Dissimilar effects of two El Nino types on PM( 2.5 )concentrations in East Asia

We investigate the effects of natural variability of meteorological fields on surface PM(2.5 )concentration changes in East Asia during El Nino periods for the past three decades (1980-2014) through GEOS-Chem 3D global chemical transport model simulations. First, our evaluation of the model with anthropogenic emissions for 2006 and a comparison against observations show that the simulated results accurately reproduced the observed spatial distribution of annual mean aerosol concentrations for 2006-2007 including inorganic (sulfate, ammonium, and nitrate) and carbonaceous (organic and black carbon) aerosols in the surface air. Based on the Oceanic Nino Index, the assimilated meteorological data used in the model simulations indicate that 10 El Nino events occurred for the past three decades (1980-2014). We further classified the 10 El Nino events into 6 central Pacific El Nino (C-type) and 4 eastern Pacific El Nino (E-type) to examine the different roles of two El Nino types in determining seasonal surface PM2.5 concentrations in East Asia. We find opposite impacts on the seasonal surface PM2.5 concentrations depending on two El Nino types, such that the surface PM2.5 concentrations during the E-type period are higher than the climatological mean value, especially in northern East Asia. The peak increase of as much as 20% occurs in winter and is sustained until the following spring. However, the C-type period shows a decrease in seasonal PM(2.5 )concentrations in northern East Asia compare to the climatological mean, and the peak decrease of as much as 10% occurs in the following spring. The different of two El Nino types also have dissimilar impacts on surface PM2.5 concentrations in southeastern China. Natural variation of aerosol concentrations driven by the different of two El Nino types appears to be significant and would be an important factor in determining the inter-annual variation of aerosol concentrations in East Asia. (C) 2018 Elsevier Ltd. All rights reserved.