Comparative statistical models for estimating potential roles of relative humidity and temperature on the concentrations of secondary inorganic aerosol: Statistical insights on air pollution episodes at Beijing during January 2013

Heavy air pollution attacked Beijing and its surrounding areas in January of 2013, which attracted large attentions from around the world. In this study, we conducted highly time-resolved measurements of inorganic ions associated with PM2.5 at an urban site in Beijing during this period. We applied the curve fitting method, the quantile regression model and the probability function model to evaluate the relationship between secondary inorganic ions (sulfate, nitrate and ammonium, SNA) and meteorological parameters (relative humidity (RH) and temperature). In our model results, RH was regarded as one of the key factors of high concentrations of SNA, and high level of RH would enhance the concentrations of SNA. In addition, the effect of temperature was also important and noticeable. We further constructed a probability function model to investigate the joint effects of RH and temperature. The model results showed higher RH and a temperature of approximately -5 degrees C (-4 similar to-6 degrees C for sulfate and nitrate; -5 similar to-7 degrees C for ammonium) would be the most suitable conditions for the high concentrations of SNA during the air pollution episodes in Beijing.