A modeling study of the peroxyacetyl nitrate (PAN) during a wintertime haze event in Beijing, China.

Abstract

Peroxyacetyl nitrate (PAN), as the second most important photochemical pollutant, could threaten human health and terrestrial ecosystem. Observed high PAN concentrations in wintertime in Beijing are comparable to those in summertime when photochemistry is strong. In this study, we investigated the variations of concentrations of PAN and other related chemical species during a haze event on February 13-16, 2017 in Beijing using a combination of in-situ observations and the WRF-Chem model simulation. During this episode, the observed mean (maximum) PAN concentrations at the urban and rural sites were 1.9 (4.7) ppb and 1.8 (4.3) ppb, respectively. Model evaluation showed that the simulation with CBM-Z gas-phase scheme performed much better than simulation using the RADM2 scheme, which mainly results from the accurate representation of the reaction rates for PAN formation through CH3C(O)O2 radicals and NO2 in the CBM-Z scheme. But simulated PAN concentrations in Beijing were still underpredicted using the CBM-Z scheme, likely due to the underestimation of VOC emissions. Since the lifetime of PAN becomes as long as several days in wintertime due to lower temperature, it could be conducive to be accumulated and transported. Sensitivity simulation demonstrated that local sources accounted for about 15%-30% of mean PAN concentrations over most of Beijing during the event, lower than that of PM2.5 (25%-60%). These suggest that during Beijing wintertime polluted days, increased PAN concentration, which is dominantly contributed by regional transport from surrounding areas, will cause more health concerns, and its control strategies need larger regional efforts than those conducted for PM2.5.