Suitability of Particulate matter low-cost sensor data for land-use-based modeling

Urban climate research is primarily based on point or location-based measurements of climate and air quality data. Advances in microsensor technologies and low cost due to production facilities, sensors for measuring air quality have spread to a large scale in the last decade. In contrast to reference devices, sensors are simple in design, lightweight, and easier to deploy in larger scales.

United Nations environment programme regards Air pollution to be a major global health concern that causes one in nine deaths worldwide. Although a number of sources and factors have been identified as a cause of air pollution it is difficult to pinpoint a particular source and manage it due to the variability of the pollutants in space, time, and the socio-economic factors involved.

This study aims to answer two questions: the possibility to use mobile measurement systems to identify the major sources that contribute to higher aerosol concentration in time and space and if low-cost particulate matter sensors are suitable for land-use-based modeling using conventional regression analysis and machine learning methods such as random forest and neural networks.

The data used was collected using the OPC-N2 sensor from Alphasense Ltd., as a part of the measurement campaign of the project Urban Climate under Change, Phase 1 and 2 [2016-2023].