Abstract：

Greenhouse gases and smoke aerosol particle emissions from fossil fuel combustion and biomass burning significantly affect several key aspects of the atmospheric environment and climate system (e.g., atmospheric constituents, air quality, optical properties, and radiative forcings from greenhouse gases and aerosols). However, large uncertainty of emissions still exists in the common bottom-up estimates, due to data gaps, a lack of knowledge in fuel statistics, and inaccurate emission factors. Atmospheric observations from satellites and other platforms offer a unique opportunity to objectively quantify the magnitude and distribution of emissions from virous processes.

In this talk, I will discuss the application of inverse modeling technique to quantify ffCO₂ and wildfire emissions by assimilating atmospheric observations. First, a Bayesian inversion framework is developed to constrain the fossil fuel CO₂ (ffCO₂) emissions from urban areas by utilizing the Orbiting Carbon Observatory 2 (OCO-2) total column CO₂ retrievals. The contribution of error components (model transport, measurement) and number of retrievals to the uncertainty of emission estimates are explicitly evaluated, with implications for future carbon monitoring missions. Next, I will discuss the application of inverse modeling to a forecasting system of wildfire smoke for the western U.S., which provides near-real-time improvements of smoke emissions based on satellite data. The performance of the forecasts is intercompared with other eleven state-of-the-art forecasting systems, which suggests pathways for the future development of smoke simulation and forecasts. Lastly, I will present our ongoing research of assessing the vertical allocation of smoke emissions by utilizing inverse modeling and airborne lidar observations.

个人简介：

叶鑫欣博士于2015年7月获得北京大学大气物理学与大气环境博士学位。2015年10月起先后在美国宾州州立大学和加州大学洛杉矶分校开展博士后研究。2020年10月起在加州大学洛杉矶分校担任助理研究员。研究兴趣为大气污染物传输与扩散，碳循环，空气质量模拟和预报。

Conference ID (For Tencent)：https://meeting.tencent.com/dm/HijcK4pUegK5

Tencent (腾讯会议) Link：620 549 777