This session will showcase best practices in using LCS for AQ monitoring. We will explore case studies where LCS enhance air quality monitoring products (integrating satellite with air sensor information), verify air quality forecasting systems, and support public health and community monitoring initiatives, particularly, but not limited, to resource-limited settings. This session promotes cutting-edge research, case studies, and best practices for using air sensors in air quality monitoring. These aspects will help to enhance air quality monitoring capabilities and foster strong local and international collaboration. It will also explore strategies for sustainable practices that empower communities and ensure equitable partnerships.
Orals: Thu, 7 May, 14:00–17:45 | Room 1.61/62
Air pollution concentration data collected by low-cost sensors and other types of monitors requires a Data Management System (DMS) for data collection and storage, quality assurance, and analysis. Currently, organizations that measure air pollution must either purchase a proprietary system or develop a custom DMS, which can lead to duplication of effort, runaway costs, systems that are not built-for-purpose, and constrained data sharing. Organizations with fewer financial resources and/or less technical capacity are particularly challenged. The Community of Practice for Air Quality Systems (COMPASS) Open Data Management System project is creating a collaboratively developed, open-source DMS to address these challenges.
The COMPASS project convenes core stakeholders representing different geographies and constituencies to develop an open-source DMS that meets the needs of organizations that collect air data or aim to do so. Core functions related to data ingestion, aggregation, harmonisation, storage, quality control, validation, and data sharing have been identified through extensive global stakeholder engagement, and a prototype DMS is now being piloted.
This presentation will describe the collaborative process underway to develop and launch, by the end of 2026, a sustainable open-source DMS for practitioners and managers who need to efficiently manage the air pollution measurements they collect.
How to cite: Hagerbaumer, C., Alimov, K., Bathan-Baterina, G., Biggs, R., Bontia, C., Carbone, S., DeWinter, J., Diez, S., Elkotbeid, S., Espita-Casanova, D., Friedman, E., Mamadov, I., Martinez, B., Opondo, G., Pavlovic, N., Tamayo, E. G., and Tejada, M.: The Community of Practice for Air Quality Systems (COMPASS) Open-Source Data Management System, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3600, https://doi.org/10.5194/egusphere-egu26-3600, 2026.
Air pollution remains a leading public health concern, disproportionately affecting densely populated urban areas in low and middle-income countries such as Nepal. Limited regulatory monitoring infrastructure and data scarcity constrains the assessment of population level exposure and vulnerability. This study employed portable aerosol sensors (Aerocet 831-Met One Instrument Inc., Grant Pass, Oregon, USA), to measure real-time personal exposure to PM1, PM2.5, PM4, and PM10 across indoor, outdoor, and mixed microenvironments in the Kathmandu valley.
The study area was sampled into urban zones with relatively higher PM concentrations and sub-urban zones with lower PM concentrations using satellite-derived PM2.5 data, enabling stratified analysis of personal exposure across spatially varying pollution levels. Households were selected using a snowball sampling method from both exposure zones, to include diverse socio-economic, occupational, and educational backgrounds, encompassing both home and workplace microenvironments. Two participants, usually one female and one male, were recruited from each selected household to understand the difference in exposure patterns based on gender. The participants carried a backpack with sensors for a continuous 72-hour period to monitor personal exposure. The data was collected between January and April 2025, capturing the winter season, which is characterised by elevated air pollution levels in the study area.
33 households participated in the study, comprising a total of 66 participants. Aerosol sensor measurements were integrated with participant-reported daily activity logs to characterise personal exposure patterns. Furthermore, combining sensor data with socio-demographic characteristics and microenvironmental information, the study aims to identify populations most vulnerable to air pollution exposure. Preliminary findings suggest variability in exposure across socio-economic groups, microenvironments, and exposure zones. Comprehensive analysis, including data normalisation, will ensure comparability across exposure zones, socio-economic characteristics, and microenvironments, thereby clarifying patterns of vulnerability.
The study approach demonstrates the application of portable sensors to understand exposure in communities with a limited regulatory monitoring network. The results will inform actionable strategies for targeted public health interventions.
How to cite: Maka Shrestha, R., Pope, F., Day, R., Sugden, F., Adhikary, B., and Bousiotis, D.: Assessing Air Pollution Vulnerability in the Kathmandu Valley: Insights from Personal Particulate Matter Exposure Monitoring with Portable Sensors, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13161, https://doi.org/10.5194/egusphere-egu26-13161, 2026.
Recent advances in low-cost air quality sensors have substantially expanded the capacity to monitor air pollution at fine spatial and temporal scales, enabling analyses that are not feasible with sparse regulatory networks. Here, we use a dense network of low-cost PM2.5 sensors to investigate neighborhood-scale variability in winter air pollution and its relationship with urban land-cover patterns in Santiago, Chile, a highly polluted metropolitan area.
We deployed 55 sensors across four residential neighborhoods during winter 2021, generating high-resolution PM2.5 measurements that capture strong intra-neighborhood variability. These data were combined with high-resolution land-cover maps to quantify both compositional (e.g., land-cover proportions) and configurational (e.g., patch size and aggregation) metrics within multiple buffer distances (30–480 m) around each sensor. Scale-dependent relationships were evaluated using linear mixed-effects models across different PM2.5 concentration ranges.
The sensor network consistently detected spatially structured PM2.5 patterns that would not be observable using sparse reference stations. Built-up land cover showed positive association with PM2.5 concentrations, particularly during high-pollution episodes, while seasonal soil and deciduous tree cover were negatively associated with PM2.5 at specific spatial scales. Configurational metrics, especially the size and aggregation of land-cover patches, were also associated with PM2.5, indicating that how land cover is arranged can be as relevant as overall land-cover extent.
Our findings demonstrate that dense low-cost sensor networks can support robust scientific analyses of urban air quality, despite higher measurement uncertainty compared to reference-grade instruments. By enabling fine-scale assessments of land-cover–air pollution interactions, low-cost sensors offer significant opportunities for advancing urban air quality research and informing neighborhood-scale mitigation strategies, particularly in cities with limited monitoring infrastructure.
How to cite: Fernández, I., Diez, S., and González, A.: Neighborhood-scale spatial PM2.5 variability from dense low-cost sensor networks: the role of urban land-cover patterns in Santiago, Chile, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4088, https://doi.org/10.5194/egusphere-egu26-4088, 2026.
Air pollution remains a critical global challenge, disproportionately impacting vulnerable communities in low- and middle-income countries. Weak policies, fragmented institutions, limited financial and computational resources, and lack of comprehensive monitoring infrastructure hinder effective air quality (AQ) management. Dar es Salaam city with population of about 7 million, is among the world’s Megacities that is undergoing rapid urbanization with accompanying infrastructure development but with underdeveloped waste management.
Monitoring of ammonia concentration gas was made at urban and peri-urban sites in Dar es Salaam city to identify its spatial and temporal variability. The monthly mean ammonia concentrations measured at two sub-urban sites (Buza Hospital and Temeke DMDP) were 34.9 and 16.6 ppm, respectively. The monthly mean ammonia concentrations at urban site (DIT, Kigamboni, Makuburi, Sinza Hospital and Mlimani city) were, 8.2, 8.85, 22.02, 16.5, and 7.2, respectively. Further, a comparison of the measured hourly data at urban and peri-urban sites showed its relative dominance at peri-urban sites during the evening hours while during the morning hours the dominance was to the urban sites. Different studies suggest that the trend of ammonia levels should be tightly affected by an increasing number of vehicles (morning traffic jams in urban areas) as well as agricultural and livestock activities (common in peri-urban areas). As such, the results of our statistical analysis point to the potentially significant role of agriculture and livestock activities in the elevation of ammonia levels in peri-urban areas of Dar es Salaam city.
How to cite: manyele, A.: Spatial and Temporal Variability of Ambient Ammonia (NH₃) and other Gas pollutants in Urban and Peri-Urban Dar es Salaam, Tanzania as Measured by Low cost Gas Sensors, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1297, https://doi.org/10.5194/egusphere-egu26-1297, 2026.
Low-cost sensor (LCS) networks can complement sparse regulatory monitoring, but their value depends on robust integration strategies that preserve data quality while exploiting dense spatial sampling. Here we assess the added value of incorporating validated LCS PM2.5 observations into the S-MESH (Satellite and ML-based Estimation of Surface air quality at High resolution) machine learning framework (Shetty et al., 2024, 2025) to generate continental-scale, 1 km resolution surface PM2.5 fields across Central Europe. Two integration strategies are evaluated for 2021–2022 within a stacked XGBoost architecture driven by satellite aerosol optical depth, meteorological predictors, and CAMS regional forecasts: a) using LCS data as an additional training target (LCST), and b) using LCS information as a model input feature (LCSI) via an inverse-distance-weighted spatial convolution layer that encodes local sensor influence. Relative to a baseline trained only on official monitoring stations, LCSI yields consistent performance gains, with RMSE reductions of ~15–20% in urban areas, whereas LCST provides less consistent improvement. The resulting high-resolution mapping product achieves skill comparable to the CAMS regional reanalysis, often considered as a modelling “gold standard” for European air-quality assessment, and in some evaluations surpasses it, with lower annual mean absolute error (2.68 vs 3.32 µg m⁻³) (Shetty et al., 2026). This demonstrates that a data-fusion ML approach including LCS information can deliver reanalysis-level performance at 1 km resolution while requiring only modest computational resources compared with running full chemical transport model reanalyses, enabling rapid updates and scalable deployment. SHAP-based attribution further suggests that LCSI improves the model’s ability to capture localized pollution variability, while performance degrades where sensor density is low, limiting representation of inter-urban transport.
Although demonstrated in Europe, the underlying methodology, namely combining globally available satellite products and meteorology with quality-controlled LCS networks in a computationally efficient ML framework, has potential to strengthen air-quality assessment also in resource-limited settings where regulatory infrastructure is scarce. A requirement for this is that appropriate sensor calibration/validation workflows are in place and equitable partnerships support sustainable sensor deployment and data stewardship.
Shetty, S., Schneider, P., Stebel, K., Hamer, P. D., Kylling, A., and Koren Berntsen, T.: Estimating surface NO2 concentrations over Europe using Sentinel-5P TROPOMI observations and Machine Learning, Remote Sens. Environ., 312, 114321, https://doi.org/10.1016/j.rse.2024.114321, 2024.
Shetty, S., Hamer, P. D., Stebel, K., Kylling, A., Hassani, A., Berntsen, T. K., and Schneider, P.: Daily high-resolution surface PM2.5 estimation over Europe by ML-based downscaling of the CAMS regional forecast, Environ. Res., 264, 120363, https://doi.org/10.1016/j.envres.2024.120363, 2025.
Shetty, S., Hassani, A., Hamer, P. D., Stebel, K., Salamalikis, V., Berntsen, T. K., Castell, N., and Schneider, P.: Evaluating the role of low-cost sensors in machine learning based European PM2.5 monitoring, Environ. Res., 291, 123558, https://doi.org/10.1016/j.envres.2025.123558, 2026.
How to cite: Schneider, P., Shetty, S., Hassani, A., Salamalikis, V., Stebel, K., Hamer, P., Berntsen, T. K., and Castell, N.: Integrating validated large-scale sensor observations into ML-based PM2.5 mapping: lessons from Europe with global relevance, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7366, https://doi.org/10.5194/egusphere-egu26-7366, 2026.
The last several decades have seen steady and expansive growth both in the development and the application of low-cost sensors (LCS) in the field of air pollution research. They are now increasingly prevalent in air quality monitoring thanks to their affordability and adaptability across diverse environments. However, in the wider academic and monitoring communities, their deployment has largely focused on the expansion of spatial coverage and generating larger datasets, either to generate data where there previously was none, or to fill spatiotemporal gaps. In this work, we argue that LCS have high potential for use in the targeted assessment of policy interventions, and that this potential remains largely underexplored. Drawing on a number of recent studies, we demonstrate the value of LCS as an effective tool for evaluating the impacts of policy measures on urban air pollution.
Utilizing these studies as an empirical basis, this work introduces a five-step rubric for guiding targeted policy assessments using LCS. These steps broadly are: 1. Identification, in which partnerships and policies are identified and established; 2. Planning, in which measurement and intervention timelines are aligned and campaigns designed; 3. Calibration, in which LCS are suitably calibrated using in-situ co-locations for the environments they are to be used in; 4. Analysis, in which LCS data is collected and analysed, with potential impacts of the policy intervention quantified; and 5. Dissemination, in which the results are published and presented in a timely manner to relevant stakeholders. These cyclical steps should be considered fluid and dynamic, as they are intended to align with policymaking timelines, which often diverge from research timelines.
In addition, we discuss the strengths and limitations of LCS for use in targeted policy assessment, to clarify key criteria for deployment in this application. These strengths include their capacity for high spatiotemporal resolution, flexible deployment options (especially outdoor), and cost-effectiveness in shorter term campaigns. These attributes enable the detection of hyperlocal pollution patterns and emission events that are often missed by sparsely populated reference monitoring networks. Key limitations to this approach include sensor drift, inter-sensor variability, cross-sensitivities to other pollutants, and the need for rigorous calibration. These factors can constrain the data quality and limit the detection of the impact signal of the policy interventions in question. However, by properly quantifying uncertainties and accounting for e.g., meteorological variability, these limitations can be taken into account and relevant results can still be delivered to stakeholders.
As such, this work argues for a shift in the research landscape surrounding LCS, and advocates for a shift away from indiscriminate large-scale sensor deployment and toward targeted assessment of individual policies at the local scale. We encourage its further uptake and remain optimistic that this approach can transform the evidence base for local policy decision-making, to create a step change in the tools and data used for mitigating air pollution and providing clean, healthy air for all.
How to cite: Schmitz, S. and von Schneidemesser, E.: Harness low-cost sensors for the targetedassessment of policy, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11987, https://doi.org/10.5194/egusphere-egu26-11987, 2026.
Hospitals and public institutions often operate under significant financial constraints. When resources are insufficient to meet core service demands, as has been the case for the United Kingdom National Health Service (NHS), investment in air quality monitoring becomes even more challenging. This is despite clear evidence that healthcare settings are not insulated from the impacts of air pollution. Low-cost air quality sensors, which cost far less than reference grade instruments, offer a practical opportunity to generate meaningful environmental data in these environments.
At two major hospitals in Northeast England, the Royal Victoria Infirmary (RVI) and Freeman Hospital (FH) of Newcastle upon Tyne Hospitals NHS Foundation Trust, the deployment of 36 passive diffusion tubes, 9 indoor PurpleAir particulate monitors, and 3 ambient multisensor devices has enabled continuous monitoring of nitrogen dioxide (NO2), and particulate matter with aerodynamic diameters <10µm (PM10) and <2.5µm (PM2.5) over a three-year period. These data provide valuable temporal and spatial insights across two large modern hospital campuses.
Results show that although many hospital pollution hotspots reflect wider urban traffic patterns, vehicle movements within the hospital premises also contribute substantially. A change in parking policy resulted in marked reductions in ambient pollution concentrations. At one section of the RVI, the mean difference in NO2 concentration before and after the policy change showed a statistically significant reduction of 2.02µg/m3 with a two-sided p value of 0.007 based on a paired sample t test. This confirms that nitrogen dioxide levels decreased following the intervention. The t value of 2.722 indicates a moderate effect size.
The cleanest locations across both hospitals were consistently a staff and patient green space and a strictly managed no idling car park. Spatial analysis of diffusion tube data shows annual mean nitrogen dioxide concentrations of 19.4µg/m3 ±5.05 at FH, and 23.6µg/m3 ±5.33 at RVI in 2023. In 2024 both hospitals recorded reductions in annual mean concentrations of 4.1µg/m3 at FH and 4.4µg/m3 at RVI. Increased standard deviation in the same year highlights substantial site level variability. All ambient monitors demonstrated high variation in pollution levels within each hospital with annual mean values exceeding the World Health Organization guideline. This underscores the need for targeted interventions even within compact hospital settings of approximately 0.14 square kilometres.
Indoor PM2.5 concentrations also showed frequent exceedances of the World Health Organization one hour guideline of 15µg/m3 and the United Kingdom 24 hour guideline of 25µg/m3. For example, the New Victoria Wing reception at RVI recorded 281 hourly exceedances in 2024. In contrast, PM10 exceedances were rare and remained below the legal limit of 35 events per year.
Both hospitals have committed to achieving Excellent status within the Clean Air Hospital Framework as part of efforts to reduce emissions and protect patients, staff and the surrounding community. The Framework provides a set of actions across Travel, Procurement, Construction, Energy, Local Air Quality, Outreach and Leadership. Low-cost sensors support implementation by enabling hotspot identification, tracking the effectiveness of interventions and providing high resolution pollution insights at an accessible cost.
How to cite: Okeowo, B.: Monitoring and Mitigating Air Pollution in Healthcare: Characterising PM2.5 and NO2 Variability to Inform Sustainability Actions at Two UK Hospitals, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2081, https://doi.org/10.5194/egusphere-egu26-2081, 2026.
Low-cost optical particle counters (OPCs) are increasingly used to supplement regulatory air quality measurements, yet their performance remains strongly dependent on aerosol characteristics and environmental conditions. In this study, five OPC-N3 sensors were co-located with a research-grade GRIMM 11-D aerosol spectrometer and evaluated using hourly measurements from an urban polluted site in western India (Ahmedabad) during September 2025. Mean ± standard deviation PM2.5 (PM10) concentrations during the colocation period were 25 ± 10 (82 ± 37) µg/m3, whereas temperature and relative humidity (RH) were 30 ± 3 °C and 74 ± 14 %.
OPC-N3 sensors show moderate to strong correlations with the GRIMM for PM1, PM2.5, and PM10 (r ≈ 0.7-0.9), demonstrating good capture of temporal variability, but consistently underestimate PM mass (PM2.5: normalized RMSE ≈ 23-37%). The regression slopes increase from PM1 (0.3-0.5) to PM2.5 (0.5-0.7) and remain similar for PM10, indicating relatively better performance for coarser fractions. To diagnose the drivers of OPC response, the influence of meteorological parameters and the reference PM2.5/PM10 ratio was quantified using their correlations with normalized OPC measurements (OPC/GRIMM PM ratio). The normalized OPC PM shows moderate positive correlations with RH (r ≈ 0.4-0.5) and moderate negative correlations with temperature (r ≈ -0.3 to -0.5), highlighting the important role of meteorology and hygroscopic growth in governing OPC response. The normalized OPC PM showed near-zero correlations with the reference PM2.5/PM10 indicating that the underlying aerosol size distribution is probably not playing a big role in OPC performance in this environment. OPC-derived aerosol size distribution captures broad features similar to the reference, with minimal diurnal variability.
Overall, the results demonstrate that OPC-N3 sensors are suitable for capturing relative variability and trends but require environment- and size-fraction-specific calibration. In particular, OPC performance is primarily governed by meteorological conditions in polluted urban environments, underscoring the need to explicitly account for meteorology for reliable PM mass estimation across diverse environments.
How to cite: Dahima, Y. and Vaishya, A.: Evaluation of OPC-N3 in an Urban Environment in Western India, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7710, https://doi.org/10.5194/egusphere-egu26-7710, 2026.
Traffic is a dominant source of urban air pollution in many low- and middle-income countries, where ageing vehicle fleets, high traffic volumes, and the resuspension of dust from paved and unpaved roads combine to degrade air quality and threaten public health. Yet the relative contributions of exhaust and non-exhaust traffic-related emissions to the total air pollution remain poorly quantified.
In partnership with the Global Alliance on Health and Pollution (GAHP), we deployed a network of low-cost air quality sensors across Kampala, Uganda, measuring particulate matter mass and size resolved number concentrations, NOx, and total volatile organic compounds (TVOCs), to complement the existing AirQo monitoring network (Sserunjogi et al., 2022). The objective was to characterise spatial variability in air pollution across the city centre and suburban areas and to quantify the contribution of transport-related sources.
Low-Cost Source Apportionment (LoCoSA) methods (Bousiotis et al., 2025) were applied to the sensor data to identify the dominant contributors to PM2.5 at multiple sites. Depending on proximity to major roads, direct traffic emissions accounted for 18–35% of total PM2.5. Resuspended dust, strongly influenced by vehicle activity, was the largest single source, contributing more than 50% at all locations. These results indicate that a substantial fraction of PM2.5 in Kampala is either directly or indirectly linked to traffic, amplifying the overall impact of transport on urban air quality.
The source-apportionment results are being integrated into a simplified version of the University of Birmingham’s air-quality life-course assessment tool (AQ-LAT; Hall et al., 2024) to quantify source-specific health impacts and attributable mortality. This low-cost, scalable framework enables cities in resource-limited settings to estimate the public-health benefits of targeted emission-control strategies, supporting evidence-based and cost-effective air-quality management.
This presentation demonstrates how the combination of low-cost sensing, low-cost source apportionment, and health-impact assessment can be used to quantify the contribution of traffic to air pollution and associated health burdens. The approach is scalable and transferable to cities worldwide.
References
Sserunjogi et al., (2022). Seeing the air in detail: Hyperlocal air quality dataset collected from spatially distributed AirQo network. Data in brief, 44, p.108512. https://doi.org/10.1016/j.dib.2022.108512
Bousiotis et al., (2025). Low-Cost Source Apportionment (LoCoSA) of air pollution-literature review of the state of the art. Science of The Total Environment, 998, p.180257. https://doi.org/10.1016/j.scitotenv.2025.180257
Hall et al., (2024). Regional impact assessment of air quality improvement: The air quality lifecourse assessment tool (AQ-LAT) for the West Midlands combined authority (WMCA) area. Environmental Pollution. https://doi.org/10.1016/j.envpol.2024.123871
How to cite: Pope, F., Bousiotis, D., Singh, A., Okure, D., Okello, G., Sanghera, D., Bartington, S., Hall, J., Okedi, D., Sserunjogi, R., and Bainomugisha, E.: Assessing the effect of traffic on air quality and public health in Kampala, Uganda using low-cost approaches, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14688, https://doi.org/10.5194/egusphere-egu26-14688, 2026.
Low-cost aerosol sensors give the opportunity to build air quality monitoring network, allowing for increased spatial resolution of particulate matter (PM) measurements [1]. This type of network has already been used in urban environments to capture variation of PM at the city scale [2]. Another environment where PM can change quickly with space is alpine valleys. Indeed, the vertical stratification of air caused by nighttime ground cooling induces important PM concentration variation along a slope. While it is well known that aerosol concentration is greater at the valley floor, the aim of this study is to describe how PM concentration evolves with altitude at the ground and what meteorological phenomenon can change the vertical distribution of PM. Indeed, complex topography influences vertical transport of aerosols through upward and downward motions as well as through local to regional sources of PM.
In the development of a monitoring station for mountainous regions, it is necessary to incorporate specific constraints, such as challenging access to measurement sites, a lack of phone network, and limited access to power. Optical particle counters are low-cost PM sensors that are both small and energy-efficient, enabling the development of portable and autonomous stations. These are key features for building a monitoring network in an alpine valley. We designed monitoring stations based on these constraints. The stations have several sensors that measure the temperature, humidity, and PM10, PM2.5, and PM1 levels. Every fifteen minutes, a microcontroller gathers data and stores them on an SD card. The energy is provided by a battery recharged by a solar panel recessed at the top of the 3D-printed framework. The overall size of the stations is a cube with a 30 cm side length.
These laboratory-made stations have been used to study PM concentration in the Arve River valley, located in the northern French Alps and encompassing cities like Sallanches, Passy, and Chamonix. The network is composed of 12 monitoring stations and 6 monitored sites positioned on the slopes around the Passy basin. One monitoring site is located on the valley floor, on an air quality agency monitoring station containing a TEOM-FDMS. The discussion will focus on the comparison of low-cost stations with the reference measurements, the benefits of using a low-cost monitoring network to study PM concentrations in mountainous terrain, and the limitations inherent to low-cost sensors and autonomous stations.
[1] Bagkis, E., Hassani, A., Schneider, P., DeSouza, P., Shetty, S., Kassandros, T., Salamalikis, V., Castell, N., Karatzas, K., Ahlawat, A., Khan, J. Evolving trends in application of low-cost air quality sensor networks: challenges and future directions. npj Clim Atmos Sci 8, 335 (2025). https://doi.org/10.1038/s41612-025-01216-4
[2] Feinberg, S. N., Williams, R., Hagler, G., Low, J., Smith, L., Brown, R., Garver, D., Davis, M., Morton, M., Schaefer, J. & Campbell, J. Examining spatiotemporal variability of urban particulate matter and application of high-time resolution data from a network of low-cost air pollution sensors. Atmospheric Environment 213, 579–584 (2019). https://doi.org/10.1016/j.atmosenv.2019.06.026
How to cite: Couëtoux, P., Fanget, P., Piot, C., and Staquet, C.: Building a low-cost monitoring network to track vertical transport of particulate matter along valley slope: the benefits of low-tech in mountainous environment, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20134, https://doi.org/10.5194/egusphere-egu26-20134, 2026.
Industrial air pollutants pose safety and health risks to workers. This study aims to better understand the spatial and temporal distribution of air pollutants in industrial working environments. We developed and installed a novel low-cost sensor network in multiple workplaces: indoors at a steel factory, outdoors at a ferrochromium mill and on a cruise ship’s car deck. The stationary sensor networks are composed of multiple custom sensing nodes. Each node is equipped with low-cost sensors to assess gaseous components, particulate matter, temperature, and humidity. Air quality measurements using validated traditional occupational hygiene methods and high-end portable direct-reading instruments are performed stationary and mobile. Mobile devices carried by workers and unmanned aerial/ground vehicles complement the measurements by the stationary sensor network. The results consist of evaluations carried out by combining the sensor data, contextual information, and the results obtained with traditional exposure assessment methods. Can this data fusion be used to assess exposure and target risk management measures? Can real time sensor measurements support the worker safety? The short answer is YES!, but further steps are necessary to improve the sensor data reliability and applicability for detailed occupational exposure assessment. We present pollutant concentration maps and time series analysis, which are valuable for planning control measures and developing worker guidance to improve industrial safety. We further discuss the advantages and disadvantages of the available sensors for industrial air quality measurements and present the next steps of research needed for wider application of these safety technologies.
This research project, Robot-assisted Environmental Monitoring for Air Quality Assessment in Industrial Scenarios (RASEM), has received funding from the Finnish Work Environment Fund, Finnish Institute of Occupational Health, and Bundesanstalt für Materialforschung und –prüfung (BAM) under Saf€ra 2018 joint call: new technologies, new trends and monitoring safety performance.
How to cite: Poikkimäki, M., Kangas, A., Winkler, N. P., Neumann, P. P., Leikas, M., and Säämänen, A.: Can low-cost sensor networks help industrial air quality?, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-42, https://doi.org/10.5194/egusphere-egu26-42, 2026.
Effective air quality management depends on the availability of reliable, locally resolved pollution data. Low-cost sensors (LCS) offer high spatial and temporal resolution but are often constrained by data reliability challenges, primarily due to their reliance on field calibration. Conventional calibration methods typically require colocation with reference stations, a limitation in regions with sparse monitoring infrastructure. This study presents a model that integrates emission inventory (EI) data with machine learning (ML) to achieve source apportionment (SA) using LCS, demonstrated through a case study in Fianarantsoa, Madagascar. Unlike conventional ML calibration approaches benchmarked solely against collocated reference monitors, the proposed method exploits a distributed sensor network in which each device is cross-validated by at least two neighbouring sensors within 500 m. Conventional calibrations frequently suffer from sensor- and site-specific biases, provide limited source-specific information, and are often hindered by proprietary algorithms. To address these issues, a Data Reliability Indicator (DRI) is introduced to evaluate LCS performance across high-, middle-, and low-income country contexts. The findings demonstrate that LCS, when supported by emission inventories and network-based cross-validation, can deliver reliable source apportionment and high-resolution air quality insights, even in regions with minimal reference-grade monitoring.
How to cite: Sharma, R., Razakamanantsoa, A., Rayssac, E., and Julian, A.: Regression Based Apportionment with Low-Cost Air Quality Sensors Using Machine Learning and Emission Inventories, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2252, https://doi.org/10.5194/egusphere-egu26-2252, 2026.
Nitrogen dioxide (NO2) is a major urban air pollutant, but current low-cost sensors for NO2 based on electrochemical cells have important drawbacks, including modest accuracy and susceptibility to temperature, humidity, and chemical interferences. These sensors are also too slow for mobile monitoring and for measuring the large and rapid fluctuations of NO2 in the transport microenvironment. These are important monitoring approaches and settings for NO2 because vehicles are the dominant source of NO2 in cities. Here we present our work in adapting cavity-enhanced absorption spectroscopy (CEAS) to develop fast (< 5 s), sensitive (±1 ppb), and portable sensors for NO2 at lower cost.
We characterise sensor performance in laboratory intercomparisons, and present adaptations to different platforms (vehicles, bicycles, and backpacks). Case studies are presented of mobile and stationary monitoring of transport emissions in Cork city and in smaller towns in Ireland. These measurements show the disproportionate impact of a small number of highly polluting vehicles. A perspective on the challenges and prospects for this approach is discussed.
How to cite: Venables, D., Dorney, C., Edmonds, A., Vikas, R., and Wang, M.: Towards lower-cost spectroscopic sensors: Applications in mobile monitoring and roadside measurements of NO2 , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8204, https://doi.org/10.5194/egusphere-egu26-8204, 2026.
The growing adoption of low-cost sensors (LCSs) has significantly enhanced environmental monitoring by enabling widespread, community-driven data collection, particularly in regions requiring dense monitoring, and in regions with limited or no reference instrumentation. Increased public awareness and demand for dense environmental monitoring have resulted in extensive air quality and meteorological datasets from diverse sources. However, the integration of such datasets into regulatory frameworks and large-scale environmental monitoring remains challenging due to persistent issues related to data quality, standardization, and interoperability.
To address these challenges, the FILTER (Framework for Improving Low-cost Technology Effectiveness and Reliability) approach developed by Hassani et al. (2025) provides a suite of algorithms to harmonize, quality-check, flag, and perform in-situ corrections on crowdsourced PM2.5 LCS datasets. While FILTER was initially designed and validated for static PM2.5 sensors, it has since been extended to address data quality challenges associated with the dynamics of mobile and wearable sensing.
Across both static and mobile LCS platforms, FILTER employs a unified processing pipeline that generates measurement-level quality flags based on multiple statistical tests, to quantify the reliability of each observation. Quality control (QC) includes statistical tests to: (a) assess physical measurement consistency (range validity test), (b) detect flatline behavior (constant value test), and (c) identify abnormal patterns (spatiotemporal outlier detection test) using both historical trends and spatial comparisons with neighboring LCSs. Beyond these mandatory QC steps, more advanced statistical procedures incorporate relative (spatial correlation test) and absolute (spatial similarity test) comparisons with nearby LCSs, higher-quality instruments, and reference monitoring stations. For mobile and wearable sensing, FILTER has been specifically adapted to support pairwise comparisons between mobile sensors and comparisons with higher-accuracy nodes, accounting for operation under dynamic environmental and operational conditions. In this context, statistical comparisons are evaluated during rendezvous events, that is, periods in which the mobile sensor and a higher-accuracy node provide temporally coincident measurements. The modified framework retains the core principles of transparency, scalability, and sensor independence, while introducing additional steps to address motion-related artifacts, intermittent time series, and location-specific uncertainties.
FILTER is developed in the open-source R environment. Its modular and hierarchical design allows flexible adaptation of quality control and correction workflows based on data availability, the spatiotemporal characteristics of LCS networks, and application-specific requirements. By improving data reliability and usability, FILTER enables crowdsourced LCS datasets to serve as a reliable complement to official monitoring networks for air quality management, urban- and regional-scale modeling, and policymaking.
References
Hassani, A., Salamalikis, V., Schneider, P., Stebel, K., and Castell, N.: A scalable framework for harmonizing, standardization, and correcting crowd-sourced low-cost sensor PM2. 5 data across Europe, J. Environ. Manage., 380, 125100, 2025.
How to cite: Salamalikis, V., Hassani, A., Schneider, P., and Castell, N.: Improving data reliability in air quality monitoring from static and mobile sensor platforms and networks using the FILTER framework, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9353, https://doi.org/10.5194/egusphere-egu26-9353, 2026.
Small sensor systems developed over the last decade have demonstrated great potential for air quality and trace pollutant monitoring. These systems are widely available on the market and can provide a fast and lower-cost alternative that is complementary to reference methods. There could, however, be a number of issues in the quality of the data from such sensor systems, as they are prone to interferences, temporal drift, low accuracy and lack of metrological traceability. Therefore, adherence to Technical Specification (TS) 17660-1 (Air quality – Performance evaluation of air quality sensor systems – Part 1: Gaseous pollutants in ambient air), issued by the European Committee for Standardization (CEN), is essential for ensuring data quality through a structured metrological assessment. The TS is designed for gaseous pollutants and entails laboratory and field tests that are complex and require high-accuracy, dedicated facilities. The laboratory validation includes preliminary tests to evaluate response time, lack of fit of the regression function and repeatability followed by a series of extended tests designed to assess long-term drift, cross-sensitivities by interfering gases, temperature effects, humidity effects and memory effects of main gas, temperature and humidity.
In this work, we outline the performance of the Multiple Atmosphere Controlled Environment (MACE) facility under several experimental regimes demonstrating how this state-of-the-art facility can be used to perform the laboratory tests described in TS 17660-1. The MACE is an advanced testing facility developed at the National Physical Laboratory (UK) designed to evaluate the performance of gas sensors and assist in the development of new products that meet the requirements of air quality legislation. The facility can create reproducible and stable environmental conditions under a range of temperature, relative humidity, and gas concentrations and compositions. The MACE consists of an environmental chamber that houses six stainless steel exposure pods and includes an insulated environment for delivering temperature-controlled tests. Accurate single and multiple test atmospheres are generated by blending zero air with traceable gas mixtures via an array of calibrated mass flow controllers. The desired relative humidities are generated by a dedicated vaporizer unit. The exposure chambers are connected to an array of reference instruments capable of measuring priority pollutants, including NO, NO2, SO2, CO, CO2 and O3. Here we present the results of tests on a set of sensor systems performed using the MACE, demonstrating that the facility can perform all the tests outlined in the TS. This makes the MACE one of only a few facilities worldwide that meet the requirements specified in the TS.
This work represents the first step toward standardizing small sensor systems, with subsequent stages involving full validation of the TS (including field tests) and its adoption as a Standard, which are currently underway. This work, along with the implementation of quality assurance and quality control practices, will ensure that the data from small sensor systems are traceable and of the highest quality possible.
How to cite: Mendes Emygdio, A. P., Ferracci, V., Garcia, G., and Martin, N.: Advanced testing facility for gas sensors validation under controlled conditions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11683, https://doi.org/10.5194/egusphere-egu26-11683, 2026.
Low-cost sensors (LCS) are increasingly deployed to enhance the spatial resolution of air quality monitoring networks, particularly in communities that are sparsely covered by regulatory-grade instruments. Despite the potential of LCS to provide new insights into local-level pollutant concentrations, issues with accuracy, calibration drift, and maintenance requirements pose significant challenges for ensuring that collected data reliably support scientific analysis and decision-making. Data fusion techniques that integrate LCS data with other observations, such as regulatory-grade ground measurements and remotely sensed satellite data, show promise for improving the accuracy and resolution of air quality predictions. However, standardized approaches for data fusion are not yet available, limiting the use of multimodal data for decision-making.
Here, we present a community-scale air quality monitoring framework that integrates data from LCS networks, regulatory-grade monitoring stations, and satellite-derived observations through a stochastic advection–diffusion (SAD) modeling approach for fine particulate matter (PM2.5). The proposed framework leverages multimodal spatiotemporal data to generate high-resolution PM2.5 fields while explicitly accounting for uncertainty arising from sparse observations, sensor heterogeneity, and measurement error within a probabilistic state-space formulation. We apply the framework for a statewide case study in Texas, USA, using archived LCS observations from the PurpleAir network, regulatory-grade PM2.5 measurements from the U.S. EPA Air Quality System (AQS), and satellite-derived PM2.5 products from NASA’s Tropospheric Emissions: Monitoring of Pollution (TEMPO) mission (early-release PM2.5 products). We examine how different combinations of data sources contribute to predictive performance, providing insight into the relative value of low-cost, regulatory, and satellite observations within the integrated framework. The results demonstrate the value of combining LCS and satellite data within a physics-informed probabilistic framework to support community-scale air quality assessment, sensor network design, and adaptive environmental decision-making.
How to cite: Hummel, M. and Choi, B.: Informing Regional-Scale Air Quality Monitoring through Multimodal Data Integration and Probabilistic Spatiotemporal Modeling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7477, https://doi.org/10.5194/egusphere-egu26-7477, 2026.
Indoor air quality is strongly influenced by both ventilation dynamics and occupant activity, yet the relationship between these factors remains insufficiently characterized. In this study, we deployed a network of low-cost MegaSense sensors across a university campus to continuously monitor several key environmental parameters, including gaseous compounds (CO, NO2 and O3), particulate matters (PM2.5 and PM10), total volatile organic compounds (TVOCs), and noise levels. People counters were also installed side-by-side for an academic year. Initial results revealed pollutants were accumulated overnight due to reduced ventilation, and rapidly diluted in the early morning once ventilation resumed. We also found that whilst PM and gaseous pollutants were strongly linked with ventilation cycles, TVOC concentrations and noise levels exhibited pronounced diurnal patterns closely aligned with occupant movement. The correlation between TVOC concentrations and with people flow (r ≈ 0.7) is strong, likely attributed to emissions from breath, skin, and personal care products, as well as redistribution of localized VOCs through airflow disturbances. Noise levels showed an even stronger correlation (r ≈ 0.8), which indicated human presence through speech, footsteps, and mechanical interactions. O3 concentrations, in contrast, displayed no discernible diurnal variation.
These findings highlight the potential of integrating occupant movement and noise monitoring as effective proxies for estimating TVOC dynamics in indoor environments. More broadly, the study demonstrates opportunities of using low-cost sensor networks in capturing the complex relationships between human activity and indoor air quality, which is able to offer valuable insights for sustainable building management and exposure assessment for human's health. As future work, such sensor applications can be scaled to diverse indoor settings (e.g. occupational, residential, etc) to further explore these relationships.
How to cite: Fung, P. L., Rebeiro-Hargrave, A., and Varjonen, S.: Linking Occupant Movement to Indoor Air Quality Dynamics: Insights from Sensor-Based Monitoring, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15685, https://doi.org/10.5194/egusphere-egu26-15685, 2026.
Low-cost particulate matter (PM) sensors can complement sparse regulatory stations and capture fine-scale urban variability, but their raw readings are affected by environmental sensitivity, drift, and inter-sensor variability. This work examines how reference instrument–sensor calibration and multi-hop rendezvous calibration jointly affect a network of 16 co-located optical PM sensors (eight Plantower PMS5003 and eight Sensirion SPS30) installed within 1 m of a Palas Fidas 200S reference analyser. Starting from experimental field co-location time series, we replay the measured sensor and reference data inside a controlled simulation framework to compare multiple calibration strategies and to identify encounter parameters that optimize the accuracy of a low-cost PM sensor network; the same 16 sensors are then virtually distributed over conceptual areas of 5, 10, and 15 km² to emulate different deployment densities. Sensor–sensor encounters and passages near the reference station are modelled as Poisson processes with a 2-minute time step; per-step interaction probabilities are derived from mean inter-event times and scaled by an effective density term so that larger areas (lower density) yield fewer interactions. Multi-hop rendezvous calibration is controlled by a multi-hop depth H (maximum number of successive updates) and a cumulative calibration influence Γ (the accumulated effect of multi-hop rendezvous calibration corrections applied to a sensor). Five calibration scenarios are compared: raw (no correction), linear regression, linear regression with robust Huber weighting, quadratic regression, and quadratic regression with Huber weighting. Reference instrument–sensor calibration uses a sliding buffer of recent sensor–reference instrument pairs with outlier filtering and time-weighted fitting, while multi-hop rendezvous calibration encounters provide additional limited corrections through a lightweight Kalman-filter update of calibration coefficients. This update provides a principled way to incorporate uncertain peer information while keeping corrections stable and limited through the filter’s weighting of prediction versus noisy observations. Performance is evaluated against the reference instrument using calibrated root-mean-square error (RMSE) for particulate matter of different sizes: PM₁ (≤1 µm), PM₂.₅ (≤2.5 µm), and PM₁₀ (≤10 µm). Compared to uncalibrated measurements, the best-performing configurations reduce the RMSE from 2 to 1.1 µg·m⁻³ for PM₁ (−45%) and from 2.7 to 1.5 µg·m⁻³ for PM₂.₅ (−45%) with robust quadratic calibration, while PM₁₀ is best handled by a quadratic (second-order) model, improving from 5.4 to 3.3 µg·m⁻³ (−39%). Across all scenarios, robust quadratic calibration provides the strongest and most consistent gains for fine particles, whereas the non-robust quadratic model is the most effective choice for coarse particles; moderate multi-hop depth H and cumulative calibration influence Γ further improve RMSE, while high H and Gamma can propagate local biases and increase variability.
How to cite: Pellegrino, J., Bulot, F., Aziza, H., Guerin, M., and Taranto, P.: Impact of multi-hop rendezvous calibration parameters on the accuracy of a network of low-cost PM sensors, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17540, https://doi.org/10.5194/egusphere-egu26-17540, 2026.
Indoor air quality is a dominant risk factor for human health, with many people spending approximately 80-90% of their time indoors. While outdoor infiltration, cooking, heating and cleaning sources are well known, human presence and activity through movement are harder to quantify and thus scarcely considered. This research investigates the relationship between occupancy, physical activity - defined by Kinetic Energy (KE), and particulate matter (PM) in real-world environments, including homes and offices. Our methodology uses a sensing network that combines low-cost air quality sensors with high-resolution radar-based motion sensors. Through this approach, we apply both simple linear regression and source apportionment modelling to define and isolate the contribution of KE-induced resuspension from indoor sources, thereby quantifying the contribution of human activity on indoor air quality levels. The first results, published recently (Bousiotis et al. 2026), establishes a significant correlation between KE thresholds and coarse particle mass (up to r = 0.74), suggesting that human-induced PM10 is a significant, yet under-quantified, contributor to personal exposure. In our current and upcoming work, we provide an update on the analysis for multiple residential and office environments, and go further by analysing the contribution of human movement to PM2.5 levels. By considering the ‘person-as-a-source’ dynamic, this research provides a scalable framework for improving indoor air quality management through low-cost, high-resolution environmental sensing, whilst contributing to the evidence base for healthier building design.
Bousiotis D., D.S. Sanghera, J. Carrington, G. Hodgkiss, F. Jajarmi, K. Rajab and F.D. Pope (2026) Parameterising the effect of human occupancy and kinetic energy on indoor air pollution. npj Climate and Atmospheric Science. https://www.nature.com/articles/s41612-025-01281-9
How to cite: Sanghera, D., Bousiotis, D., Sakenov, M., Rajab, K., and Pope, F.: Low-cost sensors to quantify activity-driven air pollution in indoor environments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7903, https://doi.org/10.5194/egusphere-egu26-7903, 2026.
To address the scarcity of high-time-resolution data on indoor pollutants, this study used low-cost air quality sensors for continuous monitoring in four typical indoor environments: an office, a residential heating room, a street kitchen, and a residential kitchen. Key pollutants (PM2.5, PM10, O3, NOx, CO, CO2, TVOC) and environmental factors (temperature, humidity, noise) were measured. Integrating wavelet analysis, peak/background decomposition, and the positive matrix factorization (PMF) model, three sources were identified: PM-related (particle intrusion/resuspension/oil aerosolization), noise-related (activity-ventilation coupling), and other-gases (combustion and material/surface processes). Wavelet analysis revealed obvious diurnal/semidiurnal cycles and multi-scale periodic characteristics dominated by human activities and ventilation. All environments exhibited distinct pollutant concentration variations linked to their specific functional uses and emission sources. The street kitchen had the highest PM and TVOC levels, while the residential heating room showed the highest CO and CO2. Health risk assessment revealed distinct drivers: office risks from particles and NO2, heating rooms from CO, street kitchens from cooking particles and near-road combustion, and residential kitchens from balanced particle and CO risks. The study confirms low-cost sensors effectively capture pollutant variations and source differences, providing scientific support for targeted indoor pollution control.
How to cite: Liu, D.: Application of Low-Cost Sensors for Pollutant Source Attribution in Various Indoor Environments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-148, https://doi.org/10.5194/egusphere-egu26-148, 2026.
Low-cost particulate matter sensors (LCPMS) offer scalable and affordable capabilities that complement regulatory-grade monitoring networks by enabling high-resolution urban air quality monitoring. However, they frequently suffer from inaccuracies arising from environmental and temporal variability, necessitating robust calibration approaches to ensure measurement reliability. Co-location studies against reference-grade monitors, combined with machine learning (ML) calibration algorithms, have emerged as effective strategies to significantly improve LCPMS performance. In this study, a long-term co-location experiment was conducted in Vishakhapatnam, India from February 2018 to January 2020, incorporating environmental and temporal co-variates: temperature, relative humidity, hour-of-day, month-of-year. The baseline Linear Regression (LR) model used only raw sensor readings as input. Subsequent models incrementally incorporated environmental variables (temperature and relative humidity), temporal features (hour of day and month of year), and finally all covariates combined. The ML approaches included LR, Random Forest (RF), eXtreme Gradient Boosting (XGB), and a hybrid ensemble combining the best-performing models, with all comparisons made relative to the baseline LR model. Results demonstrate that ML models, particularly the hybrid ensemble, yielded substantial improvements in predictive accuracy. The baseline LR model exhibited an RMSE of 17.62 µg/m³. In comparison, the best-performing RF model achieved a 58% RMSE reduction, while the hybrid ensemble model attained a 63% reduction relative to baseline, satisfying the performance criteria recommended by USEPA. Additionally, we also explore the performance of the models across the temporal, environmental and the AQI category to identify potential performance variations and inform strategies for maintaining reliable measurements across changing environmental and pollution conditions. Although the hybrid model was overall the best, the analysis highlights that no single model consistently performs optimally across all conditions, suggesting that adaptive calibration strategies, such as using different models for different seasons or environmental conditions, are more effective than relying on a single model throughout the year.
To examine the generalizability of this ML-based calibration framework, we used a publicly available co-location dataset (Campmier et al., 2023) of three Indian cities- Delhi, Hamirpur, and Bangalore, wherein RMSE of the baseline model (factory calibration) is 90.5 µg/m³, 123 µg/m³, and 75.3 µg/m³ respectively and a physics-based Köhler theory calibration model reduced RMSE by 66%, 83% and 75% respectively. In comparison, our calibration framework outperformed these results with reductions of 77%, 95%, and 97% in the respective cities demonstrating strong generalizability across different urban contexts. These improvements highlight the advantages of ML-based methods in capturing nonlinear sensor-environment interactions and addressing the limitations of physics-based or factory-derived calibration algorithms, which assume fixed aerosol properties or rely on simplified empirical relationships. Collectively, our findings indicate that ML-based calibration frameworks enhance measurement accuracy and also generalize effectively across geographically diverse urban Indian environments, which are often characterized by high PM₂.₅ levels. The proposed framework demonstrates its potential to serve as a reliable and scalable solution for improving LCPMS performance in large-scale air quality monitoring efforts and is easy to incorporate, computationally less demanding, and agnostic to sensor models, target pollutants, and calibration approaches.
How to cite: Wathore, R., Jadhao, D., Chakraborty, A., and Labhasetwar, N.: Evaluating Environmental and Temporal Performance of Machine Learning Calibration Models for Low-cost Particulate Matter Sensors: A Case Study Across 4 Indian Cities, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-976, https://doi.org/10.5194/egusphere-egu26-976, 2026.
Low-cost air quality sensors (LCS) are increasingly used to complement regulatory monitoring, but their wider adoption is constrained by challenges in data quality and calibration. To address this, we developed Modus+, a novel probabilistic machine learning framework for network calibration and quality assurance. Modus+ maintains indicative-class measurements suitable for public health communication and policy applications while eliminating the need for resource-intensive co-location with a reference.
Modus+ integrates diverse inputs – including satellite data, nearby reference monitors, and local meteorology – to generate probabilistic pollution predictions at each sensor location that serve as a proxy for a co-located reference. Where inputs lack predictive power, prediction intervals widen, providing an explicit quantification of uncertainty in space and time. Comparing predictions with LCS measurements, the system derives simple linear calibrations with confidence intervals on the slope, intercept and bias at the relevant limit value. This enables an evidence-based decision on whether and how to correct individual sensors, while preserving traceability to the underlying measurements. The framework is pollutant and sensor-agnostic and can be applied across diverse networks and operating conditions.
We validated Modus+ through a three-year co-location study and a case study of its operational deployment within the Transport for Greater Manchester (TfGM) sensor network. Twelve low-cost PM sensors were co-located at four reference sites between 2022 and 2025, and for rolling 12-week periods we compared relative expanded uncertainty from (i) uncorrected data, (ii) calibration using short-term co-location (10 days), (iii) calibration using full co-location data, and (iv) Modus+ network calibration. Modus+ significantly improves performance compared to uncorrected data and short-term co-location and achieves the 50% relative expanded uncertainty criterion for indicative measurements. Through our ongoing deployment across the TfGM network, stakeholders have gained a robust understanding of how pollution levels change across the region. This information is being used to explore the impact of local pollution sources, such as domestic wood burning, and aid public engagement.
How to cite: James, H., Stratton, S., and Sykulski, A.: Modus+: A Probabilistic Machine Learning Framework for Calibration of Low-Cost Air Quality Sensor Networks, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1368, https://doi.org/10.5194/egusphere-egu26-1368, 2026.
Solar-powered low-cost sensors revolutionize aerosol tracking and air quality monitoring in quarries, where dust from blasting, crushing, and hauling creates health and environmental hazards. These autonomous systems integrate compact sensors for particulate matter (PM2.5, PM10) and gases with photovoltaic panels and batteries, enabling grid-independent operation in remote, harsh sites lacking power infrastructure.
The study is aimed at addresses critical needs in environmental monitoring for high-pollution industrial sites like quarries. Quarries generate significant aerosol dust and pollutants from blasting, crushing, and hauling, posing health risks to workers and nearby communities, which traditional monitoring often misses due to sparse, expensive stations.
Challenges in emissions monitoring for quarrying include sensor drift from dust and humidity needing periodic calibration, power fluctuations causing data gaps in low sunlight, and high costs for ruggedized units with remote maintenance, while opportunities revealed that solar panels combined with low-cost sensors offer significant prospects for quarry aerosol tracking and air quality monitoring by enabling reliable, off-grid deployment in harsh environments. These systems address power limitations, data gaps, and compliance needs in dusty, remote quarry operations. They support proactive pollution management and regulatory adherence.
This study is necessary because Low-cost solar-powered sensors for quarry aerosol tracking and air quality monitoring offer significant advantages through affordability, scalability, and sustainability. It also spans health/safety gains by reducing worker exposure, regulatory compliance, environmental protection toward zero-emissions, and economic savings via scalability.
How to cite: Aleh, U. I.: Applications of Solar Panels Low-Cost Sensors in Quarry Aerosol Tracking, and Air Quality Monitoring: Challenges and Opportunities, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2326, https://doi.org/10.5194/egusphere-egu26-2326, 2026.
For air quality monitoring, especially at those locations where regulatory networks and challenging infrastructure often affect data collection, remote air quality sensors are an affordable ultimatum.
Clarity Movement provides advanced, IoT-enabled air quality monitoring solutions that combine precision sensing with global connectivity. The Clarity Node-S, integrates solar power, cellular communication, and weatherproof design to deliver reliable air quality data.
Two calibration systems are available: global pre-calibration and custom collocation calibration. The global calibration, applied at the factory using an extensive dataset of over six million measurements, provides consistent baseline performance across PM2.5 and NO2 monitoring meanwhile, custom collocation calibration fine-tunes sensor output can further correct for local conditions, improving measurement precision (R2 > 0.9 in optimal settings) by accounting for regional temperature, humidity, and pollution profiles.
Their ability to maintain accurate performance in remote and variable environments makes them ideal for expanding measurement coverage across urban and rural areas alike. By combining flexible calibration and autonomous operation, Clarity`s system supports accessible and reliable air quality data, advancing public health and environmental research across the world.
How to cite: O`Brien, M.: Global and custom calibration approaches for Clarity`s Node-S air quality measurements., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3436, https://doi.org/10.5194/egusphere-egu26-3436, 2026.
Low-cost air quality sensors are rapidly expanding observational capacity worldwide, particularly in regions with limited regulatory monitoring. However, increasing reliance on complex and often opaque data processing algorithms has blurred the boundary between “true” measurements and model-derived products, complicating data interpretation, comparability, and fitness-for-purpose assessments. Current performance standards largely focus on accuracy metrics, while providing limited guidance on transparency, traceability, and the nature of the underlying data-generating process (DGP).
Here, we present a conceptual and operational framework to classify sensor-derived data products based on their DGP and degree of measurement independence. Building on metrological principles and recent discussions on sensor processing levels, we introduce a formal definition of Independent Sensor Measurements (ISM), supported by five minimum criteria addressing signal dominance, admissible corrections, contemporaneity, signal provenance, and model independence from local data infrastructure. The framework distinguishes independent measurements from non-independent measurements and predictive products, and maps these categories onto an extended processing-level classification scheme.
The proposed classification enables users, manufacturers, and standardization bodies to more transparently communicate what a sensor product actually represents, supporting more appropriate data use, comparability across sites, and informed technology selection.
This work provides the foundation for integrating transparency and traceability into future sensor standards, incentivizing hardware-driven improvements, and strengthening the credibility of sensor deployments in regulatory, research, and community applications, particularly in low- and middle-income regions.
References:
Diez, S. et al. A framework for advancing independent air quality sensor measurements via transparent data generating process classification. npj Clim Atmos Sci 8, 285 (2025). https://doi.org/10.1038/s41612-025-01161-2
How to cite: Diez, S., Chacón-Mateos, M., Malings, C., and Ferracci, V.: From software-assisted predictions to hardware-driven observations: advancing independent air quality sensor measurements, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4119, https://doi.org/10.5194/egusphere-egu26-4119, 2026.
Low‑cost air quality sensors are increasingly deployed to complement reference monitoring stations due to their low cost and ease of installation. However, these sensors are susceptible to environmental conditions and long‑term nonlinear drift, which can substantially degrade data accuracy over time. Existing calibration strategies, particularly those based on machine learning and periodic co‑location with reference instruments, improve performance but they often involve considerable maintenance effort and costs, especially when managing large-scale networks.
Kunak has developed an innovative Automatic Drift Correction (ADC) algorithm that autonomously correct the baseline and sensitivity drift in electrochemical gas sensors. This new method works alongside the Kunak algorithm that compensates for environmental factors such as temperature and humidity across the full operating range. Together, they allow for accurate measurements without the need for reference data or frequent manual recalibrations.
The ADC algorithm enables continuous and sensor-specific baseline and sensitivity adjustments independently from the location and the ambient conditions, ensuring consistent data quality over time, and without using Machine Learning or Artificial Intelligence models. This significantly reduces the operational complexity and costs associated with maintaining air quality sensor networks, especially in large deployments.
We evaluated the proposed method on a NO2 sensors co‑located with a regulatory air quality monitoring station (AQMS) and compared the performance against a conventional manual calibration procedure. Results demonstrate that the ADC algorithm maintains data integrity over time with performance comparable to the periodic manual calibration, even under variable environmental conditions.
The method offers a scalable and reliable alternative to traditional approaches and supports the recommendations outlined by the WMO, which highlight the need for automated, low-effort maintenance solutions.
This work presents a practical and efficient tool for sustaining the long-term reliability of air quality data, making it especially suitable for distributed air quality monitoring networks.
How to cite: Ibarrola-Ulzurrun, E. and Lara-Ibeas, I.: Improving the long-term accuracy of low-cost sensors through a novel Automatic Drift Correction, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5187, https://doi.org/10.5194/egusphere-egu26-5187, 2026.
Calibrations of optical particle spectrometers (OPSs) are non-trivial and conventionally involve aerosolisation techniques, which are challenging for larger particles. In this paper, we present a new technique for OPS calibration that involves mounting a static fibre within the instrument sample area, measuring the scattering cross section (SCS), and then comparing the SCS with a calculated value. In addition, we present a case for the use of generalised Lorenz–Mie theory (GLMT) simulations to account for deviations in both minor- and major-axis beam intensity, which has a significant effect on particles that are large compared with the beam waist, in addition to reducing the need for a “top-hat” spatial intensity profile. The described technique is OPS independent and could be applied to a field calibration tool that could be used to verify the calibration of instruments before they are deployed. In addition to this, the proposed calibration technique would be suited for applications involving the mass production of low-cost OPSs.
How to cite: Girdwood, J.: A Noval Calibration Technique for Optical Particle Counters, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7747, https://doi.org/10.5194/egusphere-egu26-7747, 2026.
Nitrogen dioxide (NO2) is a pollutant emitted primarily by vehicles and is strongly associated with large population centres. NO2 is therefore routinely monitored in cities, but its concentrations in small towns aren’t as well documented, even though some small towns may be traversed by roads with high traffic volumes. This work presents the application of a fast, low-cost cavity-enhanced absorption spectroscopy (CEAS) sensor monitoring of NO2 levels in three small towns in County Cork, Ireland: Macroom, Killeagh and Millstreet. The respective populations of these towns are approximately 4100, 900, and 1800. The spatial and temporal characteristics of NO2 in these towns, and the influence of highly polluting vehicles on local air quality are presented. The potential of the novel CEAS system for characterising NO2 pollution in small population centres is discussed.
How to cite: Edmonds, A. E. and Venables, D. S.: Application of fast, low-cost spectroscopic sensors to measure nitrogen dioxide in small towns in Ireland, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8093, https://doi.org/10.5194/egusphere-egu26-8093, 2026.
Low-cost air quality sensors have emerged as a powerful tool to complement traditional monitoring networks by enabling high spatial and temporal resolution observations at a fraction of the cost of reference instruments. However, their application in indoor environments, particularly in residential settings affected by combustion-based heating, remains limited despite the significant health risks associated with indoor air pollution.
In this study, we present the development and laboratory validation of a low-cost, modular sensor designed for real-time monitoring of fine particulate matter (PM₂.₅), carbon monoxide (CO), temperature, and relative humidity in residential indoor environments. The system is based on open-source hardware and integrates an optical PM₂.₅ sensor, a CO gas sensor, and environmental sensors coupled to an ESP32 microcontroller, enabling continuous data acquisition, local storage (microSD), real-time visualization (OLED display), and wireless data transmission. The device is housed in a custom-designed 3D-printed enclosure optimized for airflow, sensor protection, and portability.
Following laboratory validation against certified reference instruments, the sensor units will be deployed inside dwellings equipped with different residential heating systems, including wood-burning stoves, kerosene heaters, and electric heating. The instruments will be distributed across multiple indoor spaces within each household (e.g., living rooms, bedrooms, and kitchens) to characterize the spatial distribution of pollutants and to assess how combustion emissions propagate through different indoor microenvironments under real living conditions.
A total of ten sensor units were assembled and evaluated under controlled laboratory conditions through side-by-side comparison with reference instruments. The validation protocol focused on accuracy, temporal stability, inter-sensor consistency, and operational robustness. The results show good agreement with reference measurements for both PM₂.₅ and CO, demonstrating that the system provides reliable and stable observations suitable for indoor air quality applications.
The inclusion of CO, temperature, and humidity monitoring represents a key advancement of the system, allowing for a more comprehensive characterization of combustion emissions, ventilation conditions, and indoor comfort. This integrated approach supports both chronic exposure assessment and acute risk detection, including the identification of potentially lethal CO accumulation events in poorly ventilated dwellings.
This work demonstrates the feasibility of deploying low-cost sensor networks for high-resolution indoor air quality monitoring and highlights their potential for citizen science initiatives, stove replacement programs, environmental health studies, and policy support in regions affected by residential combustion emissions.
How to cite: Ríos, B., Loyola-Valenzuela, O., Quintero-Bernal, D., López, M., Tapia, D., and Diez, S.: Development and laboratory validation of a low-cost PM₂.₅ and CO sensor network for indoor air quality assessment in residential heating environments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8205, https://doi.org/10.5194/egusphere-egu26-8205, 2026.
Monitoring air quality in urban areas is essential for assessing environmental pollution and its impact on health and climate, as well as for developing transport and urban planning measures. Legally regulated air quality measurements are based on high-precision reference measuring stations, but their high investment and operating costs mean that their spatial coverage is limited. As a result, small-scale differences in pollutant levels cannot be adequately recorded. Low-cost sensors (LCS) offer great potential here, as they enable dense, continuous and cost-efficient collection of air quality data. At the same time, however, their measurements are often distorted by sensor drift, cross-sensitivities and meteorological influences such as temperature and relative humidity, which limits their direct use for scientific analysis.
We present the UrbanAirLab, a long-term air quality monitoring network on a university campus in Heilbronn (Germany) that will be expanded to cover the city of Heilbronn in the future. The monitoring network is based on self-designed low-cost multi-sensor systems for the continuous recording of NO, NO₂, O₃, CO, PM2.5 and PM10 as well as meteorological parameters. The systems include two thermal low-cost dryers as preconditioning method for the PM and the gas sensors inlets. A central element of the concept is the permanent co-location of selected sensor boxes with an official reference measuring station of the Baden-Württemberg State Agency for the Environment (LUBW), which provides reliable comparative data over long periods of time. The UrbanAirLab is also designed as open-source real-world laboratory and serves to train and involve students and schoolchildren in practical environmental observation and data analysis.
The research design follows an empirical, data-driven approach. The aim is to develop and validate machine learning models that reconstruct reference measurements as accurately as possible from the raw data of the low-cost sensors. Data processing is carried out via a scalable pipeline that enables both the continuous storage of time series data and reproducible calibration modelling and evaluation. Various model approaches are being investigated, including multilinear regression, random forest models and gradient boosting methods.
A particular focus is on investigating seasonal effects, the long-term stability of the calibration models and their transferability to identical sensor boxes. The results presented contribute to the further development of data-driven calibration strategies for low-cost air quality monitoring networks and to the evaluation of their potential for scientific environmental observation.
How to cite: Mannschreck, K., Chacón-Mateos, M., Golder, M., Graf, P., Herrera-Carrión, E., Kieser, J., Lachnit, E., Vogt, U., and Weiland, T.: UrbanAirLab: Data-Driven Calibration of Low-Cost Air Quality Sensors Using Long-Term Co-Location Measurements, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10792, https://doi.org/10.5194/egusphere-egu26-10792, 2026.
Air pollution remains a major environmental and public health challenge. The World Health Organization (WHO) estimates that air pollution is associated with 9 million premature deaths annually. Low-cost sensors (LCS) are a promising complement to regulatory monitoring because they can deliver high frequency, hyperlocal air quality data. However, LCS data quality is affected by limitations of the measuring principle, sensor drift/aging, cross-sensitivities to other compounds, and meteorological influences like temperature (T) and relative humidity (RH), which can undermine reliability and stakeholder trust. In recent years, machine learning (ML) has been widely explored and applied to LCS data to correct systematic biases in raw sensor signals and improve the accuracy of the measurements, yet the frequent lack of explainability of black-box models can further reduce transparency and confidence in the post-processed sensor data.
In the context of the MoDa project and in collaboration with UrbanAirLab project of the University of Applied Sciences in Heilbronn, this study aims to create an explainable ML calibration workflow for LCS NO₂ measurements to enhance transparency of calibration models. The dataset consists of 1-min raw data with a co-location period from 01.06.2025 to 20.11.2025 in a regulatory measurement station located in Heilbronn (urban background). First, an exploratory data analysis (EDA) is carried out, which includes time synchronization of LCS and reference data, handling of missing values, outlier detection with Density-Based Spatial Clustering of Applications with Noise (DBSCAN), and resampling to hourly averages. Then different calibration models are trained including as input parameters the working and auxiliary electrode signals of the NO2 sensor as well as external data such as T, RH and O3 data. The tested models include Multiple Linear Regression (MLR), Support Vector Regressor (SVR), Random Forest Regressor (RF), eXtreme Gradient Boosting (XGBoost), and Artificial Neural Network (ANN). The performance evaluation is carried out using the relative expanded uncertainty as suggested in DIN CEN TS 17660-1 and also other standard metrics such as RMSE, MAE, R², and bias.
The results of these metrics suggest that RF provides the best overall performance (RMSE = 5.50 µg/m³, MAE = 3.93 µg/m³, R² = 0.69; Pearson r = 0.83) and near-zero mean bias. XGBoost performs similarly (RMSE = 5.62 µg/m³, R² = 0.69), followed by ANN (RMSE = 5.76 µg/m³, R² = 0.67).
Explainable ML techniques are implemented in a second step as an auditing layer to support data quality assurance and control (QA/QC). These include Permutation Feature Importance (PFI) to screen which predictors most affect out-of-sample performance by measuring the score drop after removing each feature, SHapley Additive exPlanations (SHAP) for global and local attributions, and Individual Conditional Expectation (ICE) and Partial Dependence (PDP) Plots to summarize average effects while exposing heterogeneity and interaction patterns. Because predictors such as T and RH are often correlated in co-location datasets, we also use Accumulated Local Effects method to obtain more reliable effect estimates under feature dependence.
By combining reproducible calibration models with systematic explainability, this work supports more transparent QA/QC practices and contributes to creating transferable workflows for deploying LCS for air-quality monitoring.
How to cite: Chacón-Mateos, M., Herrera-Carrión, E., Golder, M., Mannschreck, K., Vogt, U., Diez, S., Grein, T., Kieser, J., Reiland, S., Gaiser, N., and Köhler, M.: Opening the Black Box: Explainable machine learning techniques for air quality sensor calibration , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11877, https://doi.org/10.5194/egusphere-egu26-11877, 2026.
Dust storms (DSs) are a common phenomenon in many areas worldwide and a natural source for particulate matter (PM) in the atmosphere. Yet the knowledge about the impacts of DSs on the spatio-temporal distributions of PM in urban environments is limited. Air pollutants are regularly monitored by environmental stations in very few locations within large cities. This limits our ability to assess the exposure to PM in space and time. A kit of low-cost sensors (LCS) was developed in our lab (BGSense) to measure air-quality and meteorological data in outdoor and indoor environments, with high spatial and temporal resolutions. The BGSense is designed to be used in stationary or mobile modes of measurement networks in the city. Data from the BGSense sensors are being compared to the reference instruments of the environmental station, located in an arid urban environment. The sensors are tested under various meteorological conditions, including DSs (1-hour average PM10 concentrations < 100 µg m-3), in which hourly concentrations can reach ~2000 µg m-3. The calibration process shows strong correlations (R2=0.9) between the reference instruments and the BGSense for both air temperature and relative humidity. The PM data of BGSense vs. TEOM are well correlated for PM10 in DS (R2=0.8) and in non-DS (R2=0.7) time periods. A similar trend is obtained also for PM2.5. Preliminary measurements, done simultaneously with BGSense kits in several locations around the university campus, demonstrate variations in PM10 concentrations in space and time. The BGSense has the potential to provide high-resolution data to explore the dust-PM distribution and health exposure risk in urban environments.
How to cite: Cohen, N., Kloog, I., Naroditski, S., Yousufov, N., Dorman, M., and Katra, I.: A kit of low-cost sensors for measurements of dust-PM in arid urban environments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13945, https://doi.org/10.5194/egusphere-egu26-13945, 2026.
Ozone (O₃) is a pivotal trace gas influencing global public health, ecosystem stability, and radiative balance. However, the scarcity of long-term, high-resolution air quality observations across the Tibetan Plateau (TP) hinders the understanding of O₃ dynamics in this climatically and ecologically sensitive region. In this study, we integrated a high-density, low-cost sensor network (LCSN) along key inflow corridors in the TP with EAC4 reanalysis data for three-dimensional spatiotemporal analysis. Iterative discrete wavelet transform (IDWT) and Lagrangian transport diagnostic methods were innovatively used to quantify the dominant role of meteorologically driven regional transport and photochemical generation from local anthropogenic sources in driving surface O₃ extremes. Results show that the fusion of LCSN and reanalysis data provided a reliable, dynamic 3-D dataset at high temporal and spatial resolution for exploring regional O₃ production and transport. Key quantitative findings indicate that extreme surface O₃ pollution during summertime was driven by a coupled “stratosphere–monsoon” mechanism: stratospheric intrusion (SI) (contributing ~50.2% of the pollution signal) overlapped with monsoon-driven long-range transport of polluted air masses from upwind South Asia (~28.7%), while local photochemical generation played a lesser role (~21.1%). Dry conditions and enhanced solar radiation acted as critical amplifiers of O₃ pollution over the plateau. These findings provide the first observationally constrained, quantitative fine-scale source attribution of summertime surface O₃ extremes in the TP, demonstrating the critical role of LCSNs in supplementing traditional monitoring. The study provides a transferable framework for applying affordable LCSNs in high‑altitude or resource‑limited environments, supporting both the formulation of targeted mitigation strategies and collaborative international air‑quality management.
How to cite: Chen, W., Qin, X., Tao, S., Wang, Y., Wang, Y., Yuan, Z., Zhuoga, S., Zhang, H., Fu, Q., and Ning, Z.: Integrating Low-Cost Sensors with Multiscale Models to Quantitatively Identify Ozone Sources and Transport Patterns over the Tibetan Plateau, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15217, https://doi.org/10.5194/egusphere-egu26-15217, 2026.
Poor indoor air quality (IAQ) poses significant health risks to all, as people spend up to 90% of their time indoors, particularly affecting vulnerable groups such as children. Within the EDIAQI (Evidence Driven Air Quality Improvement) project, exposure to indoor air pollution in households with asthmatic children is being investigated through non-invasive, low-cost sensing approaches. To meet these requirements, TROPOS developed the AQBIE (Air Quality Beacon and Immission Evaluator), a compact, robust, and silent monitoring device designed to unobtrusively integrate into children’s bedrooms.
In contrast to commercial air quality monitors, AQBIE integrates three distinct particulate matter (PM) sensors, allowing for improved source classification and size differentiation. With a 10-second time resolution, the system provides detailed insights into dynamic indoor pollution events. Active user interaction supports event labeling, adding valuable context to the sensor data.
During a three-month field campaign in the city of Zagreb (Croatia), 15 AQBIE units operated continuously across households, covering the transition from late summer to the onset of the heating season. Real-time data transmission via a robust and redundant MQTT-based infrastructure enabled permanent monitoring and remote control without on-site intervention, proving to be highly reliable with data coverage vastly exceeding 99%.
AQBIE shows how open-source IoT technologies can serve scientific research while engaging stakeholders and building IAQ awareness in a playful, accessible way. Here we present the device design, data acquisition and data transmission architecture, and preliminary field campaign results. This positions AQBIE as a flexible, low-cost platform for scalable IAQ networks. Ultimately, the collected data will support lung deposition modeling and contribute to the development of health-relevant exposure metrics.
This work was supported by the affiliated institutions and the Horizon Europe project EDIAQI, grant ID: 101057497
How to cite: Förster, J.-D., Düsing, S., Cuesta-Mosquera, A., Winkler, U., Gajski, G., Gerić, M., Voigtländer, J., and Pöhlker, M. L.: Indoor Aerosol Characterization on a Shoestring – Recent Insights from the EDIAQI Project, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18636, https://doi.org/10.5194/egusphere-egu26-18636, 2026.
Air pollution is one of the main environmental risks to human health, especially in large urban centers, and is associated with cardiovascular and respiratory diseases, as well as other adverse outcomes. Despite its relevance, systematic air quality monitoring is still incipient in most Brazilian cities, particularly in the Northeast of the country. In this context, the present study aims to analyze approximately one year of monitoring of fine particulate matter (PM₂,₅) and inhalable particulate matter (PM₁₀) in the city of Recife, Brazil, using low-cost sensors, seeking to identify temporal and spatial patterns and potential risks to public health. Nine monitoring stations were installed in areas with distinct socio-environmental characteristics, considering factors such as urban density, vegetation cover, vehicle flow, and population income. The measurements, carried out between November 2023 and September 2025, were obtained using IQAir AirVisual Outdoor model sensors, with data available in near real-time. The analyses included descriptive statistics, assessment of exceedances of the limits recommended by the World Health Organization (WHO), analysis of variance, correlation with meteorological variables, and spatial clustering. The results indicate a daily average PM₂₅ value of 6 µg/m³, with significant variations between stations. On 92 days of the analyzed period (13.16%), at least one station recorded concentrations exceeding the WHO daily limit (15 µg/m³), evidencing recurring episodes of air pollution. Well-defined seasonal patterns were highlighted, with higher concentrations in the months of August and September, in addition to extreme peaks observed recurrently at the end of June, associated with local cultural events, such as the June festivities. Spatial analysis revealed intra-urban inequalities in exposure to particulate matter, with one station showing systematically higher concentrations, possibly related to local emission sources and unfavorable socio-environmental conditions. The findings demonstrate the feasibility and relevance of using low-cost sensors to expand air quality monitoring in urban contexts with data scarcity. Furthermore, the results provide important input for the formulation of intersectoral public policies in the areas of health, urban planning, and the environment, contributing to the reduction of socio-environmental inequalities and risks to public health.
How to cite: Menezes da Silva, C. E. and Vasconcelos Bezerra, A. C.: Air Quality Monitoring Network with low-cost sensors in Recife - Brazil., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-22144, https://doi.org/10.5194/egusphere-egu26-22144, 2026.
Low-cost air quality monitors are a crucial tool for communities to document exposure to pollutants. Their cost and ease of installation and maintenance allow residents to set up networks of monitors to identify air pollutant levels and sources within neighborhoods. We show results from a network of PurpleAir air quality monitors deployed in the Washington, DC neighborhoods of Buzzard Point, Brentwood, and Ivy City over the course of 3 years. The data collection was initiated by residents of these overburdened communities surrounded by a major thoroughfare, industrially-zoned areas, bus and heavy-duty vehicle parking, diesel truck traffic and railroad tracks.
Several PurpleAir monitors were deployed at volunteer residents across the neighborhoods to measure atmospheric concentrations of fine particulate matter (PM2.5). Additionally, we co-located two black carbon (BC) monitors with PurpleAir monitors in Brentwood and Ivy City. Using this dataset, we document the temporal and spatial variability of PM2.5 concentrations within the neighborhoods and compare local to city-wide concentrations of air pollutants from regulatory monitors and publicly available PurpleAir monitors. Through the relative variations of PM2.5 and BC and their correlation to the local meteorology, we can separate regional sources of air pollution from local ones (e.g. major roadways in Brentwood and construction sites in Buzzard Point).
How to cite: Dutta, A., Arminio, D., Rosa-Rivera, C., Hsieh, S., and Aquila, V.: Combining Black Carbon and PM monitoring to identify local sources of pollution in Washington, DC, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14897, https://doi.org/10.5194/egusphere-egu26-14897, 2026.
The study is aimed at investigating the awareness of risks associated with the exposure to atmospheric aerosol at quarry site and its health implication in Ebonyi state, and to examines how low-cost air quality sensors can enhance monitoring and management efforts. The questionnaire used covers the demographic information, awareness of occupational health hazards, use of personal protective equipment, health effects experienced by workers, and suggestions for improvement. A total of three hundred and fifty (350) questionnaire were distributed to respondents. a sample of one hundred and eighty-five (185) for quarry workers, quarry owners/managers, community members living near quarry sites, sixty-five (65) healthcare providers, fifty-five (55) environmental protection agencies and forty-five (45) policymakers was selected through random sampling. the data collected was statistically analyzed using frequency counts and mean. a total of three hundred and forty-seven (348) were returned, two (2) were torn and five (5) were wrongly filed. a total of three hundred and forty-three (343) were accepted and assembled for analysis. The findings on the awareness of occupational health hazards among quarry workers indicate a concerning lack of knowledge and training in this field. The findings from assessing the use of personal protective equipment (PPE) among workers indicate varying levels of compliance with safety measures. The findings from investigating the health effects of workers' exposure to aerosols at a quarry site reveal significant impacts on their well-being. The findings from the investigation aimed at identifying suggestions for improving a conducive work environment at the quarry indicate enhancement of health care. The responses from questionnaire provide valuable insights into the current state of occupational health and safety at quarry sites in Ebonyi State and help identify areas for improvement. The research reviewed that lack of low-cost air quality sensors for monitoring of aerosol from quarry station is hindering the awareness of risk of exposure. Low-cost air quality sensors offer a practical solution for monitoring these risks, enabling real-time data collection that informs both operational practices and community engagement efforts. Critically, the integration of low-cost air quality sensors into the environmental management framework at quarry sites in Ebonyi State can significantly enhance the understanding and control of air pollution. By providing real-time data and fostering community involvement, these sensors can play a pivotal role in mitigating the adverse effects of quarrying on air quality and public health. Therefore, collaborative approaches to help in having access to low-cost air quality sensors in Nigeria, research grants and sponsorship for training are the panacea for clean air quarry sites of Ebonyi State.
How to cite: Ibeh, Dr. G. F. and Aleh, U. I.: Investigating Awareness of Atmospheric Aerosol Exposure Risks and the Criticality of Low-Cost Air Quality Sensors at Quarry Sites in Ebonyi State, Nigeria , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4221, https://doi.org/10.5194/egusphere-egu26-4221, 2026.
Posters virtual: Wed, 6 May, 14:00–18:00 | vPoster spot 5
EGU26-14158 | Posters virtual | VPS4
Open-Tool Frameworks for Cross-Platform Indoor Monitoring and Optimized Air Cleaning StrategieWed, 06 May, 14:30–14:33 (CEST) vPoster spot 5
Indoor environments account for most human exposure to air pollution, yet indoor air quality (IAQ) monitoring and control remain fragmented across devices, platforms, and proprietary building automation systems. Commercial IAQ monitors and smart thermostats are widely available, but they typically operate in closed ecosystems with limited interoperability. In parallel, open-source communities have demonstrated the potential of low-cost sensing networks, yet these solutions rarely connect to building-level control systems capable of simultaneously reducing pollutant exposure and energy use. To address this gap, we present an open, interoperable framework that integrates open-source and commercial technologies for IAQ monitoring, data management, and automated building control[1]. The framework, developed within the EU-funded healthRiskADAPT project, is built on an open, production-ready IoT infrastructure for indoor environments. At the edge, low-cost sensor nodes collect and transmit environmental data. A web-based interface allows users to register locations, nodes, and sensors, and provides near-real-time visualization, historical analytics, and an interactive map of the sensor network. Beyond monitoring, the framework enables direct integration with commercial control devices such as smart thermostats, smart plugs, and filtration systems. This interoperability supports data-driven control strategies, including increasing ventilation during indoor pollution events, activating filtration during periods of poor outdoor air quality, and dynamically adjusting HVAC operation to balance comfort, energy use, and exposure reduction. By combining continuous mass-balance modeling[2] with real-time sensor data, the system will deliver actionable indoor-outdoor (I/O) ratios and exposure indicators. These outputs could drive automated responses but also support informed user behavior, such as choosing higher-efficiency filters during high-pollution episodes, using kitchen exhaust during cooking, or understanding the trade-offs between energy costs and health risks. In this way, the platform functions not only as a control system but also as an educational and decision-support tool for occupants and building managers. This presentation demonstrates how open-source hardware, open APIs, and modular integration pathways can create a flexible, transparent, and scalable ecosystem for IAQ management. The framework supports diverse use cases, homes, schools, workplaces, and research settings, while offering a roadmap toward energy-efficient, healthier indoor environments driven by interoperable technologies rather than isolated products.
[1] https://particularmatter.org
[2] Dallo, Federico, Thomas Parkinson, Carlos Duarte, Stefano Schiavon, Chai Yoon Um, Mark P. Modera, Paul Raftery, Carlo Barbante, and Brett C. Singer. "Using smart thermostats to reduce indoor exposure to wildfire fine particulate matter (PM2. 5)." Indoor Environments 2, no. 2 (2025): 100088.
How to cite: Dallo, F., Tenti, L., Palo, A., Parkinson, T., and Duarte, C.: Open-Tool Frameworks for Cross-Platform Indoor Monitoring and Optimized Air Cleaning Strategie, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14158, https://doi.org/10.5194/egusphere-egu26-14158, 2026.
EGU26-130 | ECS | Posters virtual | VPS4
Field-Calibrated Low-Cost Sensor Networks for PM2.5 Monitoring in West African Urban Environments: Insights from Abidjan and AccraWed, 06 May, 14:33–14:36 (CEST) vPoster spot 5
Low-cost air quality sensors (LCS) offer opportunities for expanding air monitoring networks in regions where reference-grade instrumentation is limited. Within the framework of the Improving Air Quality in West Africa (IAQWA) project, we deployed Real-time Affordable Multi-Pollutant sensors (RAMPs) in Abidjan (Côte d'Ivoire) and Accra (Ghana), to characterize fine particulate matter (PM2.5) in contrasting urban environments. Prior to field deployment, each RAMP underwent a co-location period with reference monitors, and city-specific multilinear calibration models were developed incorporating both RAMP-reported PM2.5 and relative humidity (RH). These calibration models were applied to correct the sensor data and improve measurement reliability under varying atmospheric conditions.
From February 2020 to June 2021, five (5) measurement sites in Abidjan and four (4) sites in Accra were monitored using a 15-second temporal resolution. These sites were selected to represent the dominant pollution sources in West Africa, particularly domestic fires and road traffic. The calibrated dataset enabled comparative analysis of diurnal, daily, and seasonal PM2.5 variability. Both cities exhibited pronounced morning PM2.5 peaks associated with traffic, while evening increases were more visible in residential areas, indicating contributions from domestic combustion. Seasonal contrasts were marked, with highest concentrations occurring during the long dry season (Harmattan), when long-range Saharan dust transport significantly enhanced particulate loading. During an intense dust episode in January 2021, calibrated RAMP data underestimated PM2.5 relative to reference measurements, highlighting a known limitation of optical LCS under high mineral dust conditions.
Annual mean PM₂.₅ concentrations ranged from 17 to 26 µg m-3 across sites, exceeding both the 2005 and 2021 WHO air quality guidelines. Variability within each city, especially between traffic-influenced and urban background locations, was greater than variability between the two cities. These findings demonstrate both the value of rigorously calibrated low-cost sensors for improving air quality knowledge in data-scarce urban regions, and the need for sensor performance considerations in environments influenced by episodic dust intrusions.
How to cite: Bahino, J., Giordano, M., Beekmann, M., Ramanchandran, S., and Yoboué, V.: Field-Calibrated Low-Cost Sensor Networks for PM2.5 Monitoring in West African Urban Environments: Insights from Abidjan and Accra, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-130, https://doi.org/10.5194/egusphere-egu26-130, 2026.
EGU26-6264 | Posters virtual | VPS4
Estimating Size-Resolved Lung Deposition Doses of Particulate Matter (PM) Using Low-Cost Sensor Data in Rural IndiaWed, 06 May, 14:36–14:39 (CEST) vPoster spot 5
Background and Objective
Exposures to fine and ultrafine particles (i.e., PM2.5 and PM1) are widely accepted as a major environmental risk factor and is known to cause adverse human health outcomes. Most epidemiological research as well as regulatory frameworks rely on using PM2.5 as the ‘reference’ exposure metric to assess health risks. However, this approach does not adequately quantify size-specific PM effects that are critical for dose-based health assessments. Size-segregated particulate matter assessment of exposures and effects are limited in resource-limited settings. The objective of this study is to estimate lung deposition doses for size-fractionated PM measured using low-cost sensors.
Methods
We utilized the data obtained from an ongoing study conducted in South Indian villages. Here, the household energy use is dominated by biomass combustion and the adoption of cleaner cooking fuels like liquefied petroleum gas (LPG) is relatively low. Ambient PM measurements were carried out continuously over a period of 1 year in 80 rural households in southern India, using real-time, optical, low-cost PM sensors. In order to capture the household-level exposure characteristics, indoor PM measurements were also carried out in a subset of households. Minute averaged, PM mass concentrations in three discrete size fractions: PM₁, PM₂.₅ and PM₁₀ were provided by the low-cost sensors. The temporal variability in PM concentrations was derived using the time-series data obtained from the sensors. Daily and monthly mean concentrations captured the short-term exposure peaks as well as day to day variability.
Results
A mathematical model using a non-linear least squares method was developed to transform the measured PM concentrations into a continuous size distribution. Respiratory deposition doses were estimated by feeding the size distribution to a computational model of the lung designed to simulate the spatial and temporal distribution of particles within the human respiratory system, incorporating various deposition models. The estimates of deposition doses ranged from ~0.2µg/min to ~1µg/min in the total lung. The coarse particles contributed to about 20% of the total lung dose, whereas the remaining 80% of the respiratory dose was predominantly of fine and ultrafine particles.
Conclusions
This study demonstrates that physiologically relevant, size-fractioned lung deposition doses can be estimated using limited size-bin data obtained from low-cost sensors. Since low-cost air quality monitoring networks are critical in regions that lack regulatory-grade instrumentation, the proposed analytical framework provides a benchmark for translating low-cost sensor-based air pollution measurements into relevant health-based dose metrics. The proposed analytical framework can be readily modified to incorporate satellite-derived PM inputs alongside low-cost sensor data, enabling improved spatial scaling of size-resolved, dose-relevant exposure estimates.
How to cite: Sai Ganesan, K. D., Puttaswamy, N., Sendhil, S., Natesan, D., Ramasami, R., Desai, M., Pillarisetti, A., Vakacherla, S., Krishnan, R., Sambandam, S., Ramaswamy, P., and Balakrishnan, K.: Estimating Size-Resolved Lung Deposition Doses of Particulate Matter (PM) Using Low-Cost Sensor Data in Rural India, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6264, https://doi.org/10.5194/egusphere-egu26-6264, 2026.
EGU26-11502 | ECS | Posters virtual | VPS4
Regional Air Quality Management: A Scalable, Data-Driven Airshed Framework using Low-Cost Sensors across the Indo-Gangetic Plain, IndiaWed, 06 May, 14:48–14:51 (CEST) vPoster spot 5
In India, ensuring clean air for all is vital and should not be limited to urbanites. However, the current air quality monitoring networks and clean air strategies are limited to cities. Notably, the air quality status across regions is yet to be measured, and comprehensive regional management plans are non-existent. To address these significant research gaps, the Ambient air quality Monitoring over Rural areas using Indigenous Technology (AMRIT) project was envisioned and implemented by the Centre of Excellence Advanced Technologies for Monitoring Air-quality iNdicators (CoE – ATMAN), Indian Institute of Technology Kanpur. For the first time, over 1,400 low-cost PM2.5 sensors were installed across the states of Uttar Pradesh and Bihar at the block level. The sensor locations encompass a diverse range of land-use and land-cover categories, demographics, and communities. By leveraging a dense network of low-cost sensors, we developed a data-driven machine learning framework to delineate airsheds for regional air quality management for the first time. We utilized a recurrent neural network–based long short-term memory (LSTM) and hierarchical clustering algorithms with PM2.5 and meteorological data to delineate airsheds. The LSTM embeddings learn latent representations from PM2.5–ventilation coefficient (VC) time series, capturing spatiotemporal patterns and inter-variable relationships. These embeddings were then hierarchically clustered to delineate airsheds. Applying this framework, our results for Bihar show five distinct airsheds; three prevail in the north of the Ganges River, and two prevail in the south of the Ganges. Notably, Airshed 1, located in the northwest region, is highly polluted. However, during the post-monsoon and winter across the airsheds, PM2.5 levels were two to three times higher than the national standard (60 µg/m³), and on ~90% of days, people breathe unhealthy air. Similarly, we identified multiple distinct airsheds in Uttar Pradesh, as well as common airsheds that prevail across Bihar and Uttar Pradesh states. This emphasizes not only shared regional influences but also the need for an integrated approach to reduce PM2.5 pollution. Therefore, the identified airsheds would be instrumental in targeting the reduction of fine particulate pollution across the Bihar and Uttar Pradesh states. Furthermore, the scalable, data-driven airshed delineation framework using low cost sensors could be implemented across India and potentially globally. Thus, this will facilitate airshed-based air quality management plans and integrated policy interventions to ensure “clean air for all.”
Keuwords: Air Quality; Low-Cost Sensors; Airshed; Indo-Gangetic Plain; Machine Learning
How to cite: P Chandrasekaran, A., Tripathi, S. N., Godhani, N., Pandey, M., Rai, P., Agrawal, N., Kumar, A., and Ballav, S.: Regional Air Quality Management: A Scalable, Data-Driven Airshed Framework using Low-Cost Sensors across the Indo-Gangetic Plain, India, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11502, https://doi.org/10.5194/egusphere-egu26-11502, 2026.
