Revealing the sulfur dioxide emission reductions in China by assimilating surface observations in WRF-Chem

Abstract

The anthropogenic emission of sulfur dioxide (SO2) over China has significantly declined as a consequence of the clean air actions. In this study, we have developed a new emission inversion system based on a four-dimensional local ensemble transform Kalman filter (4D-LETKF) and the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem) to dynamically update the SO2 emission grid by grid over China by assimilating the ground-based hourly SO2 observations. Sensitivity tests for the assimilation system have been conducted firstly to tune four system parameters: ensemble size, horizontal and temporal localization lengths, and perturbation size. Our results reveal that the same random perturbation factors used throughout the whole model grids with assimilating observations within about 180 km can efficiently optimize the SO2 emission, whereas the ensemble size has only little effect. The temporal localization by assimilating only the subsequent hourly observations can reveal the diurnal variation of the SO2 emission, which is better than updating the magnitude of SO2 emission every 12 h by assimilating all the observations within the 12 h window. The inverted SO2 emission over China in November 2016 has declined by an average of 49.4 % since 2010, which is well in agreement with the bottom-up estimation of 48.0 %. Larger reductions of SO2 emission are found over the a priori higher source regions such as the Yangtze River Delta (YRD). The simulated SO2 surface mass concentrations using two distinguished chemical reaction mechanisms are both much more comparable to the observations with the newly inverted SO2 emission than those with the a priori emission. These indicate that the newly developed emission inversion system can efficiently update the SO2 emissions based on the routine surface SO2 observations. The reduced SO2 emission induces the sulfate and PM2.5 surface concentrations to decrease by up to 10 µg m−3 over central China.