Investigating the response of East Asian ozone to Chinese emission changes using a linear approach

To illuminate the issue of trans-boundary O-3 pollution and regional O-3 reduction policies in East Asia, we have investigated the East Asian ozone (O-3) response to perturbations caused by Chinese anthropogenic emissions using the Community Multiscale Air Quality (CMAQ) model, a regional chemical transport model. The O-3 responses have been examined for the range between -100 and +100% changes from the Chinese emissions level in 2004 in 10% intervals. We have found that springtime and summertime O-3 responses both at the source and at the downwind areas can be regarded as linear over the range between -30 and +100% changes from the current emissions level. We therefore suggest that the perturbation between -30 and +100% is sufficiently small to avoid nonlinear chemical influence on O-3 formation in a model experiment to investigate East Asian scale O-3 source-receptor relationships. On the other hand, the O-3 response is strongly nonlinear in April at Hong Kong, where the current NMVOCs/NOx ratio is low and the O-3 production regime is easily moved to the NMVOCs sensitive region. The O-3 responses to the NOx emission changes have been investigated using surface O-3 concentrations at remote Japanese sites and tropospheric NO2 vertical column density (NO2 VCD) over central east China both with observations and with model simulations in springtime during 2003-2009. Analysis of satellite data shows that the observed range of NO2 VCD over central east China in 2003-2009 is the range between 25 and +34% from the 2004 level, which corresponds approximately to an emission variation between 21 and +29%. The O-3 concentration in the downwind region during 2003-2009 responds linearly to a change of the NO2 VCD over central east China both in the model and in the observation. The corresponding O-3 responses derived from surface observations at remote Japanese sites show linear features consistent with this expectation. The doubling of emissions, i.e., approximately 1.9-fold increase in the NO2 VCD from 2004, leads to O-3 increments of 5 ppbv and 8 ppbv in the model and in the observation, respectively. The modeled O-3 increase due to changes in NOx emission explains approximately 60% of the observed O-3 trend at remote Japanese sites. Thus, approximately 40% of the observed O-3 increase is unaccounted for by the NOx emissions growth. (C) 2012 Elsevier Ltd. All rights reserved.