Significant NO₂ Formation in Truck Exhaust Plumes and Its Association with Ambient O₃: Evidence from Extensive Plume-Chasing Measurements

Environmental Science & Technology, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 10, 2025

Vehicle nitrogen oxides (NOx) significantly increase dioxide (NO2) exposure in traffic-related environments. The NO2/NOx ratios are crucial for accurate NO2 modeling and closely linked to public health concerns. In 2020, we used a mobile platform follow test trucks (plume-chasing) that were installed with portable emission measuring system (PEMS) on two restricted driving tracts. Six hundred eighteen exhaust plumes collected through the PEMS-chasing measurements from seven trucks. NOx factors (EFs), ratios, calculated at distinct stages (i.e., tailpipe on-road). A significant reduction EFs (>64%) was observed normal operating after-treatment devices, except equipped diesel particulate filter (DPF). Disparities also found, attributed technologies. measured plume-chasing higher (3–4 times, p < 0.001) than measurements, providing field evidence of substantial formation plumes. We developed quantitative relationship between demonstrated robust correlation (R2 > 0.90). Since plume is not explicitly accounted modeling, (O3–NO2/NOx) could improve estimation when local inventory (tailpipe emissions) available.

Language: Английский

---

Cost-Effective Mapping of Hyperlocal Air Pollution Using Large-Scale Mobile Monitoring and Land-Use Machine Learning

ACS ES&T Air, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 15, 2025

The advent of large-scale mobile monitoring using fast-response instruments has enabled hyperlocal mapping (≤100 m) traffic-related air pollution (TRAP), with important implications for quality management. However, most related studies have been confined within small areas due to the high cost and labor intensity. This study pioneers a cost-effective TRAP method by incorporating land-use machine learning (LUML). Here, over 4.6 million 1 Hz high-frequency measurements (∼1300 h) were collected on part major roadways in Chinese megacity Shenzhen. Unmeasured locations estimated LUML models reduce measurement costs Various ML algorithms varying spatial aggregation segment lengths incorporated optimize model performance. Hyperlocal maps NO, NO2, PM2.5 predicted across entire road network covering 1700 km2. Based our results, LU-RF (random forest) NO NO2 LU-GBM (Gradient Boosting Machine) PM2.5, demonstrated superior Deep models, contrast, did not yield comparable results. Finer partitioning segments improved prediction performance, but worsened that PM2.5. By deployment optimal lengths, accuracy increased 20–80% compared conventional regression models. provides promising approach management cities worldwide.

Language: Английский

---

Using multiple machine learning algorithms to optimize the water quality index model and their applicability

Ecological Indicators, Journal Year: 2025, Volume and Issue: 172, P. 113299 - 113299

Published: March 1, 2025

Language: Английский

---

Europe-wide high-spatial resolution air pollution models are improved by including traffic flow estimates on all roads

Atmospheric Environment, Journal Year: 2024, Volume and Issue: 335, P. 120719 - 120719

Published: July 26, 2024

Road traffic is an important source of noise and air pollution. Modelling pollution therefore requires detailed information on annual average daily (AADT) flows all roads. Europe-wide estimates intensity are however not publicly available. This has hampered previous modelling, used extensively in epidemiological studies morbidity mortality. We aim to estimate AADT quantify potential improvements models. built separate random forests (RF) models for different road types OpenStreetMap (highway, primary, secondary tertiary, residential roads). collected observations from six European countries. evaluated our using 5-fold cross-validation (CV) by comparison flow with national model Switzerland the Netherlands. whether adding estimated as predictors trained more than 2000 routine monitoring sites improved performance based upon major length buffer sizes. The result showed overall captured variations between (R2 = 0.82). Our variability within types, documenting benefit framework at a continental scale. modestly NO2, PM10, PM2.5, O3, especially NO2 (3% improvement geographically-weighted regression model). Improvement was larger urban areas (5% 8% increases R2 O3). Importantly, intra-city near-road were traffic-related resulting roads across Europe will be useful further improving modelling facilitating harmonized Europe.

Language: Английский

---

PM _2.5 exposure disparities persist despite strict vehicle emissions controls in California

Science Advances, Journal Year: 2024, Volume and Issue: 10(37)

Published: Sept. 11, 2024

As policymakers increasingly focus on environmental justice, a key question is whether emissions reductions aimed at addressing air quality or climate change can also ameliorate persistent pollution exposure disparities. We examine evidence from California’s aggressive vehicle control policy 2000 to 2019. find 65% reduction in modeled statewide average PM 2.5 on-road vehicles, yet for people of color and overburdened community residents, relative disparities increased. Light-duty are the main driver disparity, although smaller contributions heavy-duty vehicles especially affect some groups. Our findings suggest that continued trend will likely reduce concentrations absolute disparity but may not without greater attention systemic factors leading this disparity.

Language: Английский

---

Province-Level Decarbonization Potentials for China’s Road Transportation Sector

Environmental Science & Technology, Journal Year: 2024, Volume and Issue: unknown

Published: Oct. 1, 2024

Decarbonizing road transportation is an important task in achieving China's climate goals. Illustrating the mitigation potentials of announced policies and identifying additional strategies for various vehicle fleets are fundamental optimizing future control pathways. Herein, we developed a comprehensive analysis carbon dioxide (CO

Language: Английский

---

U.S. Ambient Air Monitoring Network Has Inadequate Coverage under New PM_2.5 Standard

Environmental Science & Technology Letters, Journal Year: 2024, Volume and Issue: 11(11), P. 1220 - 1226

Published: Oct. 15, 2024

The Clean Air Act (CAA) in the United States relies heavily on regulatory monitoring networks, yet sites are sparsely located, especially among historically disadvantaged communities. For ambient fine particulate matter (PM2.5), we compare air quality data with spatially complete concentrations derived from empirical models to quantify gaps existing U.S. networks capturing concentration hotspots and exposure disparities. Recently, Environmental Protection Agency adopted a more stringent annual-average standard for PM2.5 (9 μg/m3). Here, demonstrate that 44% of urban areas exceeding this new standard─encompassing ∼20 million people─would remain undetected because current network. Crucially, find "uncaptured" hotspots, which contain 2.8 people census tracts misclassified as attainment standard, have substantially higher percentages minority populations (i.e., color, communities, low-income populations) compared overall population. To address these gaps, highlight 10 priority locations could reduce population uncaptured by 67%. Overall, our findings urgent need

Language: Английский

---

Exploring indoor PM2.5 pollution characteristics in Xi'an city and its health implications using interpretable machine learning

Sustainable Horizons, Journal Year: 2025, Volume and Issue: 13, P. 100131 - 100131

Published: Feb. 14, 2025

Language: Английский

---

Socioeconomic Inequalities in PM_2.5 Exposure and Local Source Contributions at Community Scales Using Hyper-Localized Taxi-Based Mobile Monitoring in Xi’an, China

Environmental Science & Technology, Journal Year: 2025, Volume and Issue: unknown

Published: March 12, 2025

The relationship between the socioeconomic status (SES) and PM2.5 exposure is rather inconclusive. We employed taxi-based measurements with 30 m resolution to characterize local source contribution (PM2.5 adjusted concentration) discerned for 2019 winter 2020 summer, in Xi'an. A big data set comprising ∼6 × 106 hourly SES from ∼5000 communities was utilized examine inequalities community-level exposure. Our results indicate that inhabitants lower are more likely be disproportionately exposed compared those higher SES. At least 92% of rural regions reside low areas, whereas a relatively smaller proportion (69–78%) urban regions. has profound impact on during summer than winter. polluted areas concentration accounted 22% 26% total However, residing low-concentration contributed only 12% while 30%. These findings provide valuable insights into SES, highlighting need sophisticated air quality policies alleviate

Language: Английский

---

Enhancing Time Resolution of Ambient VOC Measurement Data by Machine Learning: From One‐Hour to Five Minutes

Journal of Geophysical Research Atmospheres, Journal Year: 2025, Volume and Issue: 130(5)

Published: Feb. 27, 2025

Abstract Atmospheric volatile organic compounds (VOCs) significantly impact the environment and public health, necessitating precise, continuous online monitoring. Currently, VOCs monitoring primarily uses Gas Chromatography‐Mass Spectrometry (GC‐MS) Proton Transfer Reaction‐Time of Flight Mass (PTR‐ToF‐MS). GC‐MS is favored for its accurate compound identification capabilities but limited by lower temporal resolution. Conversely, PTR‐ToF‐MS, while achieving minute‐scale resolution directly ionizing samples, struggles to detect low‐proton‐affinity compounds. Here, based on 5 years long‐term data, we propose Adaptive Convolutional Tree Ensemble (ACTE) model surpass current instruments limitations accurately obtain high‐resolution (5‐min) VOCs. Our results indicate that this consistently achieves robust predictive accuracy across different major species categories, notably R 2 values 0.92 0.89 alkanes alkenes, respectively, which mostly have low‐proton‐affinity. Furthermore, comparing simulations using resolutions in ozone mechanism modeling, found models with higher more comprehensively capture rapidly occurring photochemical reactions, whereas hourly tend overlook many details, potentially leading inaccuracies understanding related mechanisms. This study underscores potential machine learning improve atmospheric pollutants enhance our chemical processes.

Language: Английский

---