High-Resolution Spatial Forecasting of Hourly Air Quality: A Fast Method for a Better Representation of Industrial Plumes and Traffic Emissions Contributions
Abstract
:1. Introduction
2. Materials and Methods
2.1. Site Description
2.2. Air Quality Monitoring in the Hauts-De-France Region
2.3. Fine-Scale Regional Concentration Maps Construction
2.3.1. Additivity Hypothesis and Methodology
- The concentration linked to a contribution of road traffic emissions, denoted ;
- The concentration linked to a contribution of industrial emissions, denoted ;
- A part , which corresponds to the residual concentration that does not come from traffic or industrial sectors.
- 1.
- The construction of a background concentration map , resulting from a spatialization of the background concentrations measured at the monitoring stations depicted on Figure 1 (Section 2.3.2 and Section 2.3.3);
- 2.
- The estimation of the traffic and industrial layers and , using a KNN statistical approach applied to the outputs of numerical simulations of regional dispersion with the software ADMS-Urban (Section 2.3.4).
2.3.2. Annual Concentration Maps with ADMS-Urban
2.3.3. Hourly Background Concentration Maps with AZUR
2.3.4. Hourly Traffic and Industrial Contributions with KNN
Prediction from a Sample of Historical Data
Dealing with the Spatial Variability of the Regional Weather
Definition of the Calendar and Meteorological Parameters
- 1.
- To avoid potential confusion associated to the fact that the wind direction (denoted ) varies between 0 and 360 degrees, which both define the same angle, we calculate two new variables to replace the wind angle, defined as and , respectively.
- 2.
- For the sake of simplification, the precipitation rate is categorized into three categories, namely “low”, “medium” and “high” to facilitate the KNN selection. The variable thus takes three values, 1 (for rates between 0 and mm·), 2 (for rates between and 2 mm·) and 3 (for rates above 2 mm·).
- 3.
- To ensure that the coordinates of historical data (the meteorological parameters registered in ADMS-Urban) are comparable with those of the targeted date (the meteorological parameters from WRF data) when computing the Euclidean distance in the KNN procedure, we apply a min-max normalization as follows:Ensuring that all variables lie between 0 and 1 also ensures that the same importance is given to each of them.
3. Results and Discussion
3.1. Industrial Plume Modeling
3.2. Temporal Variability in PM10 and NO2 Concentrations from Traffic Emissions
3.3. High-Resolution Pollution Mapping
3.4. Evaluation
4. Conclusions
Author Contributions
Funding
Data Availability Statement
Acknowledgments
Conflicts of Interest
Abbreviations
ADMS | Atmospheric Dispersion Modeling System |
CERC | Cambridge Environmental Research Consultants |
CFD | computational fluid dynamics |
CTM | chemical transport models |
HDF | Hauts-de-France |
KNNs | k-nearest neighbors |
particulate matter of size less than 10 μm | |
particulate matter of size less than 2.5 μm | |
RMSE | root mean squared error |
WRF | Weather Research Forecast |
References
- Directive (EU) 2024/2881; Ambient Air Quality and Cleaner Air for Europe (Recast). European Parliament, Council of the European Union: Brussels, Belgium, 2024.
- Patel, Z.B.; Purohit, P.; Patel, H.M.; Sahni, S.; Batra, N. Accurate and Scalable Gaussian Processes for Fine-Grained Air Quality Inference. Proc. AAAI Conf. Artif. Intell. 2022, 36, 12080–12088. [Google Scholar] [CrossRef]
- Casallas, A.; Cabrera, A.; Guevara-Luna, M.A.; Tompkins, A.; González, Y.; Aranda, J.; Belalcazar, L.C.; Mogollon-Sotelo, C.; Celis, N.; Lopez-Barrera, E.; et al. Air pollution analysis in Northwestern South America: A new Lagrangian framework. Sci. Total Environ. 2024, 906, 167350. [Google Scholar] [CrossRef]
- William, R. Stockwell, Emily Saunders, W.S.G.; Fitzgerald, R.M. A perspective on the development of gas-phase chemical mechanisms for Eulerian air quality models. J. Air Waste Manag. Assoc. 2020, 70, 44–70. [Google Scholar] [CrossRef]
- Menut, L.; Cholakian, A.; Pennel, R.; Siour, G.; Mailler, S.; Valari, M.; Lugon, L.; Meurdesoif, Y. The CHIMERE chemistry-transport model v2023r1. Geosci. Model Dev. Discuss. 2024, 17, 5431–5457. [Google Scholar] [CrossRef]
- Jacquinot, M.; Derain, R.; Armengaud, A.; Oppo, S. Spatial model for daily air quality high resolution estimation. Air Qual. Atmos. Health 2024, 17, 2141–2150. [Google Scholar] [CrossRef]
- Badawy, H. Recent trends in spatial modeling studies of air quality (2010–2023). Earth Sci. Inform. 2025, 18, 193. [Google Scholar] [CrossRef]
- Wang, Y.; Huang, L.; Huang, C.; Hu, J.; Wang, M. High-resolution modeling for criteria air pollutants and the associated air quality index in a metropolitan city. Environ. Int. 2023, 172, 107752. [Google Scholar] [CrossRef]
- Rybarczyk, Y.; Zalakeviciute, R. Machine Learning Approaches for Outdoor Air Quality Modelling: A Systematic Review. Appl. Sci. 2018, 8, 2570. [Google Scholar] [CrossRef]
- Tang, D.; Zhan, Y.; Yang, F. A review of machine learning for modeling air quality: Overlooked but important issues. Atmos. Res. 2024, 300, 107261. [Google Scholar] [CrossRef]
- Chao, B.; Guang Qiu, H. Air pollution concentration fuzzy evaluation based on evidence theory and the K-nearest neighbor algorithm. Front. Environ. Sci. 2024, 12, 1243962. [Google Scholar] [CrossRef]
- Baran, B. Air Quality Index Prediction in Besiktas District by Artificial Neural Networks and K Nearest Neighbors. MüHendislik Bilim. Ve Tasarım Derg. 2021, 9, 52–63. [Google Scholar] [CrossRef]
- Atmo Sud. Why Do We Need to Change the Air Quality Information Indexes to a HD Mapping Approach? Technical Report; Atmo Sud: Marseille, France, 2024. [Google Scholar]
- Carruthers, D.J.; Edmunds, H.A.; McHugh, C.A.; Singles, R.J. Development of Adms-Urban and Comparison with Data for Urban Areas in the UK. In Air Pollution Modeling and Its Application XII; Springer: Boston, MA, USA, 1998; pp. 467–475. [Google Scholar] [CrossRef]
- Stocker, J.; Hood, C.; Carruthers, D.; Mchugh, C. ADMS-Urban: Developments in modelling dispersion from the city scale to the local scale. Int. J. Environ. Pollut. 2012, 50, 308–316. [Google Scholar] [CrossRef]
- Snoun, H.; Krichen, M.; Chérif, H. A comprehensive review of Gaussian atmospheric dispersion models: Current usage and future perspectives. Euro-Mediterr. J. Environ. Integr. 2023, 8, 219–242. [Google Scholar] [CrossRef]
- Beauchamp, M.; de Fouquet, C.; Malherbe, L. Dealing with non-stationarity through explanatory variables in kriging-based air quality maps. Spat. Stat. 2017, 22, 18–46. [Google Scholar] [CrossRef]
- Babak, O.; Deutsch, C. Statistical approach to inverse distance interpolation. Stoch. Environ. Res. Risk Assess. 2008, 23, 543–553. [Google Scholar] [CrossRef]
- Cover, T.; Hart, P. Nearest neighbor pattern classification. IEEE Trans. Inf. Theory 1967, 13, 21–27. [Google Scholar] [CrossRef]
- Jayaraman, S. Enhancing urban air quality prediction using time-based-spatial forecasting framework. Sci. Rep. 2025, 15, 4139. [Google Scholar] [CrossRef]
- Liu, M.; Lin, J.; Wang, Y.; Sun, Y.; Zheng, B.; Shao, J.; Chen, L.; Zheng, Y.; Chen, J.; Fu, T.M.; et al. Spatiotemporal variability of NO2 and PM2.5 over Eastern China: Observational and model analyses with a novel statistical method. Atmos. Chem. Phys. 2018, 18, 12933–12952. [Google Scholar] [CrossRef]
- Van den Hooven, E.H.; Pierik, F.H.; Van Ratingen, S.W.; Zandveld, P.Y.; Meijer, E.W.; Hofman, A.; Miedema, H.M.; Jaddoe, V.W.; De Kluizenaar, Y. Air pollution exposure estimation using dispersion modelling and continuous monitoring data in a prospective birth cohort study in the Netherlands. Environ. Health 2012, 11, 9. [Google Scholar] [CrossRef]
- Krecl, P.; Castro, L.B.; Targino, A.C.; Oukawa, G.Y. Spatio-temporal variability and trends of air pollutants in the Metropolitan Area of Curitiba. Heliyon 2024, 10, e40651. [Google Scholar] [CrossRef]
- Apte, J.S.; Manchanda, C. High-resolution urban air pollution mapping. Science 2024, 385, 380–385. [Google Scholar] [CrossRef] [PubMed]
- Hammond, J.K.; Chen, R.; Mallet, V. Meta-modeling of a simulation chain for urban air quality. Adv. Model. Simul. Eng. Sci. 2020, 7, 37. [Google Scholar] [CrossRef]
- Pisoni, E.; De Marchi, D.; di Taranto, A.; Bessagnet, B.; Sajani, S.Z.; De Meij, A.; Thunis, P. SHERPA-Cloud: An open-source online model to simulate air quality management policies in Europe. Environ. Model. Softw. 2024, 176, 106031. [Google Scholar] [CrossRef]
Parameter Name | Nature | Unit |
---|---|---|
temperature | meteorological | °C |
atmospheric boundary layer height | meteorological | m |
inverse of Monin–Obukhov length | meteorological | |
wind speed | meteorological | m· |
wind direction () | meteorological | degrees |
relative humidity | meteorological | % |
precipitation rate (pr) | meteorological | mm· |
hour of the day | calendar | – |
month | calendar | – |
day of the week | calendar | – |
Parameter Name | Nature | Unit | Definition |
---|---|---|---|
cosinus of the wind angle | meteorological | – | |
sinus of the wind angle | meteorological | – | |
categorical precipitation rate | meteorological | – |
Pollutant | Influence | Sample Size | Bias | Correlation | RMSE | |
---|---|---|---|---|---|---|
all | 36,891 | 31 | 0.78 | 0.77 | 10.73 | |
back. | 28,113 | 24 | 0.60 | 0.78 | 10.13 | |
traf.+indus. | 8778 | 7 | 1.37 | 0.74 | 12.44 | |
indus. | 3792 | 3 | 3.21 | 0.79 | 11.49 | |
traf. | 4986 | 4 | −0.03 | 0.71 | 13.12 | |
all | 28,914 | 25 | 1.09 | 0.61 | 13.94 | |
back. | 20,404 | 18 | 2.35 | 0.69 | 11.83 | |
traf.+indus. | 8510 | 7 | −1.94 | 0.51 | 18.03 | |
indus. | 4771 | 4 | −3.18 | 0.59 | 9.56 | |
traf. | 3739 | 3 | −0.35 | 0.35 | 24.97 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Rorat, A.; Bouché, L.; Pasquier, M.; Cessey, H.; Pujol-Söhne, N.; Rocq, B. High-Resolution Spatial Forecasting of Hourly Air Quality: A Fast Method for a Better Representation of Industrial Plumes and Traffic Emissions Contributions. Atmosphere 2025, 16, 439. https://doi.org/10.3390/atmos16040439
Rorat A, Bouché L, Pasquier M, Cessey H, Pujol-Söhne N, Rocq B. High-Resolution Spatial Forecasting of Hourly Air Quality: A Fast Method for a Better Representation of Industrial Plumes and Traffic Emissions Contributions. Atmosphere. 2025; 16(4):439. https://doi.org/10.3390/atmos16040439
Chicago/Turabian StyleRorat, Agnieszka, Lucas Bouché, Mathis Pasquier, Hélène Cessey, Nathalie Pujol-Söhne, and Benoit Rocq. 2025. "High-Resolution Spatial Forecasting of Hourly Air Quality: A Fast Method for a Better Representation of Industrial Plumes and Traffic Emissions Contributions" Atmosphere 16, no. 4: 439. https://doi.org/10.3390/atmos16040439
APA StyleRorat, A., Bouché, L., Pasquier, M., Cessey, H., Pujol-Söhne, N., & Rocq, B. (2025). High-Resolution Spatial Forecasting of Hourly Air Quality: A Fast Method for a Better Representation of Industrial Plumes and Traffic Emissions Contributions. Atmosphere, 16(4), 439. https://doi.org/10.3390/atmos16040439