A novel spatiotemporal prediction approach to fill air pollution data gaps using mobile sensors, machine learning and citizen science techniques

Aug 4, 2026

Upload by Ezra

A novel machine learning workflow predicts PM2.5 at 30 m road segments with 10-second temporal granularity, addressing gaps in air pollution monitoring. A hybrid dataset is built from an intensive campaign in Selly Oak, Birmingham, combining citizen scientists with low-cost instruments in both static and mobile deployments. Spatial proxy variables, meteorology, and PM properties enable fine-grained PM2.5 analysis, using Standard Random Forest Regression plus sensor and road transferability evaluations. Results improve spatial resolution beyond regulatory monitoring and strengthen exposure assessment.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Abstract

Introduction

Health impacts of PM and need for spatial-temporal variability

University of Birmingham A novel spatiotemporal prediction approach to fill air pollution data gaps using mobile sensors, machine learning and citizen science techniques Baruah, Arunik; Bousiotis, Dimitrios; Damayanti, Seny; Bigi, Alessandro; Ghermandi, Grazia; Ghaffarpasand, O. ; Harrison, Roy M. ; Pope, Francis D. DOI: 10.1038/s41612-024-00859-z License: Creative Commons: Attribution (CC BY) Document Version Publisher's PDF, also known as Version of record Citation for published version (Harvard): Baruah, A, Bousiotis, D, Damayanti, S, Bigi, A, Ghermandi, G, Ghaffarpasand, O, Harrison, RM & Pope, FD 2024, 'A novel spatiotemporal prediction approach to fill air pollution data gaps using mobile sensors, machine learning and citizen science techniques', npj Climate and Atmospheric Science, vol. 7, no. 1, 310. [https://doi.org/10.1038/s41612-024-00859-z](https://doi.org/10.1038/s41612-024-00859-z) Link to publication on Research at Birmingham portal General rights Unless a licence is specified above, all rights (including copyright and moral rights) in this document are retained by the authors and/or the copyright holders. The express permission of the copyright holder must be obtained for any use of this material other than for purposes permitted by law. •Users may freely distribute the URL that is used to identify this publication. •Users may download and/or print one copy of the publication from the University of Birmingham research portal for the purpose of private study or non-commercial research. •User may use extracts from the document in line with the concept of ‘fair dealing’ under the Copyright, Designs and Patents Act 1988 (?) •Users may not further distribute the material nor use it for the purposes of commercial gain. Where a licence is displayed above, please note the terms and conditions of the licence govern your use of this document. When citing, please reference the published version. Take down policy While the University of Birmingham exercises care and attention in making items available there are rare occasions when an item has been uploaded in error or has been deemed to be commercially or otherwise sensitive. If you believe that this is the case for this document, [please contact UBIRA@lists.bham.ac.uk](please contact UBIRA@lists.bham.ac.uk) providing details and we will remove access to the work immediately and investigate. Download date: 03. Aug. 2026 npj | climate and atmospheric science Article Published in partnership with CECCR at King Abdulaziz University [https://doi.org/10.1038/s41612-024-00859-z](https://doi.org/10.1038/s41612-024-00859-z) A novel spatiotemporal prediction approach to ﬁ ll air pollution data gaps using mobile sensors, machine learning and citizen science techniques Check for updates | Arunik Baruah 1,2, Dimitrios Bousiotis 3, Seny Damayanti3, Alessandro Bigi O. Ghaffarpasand3, Roy M. Harrison 3,4 & Francis D. Pope 3 | | 1, Grazia Ghermandi1, | | --- | --- | --- | | Particulate Matter (PM) air pollution poses signiﬁcant threats to public health. We introduce a novel machine learning methodology to predict PM2.5 levels at 30 m long segments along the roads and at a temporal scale of 10 seconds. A hybrid dataset was curated from an intensive PM campaign in Selly Oak, Birmingham, UK, utilizing citizen scientists and low-cost instruments strategically placed in static and mobile settings. Spatially resolved proxy variables, meteorological parameters, and PM properties were integrated, enabling a ﬁne-grained analysis of PM2.5. Calibration involved three approaches: Standard Random Forest Regression, Sensor Transferability and Road Transferability Evaluations. This methodology signiﬁcantly increased spatial resolution beyond what is possible with regulatory monitoring, thereby improving exposure assessments. The ﬁndings underscore the importance of machine learning approaches and citizen science in advancing our understanding

FAQs

What prediction task does the study focus on?

The study predicts PM2.5 concentrations at 30 m long road segments with a temporal scale of 10 seconds.

How is the dataset collected?

It uses a hybrid dataset from an intensive PM campaign in Selly Oak, Birmingham, combining citizen scientists with low-cost instruments in static and mobile settings.

What methods are used to calibrate the model?

Calibration uses Standard Random Forest Regression and evaluates sensor transferability and road transferability.

Aug 4, 2026

Upload by Ezra