A Distributed Low-Cost Pollution Monitoring Platform

Abstract

Personal exposure to heightened levels of fine airborne particulate matter (PM) has been linked to numerous adverse health effects in sensitive groups. However, researchers investigating these correlations are struggling to find the spatiotemporal datasets that are sufficient for study. Current airborne PM monitoring solutions are highly accurate, but expensive. Therefore, they are not feasible candidates for spatially dense deployments, and cannot be used to analyze the effects of exposure to pollution microclimates. In this article, we present a low-cost pollution monitoring station that operates as a single node in a wireless network. Each node periodically collects airborne pollution and supporting meteorological data and uploads measurements to a central, open-source database. A total of 50 nodes were deployed across a large metropolitan area (roughly 100 km <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ) over a six-month campaign. The experimental results show good correlation (R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> = 0.88) between devices co-located with the federal equivalent methods, which have an accuracy traceable to the National Institute of Standards and Technology. By applying linear corrections derived from in situ field measurements to each PM sensor, we were able to demonstrate a 1.8× decrease in root-mean-squared error over the raw measurements.