The effects of meteorological parameters on PM10: Health impacts assessment using AirQ+ model and prediction by an artificial neural network (ANN)

The estimation of PM10 health effects and air quality forecasting plays an essential role in protecting public health against harmful air pollutants. This study aimed to estimate the impact of meteorological parameters on PM10, evaluate its health impacts, and its prediction using an artificial neural network (ANN) in Yasuj. Demographic, meteorological, and PM10 data were collected from March 2013 to March 2018 in one of Iran's cities. The health impacts of PM10 were estimated using the AirQ+ software. Furthermore, the daily average of PM10 concentrations combined with the meteorological data was used to predict PM10 concentration. The results showed a greater risk of respiratory symptoms for the incidence of asthma symptoms in asthmatic children, the prevalence of bronchitis in children, incidence of chronic bronchitis in adults, and postneonatal infant mortality due to exposure to PM10 with a relative risk of 1.028, 1.08, 1.117, and 1.04 respectively. Also, the best MLP-ANN model predicted PM10 value with correlation coefficient (R2) of 0.87. It can be concluded that decreased PM10 levels were associated with reduced symptoms of bronchitis and asthma in children and bronchitis in adults, and ANN modeling provides a feasible procedure for managerial planning in the view of air pollution. © 2021 Elsevier B.V