Urban Air Quality Monitoring with EO

CAMS runs both global and European-scale atmospheric models that assimilate satellite and in-situ data to produce daily forecasts of air quality indicators such as nitrogen dioxide (NO₂), particulate matter (PM), and ozone. At the global level, CAMS operates at a spatial resolution of 40 km, while its European component reaches 10 km resolution. The service includes tools for source attribution—showing, for instance, how traffic or residential heating contributes to pollution events—and custom scenario modelling, allowing users to simulate the impact of emission reductions by sector. All datasets are openly accessible, and CAMS data is frequently used as a base layer in national and local services. As part of its National Collaboration Programme, CAMS is supporting the development of localised applications in countries such as Bulgaria, Italy, Germany, and Hungary, with use cases ranging from PM source identification to regional emission forecasting and policy support tools.

Figure: Global and Regional CAMS systems

Lobelia, a Barcelona-based company, has developed an advanced urban air quality monitoring platform (Lobelia Air) in collaboration with KNMI using the RETINA algorithm. Lobelia Air integrates a variety of datasets—including CAMS forecasts, Sentinel-5P satellite measurements, ground-based sensors (where available), meteorological data, and urban emission proxies like traffic density and population distribution—to produce high-resolution maps of air pollution across entire urban areas. In cities like Barcelona, Madrid, and Sabadell, the platform offers near street-level mapping of NO₂ at resolutions between 25 and 100 metres. It provides historical trends (from 2018 onwards), real-time monitoring, and 48-hour forecasts. Crucially, the platform has been adopted by public authorities in the Barcelona metropolitan region and Sabadell to support the planning of low-emission zones and assess pollution exposure near schools. For citizens, Lobelia Air also offers visual tools such as "healthier route" maps that suggest cleaner commuting paths, helping to raise awareness and encourage behavioural change.

Figure: Street-level resolution of air quality data in Barcelona in the Lobelia Air platform  

Murmuration’s case in Sarajevo, funded through the European Bank for Reconstruction and Development (EBRD) and ESA’s Global Development Assistance (GDA) programme, focused on assessing the environmental impact of new public transport infrastructure on local air quality. The Murmuration approach combines Copernicus atmospheric data with machine learning (XGBoost) to downscale NO₂ and PM concentrations to 1 km resolution. The model incorporates meteorological inputs, emissions data from the EDGAR inventory, population density, road networks, and in situ sensor data. Despite having only five air quality monitoring stations available for training and validation, the team developed an interactive dashboard that visualises daily pollution levels and compares them against WHO thresholds using a "polluted days" indicator. Early findings indicated that particulate matter levels rose during the construction phase (2020–2024), while post-construction NO₂ levels remained high, highlighting the need for longer observation periods and more robust sensor networks. Nonetheless, the dashboard proved to be an effective demonstration of how EO-based tools can be tailored for urban policy evaluation in cities with limited monitoring infrastructure.

Figure: User interface for air quality monitoring in Sarajevo provided by Murmuration