Understanding the origins of urban particulate matter pollution based on high-density vehicle-based sensor monitoring and big data analysis

Abstract

This study presents an innovative method for air quality monitoring and identifying pollution sources in Rizhao, a coastal city in northern China, by deploying a network of low-cost sensors mounted on 102 taxis. Over a one-year period, we collected a set of high-resolution PM₁₀ and PM_2.5 data. Using big data analysis, including downwind-calm wind analysis, hotspot detection, and time-series clustering analysis, we traced the pollution back to the urban origins of pollutant. Our extensive study uncovered significant spatial and seasonal variations in PM₁₀ and PM_2.5 concentrations, pinpointing substantial PM₁₀ emissions from steel plants and a notable influence of industrial activities on ambient PM_2.5 concentrations. Through the application of bivariate polar plots and hotspot mapping, we linked major particulate matter sources to industrial activities especially steel plant emissions, and road traffic, which significantly elevated the particulate matter levels in residential and industrial zones. Our time-series clustering analysis further distinguishes traffic and industrial activities as key contributors to particulate matter pollution. This study advances the application of low-cost sensor technologies in urban air quality management and offers a detailed insight into the pollution sources and their diverse impacts on particulate matter levels in urban settings.