The applicability and challenges of black carbon sensors in monitoring networks

Abstract

Black carbon (BC) is a particulate pollutant emitted as a by-product of combustion. BC has an emerging role in air quality monitoring with the current recommendations by the World Health Organization to monitor BC to capture its temporal and spatial variability. To observe this variability, especially in urban areas, a large quantity of sensor-type measurements is required. In this study, four different types of small-scale filter-based BC sensors (AE51, MA200, MA350, and Observair) were used to build a sensor network in Kumpula campus, Helsinki, Finland. Our aim was to test the applicability of the sensors to monitor ambient BC concentrations in field conditions and to study the variation of BC at high resolution. The results were compared to a reference level instrument (multi-angle absorption photometer, MAAP) for validation. During intercomparisons, the sensors had a good correlation with the reference and, after a simple orthogonal regression calibration, were deemed suitable for deployment in the sensor network. During deployment, the sensor network proved to be able to capture small-scale temporal and spatial differences in BC concentrations. Changes in temperature (T ) and relative humidity (RH) were observed to induce error in the BC measurements. This error was amplified by the dual-spot correction, which was worsening the measurement result under unstable conditions of T and RH. This should be considered when using sensors that apply the dual-spot correction automatically. The environmental compensation used by the Observair sensors reduced the error from the changing T and RH. To reduce the effect of changing T and RH, more robust environmentally controlled boxes should be developed, or correction algorithms, such as environmental compensation, should be applied.