Impact of sampling frequency on low-cost PM sensor performance including short-term temporal events in high PM environments

Abstract

Low-cost sensors (LCSs) for particulate matter (PM) monitoring have gained popularity due to their affordability, compact size, and low power requirements. These sensors typically offer the capability to collect data at sampling rates that can be adjusted according to the application. However, the effect of varying the sampling frequency on sensor performance has not been thoroughly examined. This study explores how variations in sampling frequency influence the performance of low-cost PM sensors and identifies some possible use cases for different sampling rates. During this study conducted over a 1-month period, data from five SPS30 sensors were collected at 15 s intervals and then aggregated into 5, 10, 15, 30, and 60 min intervals. During the study, hourly PM2.5 levels ranged from 117 to 303.3 µg m−3, with significant diurnal variations, influenced by temperature and humidity. It was found that changes in sampling frequency had minimal impact on sensor performance, as evidenced by comparable linearity and error metrics across different sampling intervals. However, the study also revealed that short-lived plume events could be missed at lower sampling frequencies. This suggests that for monitoring gradual changes in PM2.5 levels, higher sampling frequencies do not necessarily improve measurement accuracy but are crucial for capturing transient events. This study underscores the importance of optimizing sampling frequency based on specific monitoring objectives and the need to balance power consumption with data resolution, particularly in remote or battery-based deployments.