Advancing air pollution exposure assessment model: challenges and future perspectives
0 
, Abstract
In recent years, air pollution exposure assessment models have experienced significant advancements, particularly in integrating advanced technologies. However, the intrinsic deficiency of the geostatistical model in existing studies restricted further development of the air pollution exposure model. In this perspective, we summarized several emerging technologies that can overcome the limitations and estimate air pollution exposures with high spatial and temporal resolutions. As these technologies evolve, they are expected to play an increasingly significant role in improving public health and managing environmental challenges.
Keywords
INTRODUCTION OF AIR POLLUTION EXPOSURE MODEL
Air pollution is a complex environmental issue influenced by various factors such as natural and anthropogenic sources, meteorology, and topography. Traditional air quality monitoring typically relies on fixed monitoring stations, as seen in studies like the Harvard Six Cities Study[1]. However, this approach has limitations[2,3]. First, the number of monitoring stations is limited, and populations residing far from these stations are excluded from the study. Second, populations near adjacent stations are assigned the same pollution concentration despite the highly uneven spatial distribution of air pollutants, with significant concentration differences even between points in close proximity[4]. Third, in specific contexts, the spatial differences within a city can be as substantial as those between different cities. Epidemiological research has shown that the disparities in particulate matter air pollution within cities often exceed those observed between cities[5].
As a result, exposure modeling has become a primary method for prediction in large-scale studies[6,7]. High spatiotemporal exposure models, through dynamic data collection and analysis, offer more detailed and accurate information on air pollution exposure[8]. These models can capture the temporal and spatial variations in pollutant concentrations, revealing their distribution characteristics under specific environmental conditions. This is crucial for understanding the health impacts of air pollution, as health outcomes are closely linked to the surrounding air quality. In epidemiological studies, estimating individual-level exposure to air pollution is essential for assessing the relationship between pollution and participants’ health outcomes. This provides new perspectives for scientific research and supports the development of public policy and public health protection. Over the past two decades, significant progress has been made in air pollution exposure modeling, evolving from simple spatial interpolation methods (e.g., kriging, inverse distance weighing) and exposure indicator variables (e.g., traffic intensity at the residential address or distance to a major road) to statistical techniques like land use regression[9-11] and advancing to machine learning approaches that integrate multiple data sources[12]. These advancements have created high-resolution spatiotemporal models at different scales, laying a solid data foundation for air pollution epidemiology research.
LIMITATIONS OF CURRENT SPATIAL MODELS OF AIR POLLUTION EXPOSURE
In air pollution exposure assessment, spatial analysis is a critical tool that leverages geographic information systems (GIS) and spatial statistical techniques to study pollutant distributions and population exposure. However, this approach faces several limitations.
First, the accuracy of spatial data represents a fundamental challenge. Spatial analysis relies on high-quality geographic data to pinpoint pollution sources and affected populations[5], but in practice, data resolution - such as satellite image granularity and the uneven distribution of monitoring stations - often affects the practical application of the model[11].
Second, temporal scale limitations are evident. Air pollution is a dynamic phenomenon influenced by seasonal variations, climatic conditions, and human activities[13,14]. However, spatial analysis typically provides static “snapshots”, making capturing temporal pollutant-level fluctuations difficult. This temporal mismatch can result in exposure assessments failing to reflect actual conditions, as they overlook variations in pollutant concentrations across different periods[11].
Third, the choice of spatial scale significantly impacts the analysis results. Different spatial scales can produce vastly different outcomes. At larger scales, local pollution hotspots may be overlooked, whereas, at finer scales, regional pollution trends may not be effectively captured. Thus, selecting an appropriate spatial scale is crucial for enhancing assessment accuracy[5].
Fourth, uncertainty is a critical issue. Incomplete data, measurement errors, and model assumptions contribute to uncertainty in assessment results[15]. Although advanced statistical models can address such uncertainties to some degree, they also increase model complexity and place higher demands on data quality.
Finally, individual activity patterns play an essential role in exposure assessment. Personal exposure levels are closely linked to the time spent at different locations, individual behaviors, and pollutant concentrations at those locations[16]. However, spatial analysis often fails to account for variations in individual activity patterns, such as differences between weekdays and weekends or exposure variations due to different modes of transportation[17]. This limitation primarily stems from the difficulty of collecting data on individual activities across time and space, as well as pollutant concentrations in microenvironments.
Based on these limitations, this study highlights current technological advancements in exposure monitoring techniques that offer potential directions for the development of air pollution exposure models.
DEVELOPMENT OF AIR POLLUTION EXPOSURE MONITORING TECHNIQUES
Personal monitoring based on wearable devices
In the context of rapid technological advancement, wearable devices have increasingly become an integral part of daily life, particularly in the fields of personal health and environmental monitoring. These devices enable users to access real-time health data, track physical activity, and monitor environmental air quality, thereby improving quality of life and driving innovation in health management. The application of wearable devices in health monitoring is extensive. Smartwatches and fitness trackers, such as the Huawei Watch, Apple Watch, and Fitbit, offer features like heart rate monitoring, sleep analysis, and step counting to help users manage their health[18-20]. Additionally, these devices can track running distance, duration, and calorie expenditure, assisting athletes in planning their training scientifically and reducing the risk of injury[21].
As public awareness of air quality grows, some wearable devices have started to integrate environmental monitoring functions, allowing users to track the levels of air pollution, such as PM2.5 and carbon dioxide concentrations, in real time. This feature is precious for individuals living in highly polluted areas, as it provides timely alerts for preventive actions[22]. In addition, most people spend their time indoors, so their exposure is driven by the exposure in the home, office, school, etc. However, few models ever try to account for these drivers of exposure. The wearable devices would provide a good opportunity for model development.
However, wearable devices face data privacy and security challenges, as well as issues concerning their accuracy and reliability. Manufacturers need to enhance data protection measures, and users should exercise caution by consulting professional medical advice when using these devices.
Wearable devices have opened new avenues for personal health and environmental monitoring, contributing to smart and personalized health management. In the future, they are expected to play an even more significant role in addressing health and environmental challenges.
Human mobility
Applying population mobility big data in air pollution exposure assessment is increasingly becoming an important research direction in environmental science and public health[23]. This technology leverages data sources such as mobile devices[24-26] and location-based services[27,28] to track population activity patterns and locations in real time, providing precise information for assessing exposure to air pollution. By analyzing the movement trajectories of populations, researchers can gain insights into the amount of time individuals spend in different environments and the frequency of their activities, enabling the evaluation of how air quality in specific areas affects the population[29]. This approach is more flexible and efficient than traditional monitoring methods, offering timelier risk assessments.
Furthermore, by integrating meteorological data and air quality monitoring networks, mobility data can facilitate the development of more complex models to predict exposure levels under different temporal and spatial conditions. By optimizing urban planning, traffic management, and health education measures, it is possible to reduce population exposure in high-pollution areas.
In summary, population mobility data provide new perspectives and methodologies for air pollution exposure assessment, fostering more profound research into environmental and health issues. As data collection and analysis technologies continue to advance, the potential applications in this field will expand, offering significant support for improving public health and enhancing quality of life.
Low-cost sensors
Emerging low-cost sensors have the potential to significantly alter how, where, and when air pollution monitoring is done[30]. Low-cost sensing technologies offer benefits in enhancing the spatial resolution of air pollution measurements and supplementing regulatory data at a significantly lower cost[31,32]. The advantages of low-cost sensor technologies have opened up new avenues for air pollutant exposure modeling. When deployed in dense arrays, these low-cost sensors can deliver near real-time data on air pollutants with a spatial resolution that reflects neighborhood dynamics. They can reveal the impact of local pollution sources over various temporal and spatial dimensions often overlooked by the typically sparse regulatory monitoring systems[33-35]. Consequently, the increasing availability of data from these sensor networks has spurred research on integrating continuous measurements from low-cost sensors with land use information to produce comprehensive air quality data across space and time[36].
Mobile monitoring technologies
Vehicle-based mobile observation technology has made innovative progress in air pollution research in recent years[37,38]. Compared to fixed monitoring stations, this technology offers higher spatial coverage and flexibility, allowing for a more detailed capture of pollution variations across urban streets. By equipping vehicles with air quality monitoring devices, air pollution data can be collected in real time at different times and locations, including concentrations of harmful gases such as particulate matter, nitrogen oxides, and ozone, as well as some unregulated pollutants (ultrafine particles, black carbon, volatile organic compounds)[39,40]. Onboard monitoring provides real-time data and covers areas difficult for traditional monitoring stations to reach, such as street canyons, busy intersections, and commercial and residential districts. These high-resolution data offer new perspectives for understanding the complexity and dynamic changes in urban air quality[40,41].
When constructing air pollution exposure models, the data obtained from mobile observation technologies reveal the spatiotemporal distribution characteristics of pollutants, which is critical for accurately assessing population exposure levels. The high temporal resolution data collected by vehicle-based monitoring can highlight the sharp fluctuations in pollutant concentrations during peak traffic hours, helping researchers identify hotspots at specific times and locations. This information is crucial for developing targeted intervention measures and policies, effectively guiding resource allocation, and implementing pollution control strategies.
Artificial intelligenc and machine learning
Applying machine learning and artificial intelligence (AI) technologies has significantly enhanced the understanding and performance of air pollution exposure models[42]. These technologies help researchers and policymakers accurately assess the impact of air pollution on human health and ecosystems by processing and analyzing large volumes of environmental data. Machine learning algorithms can learn patterns from historical air quality data, identifying the sources and trends of pollutants. Machine learning models can construct complex air pollution prediction models using diverse data sources, such as meteorological data, traffic flow, industrial emissions, and satellite remote sensing.
More advanced AI techniques also show significant potential in exposure models[43,44]. These technologies can automatically extract features from unlabeled, complex data, helping to identify and classify pollution sources, thereby improving the accuracy of air pollution source apportionment. AI technologies can offer more precise pollution predictions and spatial distribution analyses by integrating various data sources, such as real-time sensor data, social media information, and traffic data.
CONCLUSION
The rapid advancement of cutting-edge monitoring technologies, coupled with the exponential growth of high-resolution environmental data, has revolutionized the field of air pollution exposure assessment. These developments have enabled unprecedented precision in characterizing ambient air pollutant exposure, offering transformative opportunities for public health research and policy interventions. However, the integration of traditional exposure models with emerging technologies remains a critical challenge, requiring innovative approaches to bridge the gap between conventional methodologies and modern data-driven paradigms. Future research must prioritize the development of sophisticated model fusion frameworks that effectively combine the strengths of mechanistic models with machine learning algorithms and AI capabilities. This integration should be complemented by advancements in big data processing, including edge computing for real-time analysis and federated learning for secure data sharing. Furthermore, the construction of next-generation exposure models should incorporate multi-omics data, personal exposure monitoring, and socio-economic factors to create a comprehensive exposure assessment ecosystem. Such advanced models will not only enhance our understanding of the complex interactions between air pollution and human health but also enable predictive capabilities for early warning systems and personalized exposure mitigation strategies. Ultimately, these technological advancements promise to support more informed decision making and evidence-based policy formulation, potentially leading to transformative changes in urban planning, transportation systems, and public health interventions to mitigate the global burden of air pollution.
DECLARATIONS
Authors’ contributions
Conceptualization, methodology, investigation, writing - original draft, writing - review and editing, project administration, funding acquisition: Han, B.
Methodology, investigation, writing - original draft, writing - review and editing: Xu, J.
Conceptualization, methodology, investigation, writing - review and editing: Zhang, K.
Availability of data and materials
Not applicable.
Financial support and sponsorship
This work was supported by the National Key R&D Program of China (2019YFE0115100) and the National Natural Science Foundation of China (42277429).
Conflicts of interest
All authors declared that there are no conflicts of interest.
Ethical approval and consent to participate
Not applicable.
Consent for publication
Not applicable.
Copyright
© The Author(s) 2025.
REFERENCES
1. Dockery, D. W.; Pope, C. A.; Xu, X.; et al. An association between air pollution and mortality in six U.S. cities. N. Engl. J. Med. 1993, 329, 1753-9.
2. Vedal, S.; Han, B.; Xu, J.; Szpiro, A.; Bai, Z. Design of an air pollution monitoring campaign in beijing for application to cohort health studies. Int. J. Environ. Res. Public. Health. 2017, 14, 1580.
3. Özkaynak, H.; Baxter, L. K.; Dionisio, K. L.; Burke, J. Air pollution exposure prediction approaches used in air pollution epidemiology studies. J. Expo. Sci. Environ. Epidemiol. 2013, 23, 566-72.
4. Xie, X.; Semanjski, I.; Gautama, S.; et al. A review of urban air pollution monitoring and exposure assessment methods. IJGI. 2017, 6, 389.
5. Miller, K. A.; Siscovick, D. S.; Sheppard, L.; et al. Long-term exposure to air pollution and incidence of cardiovascular events in women. N. Engl. J. Med. 2007, 356, 447-58.
6. Eeftens, M.; Tsai, M.; Ampe, C.; et al. Spatial variation of PM2.5, PM10, PM2.5 absorbance and PMcoarse concentrations between and within 20 European study areas and the relationship with NO2 - Results of the ESCAPE project. Atmos. Environ. 2012, 62, 303-17.
7. Cyrys, J.; Eeftens, M.; Heinrich, J.; et al. Variation of NO2 and NOx concentrations between and within 36 European study areas: results from the ESCAPE study. Atmos. Environ. 2012, 62, 374-90.
8. Keller, J. P.; Olives, C.; Kim, S. Y.; et al. A unified spatiotemporal modeling approach for predicting concentrations of multiple air pollutants in the multi-ethnic study of atherosclerosis and air pollution. Environ. Health. Perspect. 2015, 123, 301-9.
9. Hoek, G.; Beelen, R.; de Hoogh, K.; et al. A review of land-use regression models to assess spatial variation of outdoor air pollution. Atmos. Environ. 2008, 42, 7561-78.
10. Han, B.; Hu, L.; Bai, Z. Human exposure assessment for air pollution. In: Dong G, editor. Ambient air pollution and health impact in China. Singapore: Springer; 2017. pp. 27-57.
11. Ma, X.; Zou, B.; Deng, J.; et al. A comprehensive review of the development of land use regression approaches for modeling spatiotemporal variations of ambient air pollution: a perspective from 2011 to 2023. Environ. Int. 2024, 183, 108430.
12. Bellinger, C.; Mohomed, J. M. S.; Zaïane, O.; Osornio-Vargas, A. A systematic review of data mining and machine learning for air pollution epidemiology. BMC. Public. Health. 2017, 17, 907.
13. Zhang, Z.; Zhang, G.; Su, B. The spatial impacts of air pollution and socio-economic status on public health: empirical evidence from China. Socio. Econ. Plan. Sci. 2022, 83, 101167.
14. Kuerban, M.; Waili, Y.; Fan, F.; et al. Spatio-temporal patterns of air pollution in China from 2015 to 2018 and implications for health risks. Environ. Pollut. 2020, 258, 113659.
15. Wu, Y.; Xu, J.; Liu, Z.; Han, B.; Yang, W.; Bai, Z. Comparison of population-weighted exposure estimates of air pollutants based on multiple geostatistical models in Beijing, China. Toxics 2024, 12, 197.
16. Chatzidiakou, L.; Krause, A.; Kellaway, M.; et al. Automated classification of time-activity-location patterns for improved estimation of personal exposure to air pollution. Environ. Health. 2022, 21, 125.
17. Yu, X.; Ivey, C.; Huang, Z.; et al. Quantifying the impact of daily mobility on errors in air pollution exposure estimation using mobile phone location data. Environ. Int. 2020, 141, 105772.
18. Liu, C.; Tai, M.; Hu, J.; et al. Application of smart devices in investigating the effects of air pollution on atrial fibrillation onset. NPJ. Digit. Med. 2023, 6, 42.
19. Li, A.; Zhang, Q.; Yao, Y.; et al. Higher ambient temperatures may worsen obstructive sleep apnea: a nationwide smartwatch-based analysis of 6.2 million person-days. Sci. Bull. 2024, 69, 2114-21.
20. Zhang, Q.; Wang, H.; Zhu, X.; et al. Air pollution may increase the sleep apnea severity: a nationwide analysis of smart device-based monitoring. Innovation 2023, 4, 100528.
21. Glasgow, M. L.; Rudra, C. B.; Yoo, E. H.; et al. Using smartphones to collect time-activity data for long-term personal-level air pollution exposure assessment. J. Expo. Sci. Environ. Epidemiol. 2016, 26, 356-64.
22. Bernasconi, S.; Angelucci, A.; Aliverti, A. A scoping review on wearable devices for environmental monitoring and their application for health and wellness. Sensors 2022, 22, 5994.
23. Park, Y. M.; Kwan, M. P. Individual exposure estimates may be erroneous when spatiotemporal variability of air pollution and human mobility are ignored. Health. Place. 2017, 43, 85-94.
24. Yu, X.; Stuart, A. L.; Liu, Y.; et al. On the accuracy and potential of Google Maps location history data to characterize individual mobility for air pollution health studies. Environ. Pollut. 2019, 252, 924-30.
25. Yu, H.; Russell, A.; Mulholland, J.; Huang, Z. Using cell phone location to assess misclassification errors in air pollution exposure estimation. Environ. Pollut. 2018, 233, 261-6.
26. Eastman, E.; Stevens, K. A.; Ivey, C.; Yu, H. On the potential of iPhone significant location data to characterize individual mobility for air pollution health studies. Front. Environ. Sci. Eng. 2022, 16, 1542.
27. Yu, M.; Zhang, S.; Zhang, K.; Yin, J.; Varela, M.; Miao, J. Developing high-resolution PM2.5 exposure models by integrating low-cost sensors, automated machine learning, and big human mobility data. Front. Environ. Sci. 2023, 11, 1223160.
28. Li, Y.; Huang, Y.; Li, R.; Zhang, K. Historical redlining and park use during the COVID-19 pandemic: evidence from big mobility data. J. Expo. Sci. Environ. Epidemiol. 2024, 34, 399-406.
29. Lu, X.; Wang, Y.; Huang, L.; Yang, W.; Shen, Y. Temporal-spatial aggregated urban air quality inference with heterogeneous big data. In: Yang Q, Yu W, Challal Y, editors. Wireless algorithms, systems, and applications. Cham: Springer International Publishing; 2016. pp. 414-26.
30. Giordano, M. R.; Malings, C.; Pandis, S. N.; et al. From low-cost sensors to high-quality data: a summary of challenges and best practices for effectively calibrating low-cost particulate matter mass sensors. J. Aerosol. Sci. 2021, 158, 105833.
31. Jiao, W.; Hagler, G.; Williams, R.; et al. Community Air Sensor Network (CAIRSENSE) project: evaluation of low-cost sensor performance in a suburban environment in the southeastern United States. Atmos. Meas. Tech. 2016, 9, 5281-92.
32. Morawska, L.; Thai, P. K.; Liu, X.; et al. Applications of low-cost sensing technologies for air quality monitoring and exposure assessment: How far have they gone? Environ. Int. 2018, 116, 286-99.
33. Feinberg, S. N.; Williams, R.; Hagler, G.; et al. Examining spatiotemporal variability of urban particulate matter and application of high-time resolution data from a network of low-cost air pollution sensors. Atmos. Environ. 2019, 213, 579-84.
34. Lim, C. C.; Kim, H.; Vilcassim, M. J. R.; et al. Mapping urban air quality using mobile sampling with low-cost sensors and machine learning in Seoul, South Korea. Environ. Int. 2019, 131, 105022.
35. Masiol, M.; Squizzato, S.; Chalupa, D.; Rich, D. Q.; Hopke, P. K. Spatial-temporal variations of summertime ozone concentrations across a metropolitan area using a network of low-cost monitors to develop 24 hourly land-use regression models. Sci. Total. Environ. 2019, 654, 1167-78.
36. Weissert, L.; Alberti, K.; Miles, E.; et al. Low-cost sensor networks and land-use regression: interpolating nitrogen dioxide concentration at high temporal and spatial resolution in Southern California. Atmos. Environ. 2020, 223, 117287.
37. Fujita, E. M.; Campbell, D. E.; Arnott, W. P.; Johnson, T.; Ollison, W. Concentrations of mobile source air pollutants in urban microenvironments. J. Air. Waste. Manag. Assoc. 2014, 64, 743-58.
38. Levy, I.; Mihele, C.; Lu, G.; Narayan, J.; Hilker, N.; Brook, J. R. Elucidating multipollutant exposure across a complex metropolitan area by systematic deployment of a mobile laboratory. Atmos. Chem. Phys. 2014, 14, 7173-93.
39. Xu, J.; Yang, W.; Bai, Z.; et al. Modeling spatial variation of gaseous air pollutants and particulate matters in a Metropolitan area using mobile monitoring data. Environ. Res. 2022, 210, 112858.
40. Blanco, M. N.; Bi, J.; Austin, E.; Larson, T. V.; Marshall, J. D.; Sheppard, L. Impact of mobile monitoring network design on air pollution exposure assessment models. Environ. Sci. Technol. 2023, 57, 440-50.
41. Clark, S. N.; Kulka, R.; Buteau, S.; et al. High-resolution spatial and spatiotemporal modelling of air pollution using fixed site and mobile monitoring in a Canadian city. Environ. Pollut. 2024, 356, 124353.
42. Watson, G. L. Machine learning in environmental exposure assessment. In: Wang Q, Cai C, editors. Machine learning in chemical safety and health. Wiley; 2022. pp. 251-65.
43. Di, Q.; Kloog, I.; Koutrakis, P.; Lyapustin, A.; Wang, Y.; Schwartz, J. Assessing PM2.5 exposures with high spatiotemporal resolution across the continental United States. Environ. Sci. Technol. 2016, 50, 4712-21.
Cite This Article
Open Access
, ... Kai ZhangHow to Cite
Han, B.; Xu, J.; Zhang, K. Advancing air pollution exposure assessment model: challenges and future perspectives. J. Environ. Expo. Assess. 2025, 4, 6. http://dx.doi.org/10.20517/jeea.2024.56
Download Citation
Export Citation File:
Type of Import
Tips on Downloading Citation
Citation Manager File Format
Type of Import
Direct Import: When the Direct Import option is selected (the default state), a dialogue box will give you the option to Save or Open the downloaded citation data. Choosing Open will either launch your citation manager or give you a choice of applications with which to use the metadata. The Save option saves the file locally for later use.
Indirect Import: When the Indirect Import option is selected, the metadata is displayed and may be copied and pasted as needed.
About This Article
Special Issue
Copyright
Data & Comments
Data
0
Comments
Comments must be written in English. Spam, offensive content, impersonation, and private information will not be permitted. If any comment is reported and identified as inappropriate content by OAE staff, the comment will be removed without notice. If you have any queries or need any help, please contact us at [email protected].