strict nationwide lockdown to inhibit spread of Coronavirus Disease 2019 (COVID-19). Despite abrupt decreases in gaseous emissions caused by record-low anthropogenic activities, severe haze pollution occurred northern China during COVID lockdown. This paradox has attracted attention both public and scientific community. By analyzing comprehensive measurements air pollutants, planetary boundary layer (PBL) height, surface meteorology, we show that episode over coincided with abnormally low PBL which had reduced 45%, triggering strong aerosol-PBL interactions. dynamical processes initiated temperature inversion, Beijing metropolitan area experienced period continuously shallow PBLs unprecedented event provided an experiment showcasing role particular interactions affecting quality.

Language: Английский

---

Satellite remote sensing of atmospheric particulate matter mass concentration: Advances, challenges, and perspectives

Fundamental Research, Journal Year: 2021, Volume and Issue: 1(3), P. 240 - 258

Published: May 1, 2021

Mapping the mass concentration of near-surface atmospheric particulate matter (PM) using satellite observations has become a popular research niche, leading to development variety instruments, algorithms, and datasets over past two decades. In this study, we conducted holistic review major advances challenges in quantifying PM, with specific focus on datasets, modeling methods that have been developed 20 years. The aim study is provide general guide for future satellite-based PM mapping practices better support air quality monitoring management environmental health. Specifically, evolution platforms, sensors, inversion can be used aerosol properties. We then compare various practical techniques estimate concentrations group them into four primary categories: (1) univariate regression, (2) chemical transport models (CTM), (3) multivariate (4) empirical physical approaches. Considering main encountered practices, example, data gaps discontinuity, hybrid method proposed generating maps are both spatially continuous high precision.

Language: Английский

---

Significant contribution of organics to aerosol liquid water content in winter in Beijing, China

Atmospheric chemistry and physics, Journal Year: 2020, Volume and Issue: 20(2), P. 901 - 914

Published: Jan. 23, 2020

Abstract. The aerosol liquid water (ALW) content (ALWC), an important component of atmospheric particles, has a significant effect on optical properties, visibility and multiphase chemical reactions. In this study, ALWC is determined from hygroscopic growth factor (GF) particle number size distribution (PNSD) measurements also simulated by ISORROPIA II, thermodynamic equilibrium model, with measured composition data taken at urban site in Beijing 8 November to 15 December 2017. Rich made during the experiment concerning virtually all properties allow us not only derive but study contributions various species for which little been done region. including contribution organics calculated are highly correlated (coefficient determination R2=0.92). contributed (ALWCOrg) accounts 30 %±22 % total sampling period. These results suggest ALWC, rather different previous studies that showed negligible organics. Our show correlates well mass concentrations sulfate, nitrate, secondary organic aerosols (SOAs) (R2=0.66, 0.56 0.60, respectively). We further noted accumulation mode particles play key role determining dominating among modes. exponential function ambient relative humidity (RH), whose strong diurnal variation influence ALWC. However, there 3 h lag between extremes RH values, due variations PNSD composition. Finally, case reveals ALWCOrg plays formation through reactions initial stage heavy-haze episode.

Language: Английский

---