Community based air pollution measurements in Europe: Opportunities and challenges

Exposure to particles with a diameter smaller than 2.5 μm (PM 2.5 ) poses a threat to human health. Monitoring of PM 2.5 often relies on sparse networks of standardized instruments, but in many locations, official networks are supplemented by networks of low-cost PM sensors operated by members of the community. This study aims to evaluate the open low-cost PM 2.5 sensor networks across Europe by analysing data from Sensor.Community and OpenAQ, assessing completeness and reliability as well as growth, coverage, locations, and correlation in comparison with official measurements. As of 2023, there were 8735 low-cost PM 2.5 sensors in Europe publicly sharing data on these networks, or more than three times the number of official stations. Results indicate these sensors are primarily in residential urban areas with higher population densities, higher income levels, and average levels of ambient PM 2.5 . The analysis shows that although official measurement stations provide more consistent coverage across Europe, in many cities the low-cost PM 2.5 sensor network is denser and can potentially complement the official network and provide insights for authorities, researchers and citizens. The rapid growth of openly available low-cost PM 2.5 sensor data presents a growing opportunity to enhance the spatial coverage of monitored PM 2.5 exposure estimates and understanding of local PM 2.5 sources, both indoors and outdoors. However, challenges exist with inconsistent methods for data quality control, absent metadata, and sensor placement and location biases which need to be addressed for example through harmonized measurement protocols, collecting more complete metadata and with advanced data cleaning and processing methods. • Low-cost community operated PM sensors show rapid growth across Europe. • Community operated sensors provide useful PM 2.5 data in many European cities. • Sensors are biased towards residential, higher-income locations. • Issues with data quality, reliability, and representativeness are key challenges.