1
|
Wei P, Hao S, Shi Y, Anand A, Wang Y, Chu M, Ning Z. Combining Google traffic map with deep learning model to predict street-level traffic-related air pollutants in a complex urban environment. ENVIRONMENT INTERNATIONAL 2024; 191:108992. [PMID: 39250881 DOI: 10.1016/j.envint.2024.108992] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 03/28/2024] [Revised: 08/26/2024] [Accepted: 08/29/2024] [Indexed: 09/11/2024]
Abstract
BACKGROUND Traffic-related air pollution (TRAP) is a major contributor to urban pollution and varies sharply at the street level, posing a challenge for air quality modeling. Traditional land use regression models combined with data from fixed monitoring stations may be unable to predict and characterize fine-scale TRAP, especially in complex urban environments influenced by various features. This study aims to estimate fine-scale (50 m) concentrations of nitrogen oxides (NO and NO₂) in Hong Kong using a deep learning (DL) structured model. METHODS We collected data from mobile air quality sensors on buses and crowd-sourced Google real-time traffic status as a proxy for real-time traffic emissions. Our DL model was compared with existing machine learning models to assess performance improvements. Using an interpretable machine learning method, we hierarchically evaluated the global, local, and interaction effects for different features. RESULTS Our DL model outperformed existing machine learning models, achieving R2 values of 0.72 for NO and 0.69 for NO₂. The incorporation of traffic status as a key predictor improved model performance by 9% to 17%. The interpretable machine learning method revealed the importance of traffic-related features and their pairwise interactions. CONCLUSION The results indicate that traffic-related features significantly contribute to TRAP and provide insights and guidance for urban planning. By incorporating crowd-sourced Google traffic information, we assessed traffic abatement scenarios that could inform targeted strategies for improving urban air quality.
Collapse
Affiliation(s)
- Peng Wei
- College of Geography and Environment, Shandong Normal University, Jinan, China; Division of Environment and Sustainability, The Hong Kong University of Science and Technology, Hong Kong, China
| | - Song Hao
- State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, China.
| | - Yuan Shi
- Department of Geography & Planning, University of Liverpool, Liverpool, UK.
| | - Abhishek Anand
- Department of Mechanical Engineering, Carnegie Mellon University, United States
| | - Ya Wang
- Division of Environment and Sustainability, The Hong Kong University of Science and Technology, Hong Kong, China
| | - Mengyuan Chu
- Division of Environment and Sustainability, The Hong Kong University of Science and Technology, Hong Kong, China
| | - Zhi Ning
- Division of Environment and Sustainability, The Hong Kong University of Science and Technology, Hong Kong, China.
| |
Collapse
|
2
|
Jin MY, Gallagher J, Li XB, Lu KF, Peng ZR, He HD. Characterizing the distribution pattern of traffic-related air pollutants in near-road neighborhoods. ENVIRONMENTAL MONITORING AND ASSESSMENT 2024; 196:767. [PMID: 39073498 DOI: 10.1007/s10661-024-12917-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 02/03/2024] [Accepted: 07/11/2024] [Indexed: 07/30/2024]
Abstract
In near-road neighborhoods, residents are more frequently exposed to traffic-related air pollution (TRAP), and they are increasingly aware of pollution levels. Given this consideration, this study adopted portable air pollutant sensors to conduct a mobile monitoring campaign in two near-road neighborhoods, one in an urban area and one in a suburban area of Shanghai, China. The campaign characterized spatiotemporal distributions of fine particulate matter (PM2.5) and black carbon (BC) to help identify appropriate mitigation measures in these near-road micro-environments. The study identified higher mean TRAP concentrations (up to 4.7-fold and 1.7-fold higher for PM2.5 and BC, respectively), lower spatial variability, and a stronger inter-pollutant correlation in winter compared to summer. The temporal variations of TRAP between peak hour and off-peak hour were also investigated. It was identified that district-level PM2.5 increments occurred from off-peak to peak hours, with BC concentrations attributed more to traffic emissions. In addition, the spatiotemporal distribution of TRAP inside neighborhoods revealed that PM2.5 concentrations presented great temporal variability but almost remained invariant in space, while the BC concentrations showed notable spatiotemporal variability. These findings provide valuable insights into the unique spatiotemporal distributions of TRAP in different near-road neighborhoods, highlighting the important role of hyperlocal monitoring in urban micro-environments to support tailored designing and implementing appropriate mitigation measures.
Collapse
Affiliation(s)
- Meng-Yi Jin
- Center for Intelligent Transportation Systems and Unmanned Aerial Systems Applications Research, School of Naval Architecture, Ocean and Civil Engineering, State Key Laboratory of Ocean Engineering, Shanghai Jiao Tong University, Shanghai, 200240, China
- Department of Civil, Structural & Environmental Engineering, Trinity College Dublin, The University of Dublin, Dublin, D02 PN40, Ireland
| | - John Gallagher
- Department of Civil, Structural & Environmental Engineering, Trinity College Dublin, The University of Dublin, Dublin, D02 PN40, Ireland
| | - Xiao-Bing Li
- Institute for Environmental and Climate Research, Jinan University, Guangzhou, 510632, China
| | - Kai-Fa Lu
- iAdapt: International Center for Adaptation Planning and Design, College of Design, Construction and Planning, University of Florida, Gainesville, FL, 32611-5706, USA
| | - Zhong-Ren Peng
- iAdapt: International Center for Adaptation Planning and Design, College of Design, Construction and Planning, University of Florida, Gainesville, FL, 32611-5706, USA.
- Healthy Building Research Center, Ajman University, Ajman, United Arab Emirates.
| | - Hong-Di He
- Center for Intelligent Transportation Systems and Unmanned Aerial Systems Applications Research, School of Naval Architecture, Ocean and Civil Engineering, State Key Laboratory of Ocean Engineering, Shanghai Jiao Tong University, Shanghai, 200240, China.
| |
Collapse
|
3
|
Pan J, Li X, Zhu S. High-resolution estimation of near-surface ozone concentration and population exposure risk in China. ENVIRONMENTAL MONITORING AND ASSESSMENT 2024; 196:249. [PMID: 38340249 DOI: 10.1007/s10661-024-12416-5] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 10/14/2023] [Accepted: 01/29/2024] [Indexed: 02/12/2024]
Abstract
Considering the spatial and temporal effects of atmospheric pollutants, using the geographically and temporally weighted regression and geo-intelligent random forest (GTWR-GeoiRF) model and Sentinel-5P satellite remote sensing data, combined with meteorological, emission inventory, site observation, population, elevation, and other data, the high-precision ozone concentration and its spatiotemporal distribution near the ground in China from March 2020 to February 2021 were estimated. On this basis, the pollution status, near-surface ozone concentration, and population exposure risk were analyzed. The findings demonstrate that the estimation outcomes of the GTWR-GeoiRF model have high precision, and the precision of the estimation results is higher compared with that of the non-hybrid model. The downscaling method enhances estimation results to some extent while addressing the issue of limited spatial resolution in some data. China's near-surface ozone concentration distribution in space shows obvious regional and seasonal characteristics. The eastern region has the highest ozone concentrations and the lowest in the northeastern region, and the wintertime low is higher than the summertime high. There are significant differences in ozone population exposure risks, with the highest exposure risks being found in China's eastern region, with population exposure risks mostly ranging from 0.8 to 5.
Collapse
Affiliation(s)
- Jinghu Pan
- College of Geography and Environmental Science, Northwest Normal University, No. 967 Anning East Road, Lanzhou, Gansu Province, People's Republic of China.
| | - Xuexia Li
- College of Geography and Environmental Science, Northwest Normal University, No. 967 Anning East Road, Lanzhou, Gansu Province, People's Republic of China
| | - Shixin Zhu
- College of Geography and Environmental Science, Northwest Normal University, No. 967 Anning East Road, Lanzhou, Gansu Province, People's Republic of China
| |
Collapse
|
4
|
Bortoluzzi MG, Neckel A, Bodah BW, Cardoso GT, Oliveira MLS, Toscan PC, Maculan LS, Lozano LP, Bodah ET, Silva LFO. Detection of atmospheric aerosols and terrestrial nanoparticles collected in a populous city in southern Brazil. ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH INTERNATIONAL 2024; 31:3526-3544. [PMID: 38085483 DOI: 10.1007/s11356-023-31414-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 09/11/2023] [Accepted: 12/04/2023] [Indexed: 01/19/2024]
Abstract
The main objective of this study is to analyze hazardous elements in nanoparticles (NPs) (smaller than 100 nm) and ultrafine particles (smaller than 1 µm) in Porto Alegre City, southern Brazil using a self-made passive sampler and Sentinel-3B SYN satellite images in 32 collection points. The Aerosol Optical Thickness proportion (T550) identification was conducted using images of the Sentinel-3B SYN satellite at 634 points sampled in 2019, 2020, 2021, and 2022. Focused ion beam scanning electron microscopy analyses were performed to identify chemical elements present in NPs and ultrafine particles, followed by single-stage cascade impactor to be processed by high-resolution transmission electron microscopy. This process was coupled with energy-dispersive X-ray spectroscopy and later analysis via secondary ion mass spectrometry. Data was acquired from Sentinel-3B SYN images, normalized to a standard mean of 0.83 µg/mg, at moderate spatial resolution (260 m), and modeled in the Sentinel Application Platform (SNAP) software v.8.0. Statistical matrix data was generated in the JASP software (Jeffreys's Amazing Statistics Program) v.0.14.1.0 followed by a K-means cluster analysis. The results demonstrate the presence of between 1 and 100 nm particles of the following chemical elements: Si, Al, K, Mg, P, and Ti. Many people go through these areas daily and may inhale or absorb these elements that can harm human health. In the Sentinel-3B SYN satellite images, the sum of squares in cluster 6 is 168,265 and in cluster 7 a total of 21,583. The use of images from the Sentinel-3B SYN satellite to obtain T550 levels is of great importance as it reveals that atmospheric pollution can move through air currents contaminating large areas on a global scale.
Collapse
Affiliation(s)
| | - Alcindo Neckel
- Atitus Educação, 304 - Villa Rodrigues, Passo Fundo, RS, 99070-220, Brazil.
- University of Minho, UMINHO, 4710-057, Porto, Portugal.
| | - Brian William Bodah
- Thaines and Bodah Center for Education and Development, 840 South Meadowlark Lane, Othello, WA, 99344, USA
- Workforce Education & Applied Baccalaureate Programs, Yakima Valley College, South 16th Avenue & Nob Hill Boulevard, Yakima, WA, 98902, USA
| | | | - Marcos L S Oliveira
- Department of Civil and Environmental Engineering, Universidad de La Costa, CUC, Calle 58 # 55-66, Barranquilla, Atlántico, Colombia
- Santa Catarina Research and Innovation Support Foundation (Fapesc), Florianópolis, SC, 88030-902, Brazil
| | | | | | - Liliana P Lozano
- Department of Civil and Environmental Engineering, Universidad de La Costa, CUC, Calle 58 # 55-66, Barranquilla, Atlántico, Colombia
- Postgraduate Doctoral Program in Society, Nature and Development, Universidade Federal Do Oeste Do Pará, UFOPA, Paraná, 68040-255, Brazil
| | - Eliane Thaines Bodah
- Thaines and Bodah Center for Education and Development, 840 South Meadowlark Lane, Othello, WA, 99344, USA
- State University of New York, Onondaga Community College, 4585West Seneca Turnpike, Syracuse, NY, 13215, USA
| | - Luis F O Silva
- Department of Civil and Environmental Engineering, Universidad de La Costa, CUC, Calle 58 # 55-66, Barranquilla, Atlántico, Colombia
- Postgraduate Doctoral Program in Society, Nature and Development, Universidade Federal Do Oeste Do Pará, UFOPA, Paraná, 68040-255, Brazil
- CDLAC - Data Collection Laboratory and Scientific Analysis LTDA, Nova Santa Rita, 92480-000, Brazil
| |
Collapse
|
5
|
Chen D, Billmire M, Loughner CP, Bredder A, French NHF, Kim HC, Loboda TV. Simulating spatio-temporal dynamics of surface PM 2.5 emitted from Alaskan wildfires. THE SCIENCE OF THE TOTAL ENVIRONMENT 2023; 898:165594. [PMID: 37467978 DOI: 10.1016/j.scitotenv.2023.165594] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/15/2022] [Revised: 07/13/2023] [Accepted: 07/15/2023] [Indexed: 07/21/2023]
Abstract
Wildfire is a major disturbance agent in Arctic boreal and tundra ecosystems that emits large quantities of atmospheric pollutants, including PM2.5. Under the substantial Arctic warming which is two to three times of global average, wildfire regimes in the high northern latitude regions are expected to intensify. This imposes a considerable threat to the health of the people residing in the Arctic regions. Alaska, as the northernmost state of the US, has a sizable rural population whose access to healthcare is greatly limited by a lack of transportation and telecommunication infrastructure and low accessibility. Unfortunately, there are only a few air quality monitoring stations across the state of Alaska, and the air quality of most remote Alaskan communities is not being systematically monitored, which hinders our understanding of the relationship between wildfire emissions and human health within these communities. Models simulating the dispersion of pollutants emitted by wildfires can be extremely valuable for providing spatially comprehensive air quality estimates in areas such as Alaska where the monitoring station network is sparse. In this study, we established a methodological framework that is based on an integration of the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model, the Wildland Fire Emissions Inventory System (WFEIS), and the Arctic-Boreal Vulnerability Experiment (ABoVE) Wildfire Date of Burning (WDoB) dataset, an Arctic-oriented fire product. Through our framework, daily gridded surface-level PM2.5 concentrations for the entire state of Alaska between 2001 and 2015 for which wildfires are responsible can be estimated. This product reveals the spatio-temporal patterns of the impacts of wildfires on the regional air quality in Alaska, which, in turn, offers a direct line of evidence indicating that wildfire is the dominant driver of PM2.5 concentrations over Alaska during the fire season. Additionally, it provides critical data inputs for research on understanding how wildfires affect human health which creates the basis for the development of effective and efficient mitigation efforts.
Collapse
Affiliation(s)
- Dong Chen
- Department of Geographical Sciences, University of Maryland, College Park, MD, USA.
| | - Michael Billmire
- Michigan Tech Research Institute, Michigan Technological University, Ann Arbor, MI, USA.
| | - Christopher P Loughner
- Air Resources Laboratory, National Oceanic and Atmospheric Administration, College Park, MD, USA.
| | - Allison Bredder
- Department of Geographical Sciences, University of Maryland, College Park, MD, USA.
| | - Nancy H F French
- Michigan Tech Research Institute, Michigan Technological University, Ann Arbor, MI, USA.
| | - Hyun Cheol Kim
- Air Resources Laboratory, National Oceanic and Atmospheric Administration, College Park, MD, USA; Cooperative Institute for Satellite Earth System Studies, University of Maryland, College Park, MD, USA.
| | - Tatiana V Loboda
- Department of Geographical Sciences, University of Maryland, College Park, MD, USA.
| |
Collapse
|
6
|
Lu Z, Guan Y, Shao C, Niu R. Assessing the health impacts of PM 2.5 and ozone pollution and their comprehensive correlation in Chinese cities based on extended correlation coefficient. ECOTOXICOLOGY AND ENVIRONMENTAL SAFETY 2023; 262:115125. [PMID: 37331289 DOI: 10.1016/j.ecoenv.2023.115125] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 02/01/2023] [Revised: 06/02/2023] [Accepted: 06/07/2023] [Indexed: 06/20/2023]
Abstract
The coordinated control of PM2.5 and ozone pollution is becoming more and more important in the current and next stage of Chinese environmental pollution control. Existing studies are unable to provide sufficient quantitative assessments of the correlation of PM2.5 and ozone pollution to support the coordinated control of the two air pollutants. This study develops a systematic method to comprehensively assess the correlation between PM2.5 and ozone pollution, including the evaluation of the impact of two air pollutants on human health and the extended correlation coefficient (ECC) for assessing the bivariate correlation index of PM2.5-ozone pollution in Chinese cities. According to the latest studies on epidemiology conducted in China, we take cardiovascular and cerebrovascular diseases and respiratory diseases as the ozone pollution's health burden when evaluating the health impact of ozone pollution. The results show that the health impact of PM2.5 in China decreases by 25.9 % from 2015 to 2021, while the health impact of ozone increases by 11.8 %. The ECC of 335 cities in China shows an increasing-decreasing trend but has generally increased from 2015 to 2021. The study provides important support for an in-depth understanding of the correlation and development trend of Chinese PM2.5 and ozone pollution by classifying the comprehensive PM2.5-ozone correlation performances of Chinese cities into four types. China or other countries will get better environmental benefits by implementing different coordinated management approaches for different correlative types of regions based on the assessment method in this study.
Collapse
Affiliation(s)
- Zhirui Lu
- College of Environmental Science and Engineering, Nankai University, Tianjin 300350, China
| | - Yang Guan
- Institute of Strategic Planning, Chinese Academy of Environmental Planning, Beijing 100041, China; The Center for Beautiful China, Chinese Academy of Environmental Planning, Beijing 100041, China
| | - Chaofeng Shao
- College of Environmental Science and Engineering, Nankai University, Tianjin 300350, China.
| | - Ren Niu
- Institute of Strategic Planning, Chinese Academy of Environmental Planning, Beijing 100041, China.
| |
Collapse
|