Improved Inversion of Monthly Ammonia Emissions in China Based on the Chinese Ammonia Monitoring Network and Ensemble Kalman Filter

• DOI: 10.1021/acs.est.9b02701
• Publication Date: 2019-10-02T09:12:34Z
• Authors: Lei Kong; Xiao Tang; Jiang Zhu; Zifa Wang; Yuepeng Pan; Huangjian Wu; Lin Wu; Qizhong Wu; Yuexin He; Shili Tian; Yuzhu Xie; Zirui Liu; Wenxuan Sui; Lina Han; Greg Carmichael
• Keywords: Air Pollutants; China; Ammonia; 13. Climate action; Air Pollution; 11. Sustainability; 01 natural sciences; 6. Clean water; Environmental Monitoring; 0105 earth and related environmental sciences

<p>Ammonia (NH<sub>3</sub>) emission inventories are an essential input in chemical transport models and are helpful for policy-makers to refine mitigation strategies. However, current estimates of Chinese NH<sub>3</sub> emissions still have large uncertainties. In this study, an improved inversion estimation of NH<sub>3</sub> emissions in China has been made using an ensemble Kalman filter and the Nested Air Quality Prediction Modeling System. By first assimilating the surface NH<sub>3</sub> observations from the Ammonia Monitoring Network in China at a high resolution of 15 km, our inversion results have provided new insights into the spatial and temporal patterns of Chinese NH<sub>3</sub> emissions. More enhanced NH<sub>3</sub> emission hotspots, likely associated with industrial or agricultural sources, were captured in northwest China, where the a posteriori NH<sub>3</sub> emissions were more than twice the a priori emissions. Monthly variations of NH<sub>3</sub> emissions were optimized in different regions of China and exhibited a more distinct seasonality, with the emissions in summer being twice those in winter. The inversion results were well-validated by several independent datasets that traced gaseous NH<sub>3</sub> and related atmospheric processes. These findings highlighted that the improved inversion estimation can be used to advance our understanding of NH<sub>3</sub> emissions in China and their environmental impacts.</p>