10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
The Impact of Ambient Conditions on the Performance of a Low-cost Air Quality Sensor Package (Koala)
XIAOTING LIU, Rohan Jayaratne, Phong Thai, Matthew Dunbabin, Lidia Morawska, Queensland University of Technology
Abstract Number: 1132 Working Group: Low-Cost and Portable Sensors
Abstract Low-cost sensors have been widely used in environmental monitoring. With respect to air quality monitoring, the application of large numbers of low-cost sensors can implement the results of regulatory air quality monitoring. The flexibility of the low-cost sensors enables them to be installed to cover a wide proportion of the targeted area. In this study, the performance of a range of low-cost sensors was tested.
In this study, a low-cost sensors package equipped with a particulate mass (PM) and a carbon monoxide (CO) were developed, calibrated and tested by the Queensland University of Technology, Australia. They were installed at an established air quality monitoring station at a typical roadside location and operated in parallel with standard instruments such as a tapered element oscillating microbalance (TEOM) and a CO analyser for a continuous period of three weeks. Meteorological parameters such as temperature, relative humidity, wind speed and wind direction were also obtained from the air quality monitoring station. The PM and CO data acquired were compared with the data collected by the reference station.
Here, we present the results obtained on a typical 24-hour day of measurement – the 26th of July 2017, by one of the sensor packages. We observed that both PM and CO concentrations from the sensor package followed a similar trend, with the measured values from the reference station. Fairly good linear relationships (Pearson r = 0.80 for PM2.5 and r=0.92 for CO) were also found between the sensor package and the reference instruments. The PM2.5 concentrations from both reference station and sensor package showed high correlation coefficients (Pearson r = 0.72 for reference instrument and r=0.90 for sensor package) with relative humidity. The possible reason could be the temporal distribution of the PM2.5 during this day had the same variation trend as the relative humidity which had a high value during the night time and low value during the day time. Meanwhile, the CO concentration from both reference station and sensor package had low correlation relationships (Pearson r = 0.48 for reference instrument and r=0.45 for sensor package) with relative humidity.
A limitation observed in this study is that the PM2.5 sensor in the sensor packages underestimated the particle number concentration from motor vehicle emissions. This was because the lower detection limit of the low-cost PM sensors was 300 nm, which is significantly higher than the size of most of the vehicle emission particles. The CO concentration value shown by the sensor package was consistently lower than the value from the reference instrument. This showed that an offset value needs to be added to the observed readings in future applications. Analysis of more collected data from sensor packages and reference station need to be completed to have a better understanding of the performance of the low-cost air quality sensor package.