Modeling PM2.5 Urban Pollution Using Machine Learning and Selected Meteorological Parameters

TLDR: A machine learning approach based on six years of meteorological and pollution data analyses is proposed to predict the concentrations of PM2.5 from wind (speed and direction) and precipitation levels and demonstrates that the use of statistical models based on machine learning is relevant to predict PM 2.5 concentrations from meteorological data.

Abstract: Outdoor air pollution costs millions of premature deaths annually, mostly due to anthropogenic fine particulate matter (or PM2.5). Quito, the capital city of Ecuador, is no exception in exceeding the healthy levels of pollution. In addition to the impact of urbanization, motorization, and rapid population growth, particulate pollution is modulated by meteorological factors and geophysical characteristics, which complicate the implementation of the most advanced models of weather forecast. Thus, this paper proposes a machine learning approach based on six years of meteorological and pollution data analyses to predict the concentrations of PM2.5 from wind (speed and direction) and precipitation levels. The results of the classification model show a high reliability in the classification of low ( 25 µg/m3) and low (<10 µg/m3) versus moderate (10–25 µg/m3) concentrations of PM2.5. A regression analysis suggests a better prediction of PM2.5 when the climatic conditions are getting more extreme (strong winds or high levels of precipitation). The high correlation between estimated and real data for a time series analysis during the wet season confirms this finding. The study demonstrates that the use of statistical models based on machine learning is relevant to predict PM2.5 concentrations from meteorological data.