Assessment and application of a low-cost air quality sensor network

Lead Research Organisation: University of Southampton
Department Name: School of Ocean and Earth Science


Rationale: The issue of air quality has an impact on every human on the planet and there are many initiatives to reduce pollution sources. These initiatives can only target pollution hotspots and measure the effects if we have reliable sensor data. However, monitoring air quality is not a simple task and pollution sources can be very localized, e.g. a single traffic light, road or industry. Reference standard pollution sensors are very expensive (£25k+) resulting in relatively few sensor stations to cover large cities with multiple pollution sources. This PhD will focus on low-cost sensors, applied to air quality.
Seven cities in the UK have been identified as breaching air quality regulations on a regular basis, Southampton is one of them. A government report has recently outlined a loose strategy to improve this situation, but it is not clear if the mitigation will be sufficient to bring Southampton to within regulation limits, particularly given the lack of monitoring currently in place. Indeed, similar measures implemented elsewhere have frequently failed to yield their stated aims potentially due to lack of data with sufficient granularity.
Ultra-low cost gas sensors exist (under £50) but are generally thought to be of such low quality that they are largely inappropriate for air quality monitoring, but offer the potential for a relatively inexpensive large scale network that may be useful, provided sufficient data quality checks/screening are in place. Conversely there exist a range of sensors and particulate monitors in the low-medium cost (£500-5000) category that can potentially be used gain insights into air quality, this however would provide a less resolved network but one with more reliable data.
The solution to which strategy (low-cost/high resolution vs. higher-cost/lower resolution) provides the optimal air quality sensor network remains to be determined.


10 25 50