Atmospheric composition is affected by emissions from multiple sources at different spatial and temporal scales. Traditional methods for quantifying these emissions rely on surface-based survey or observation data, and the results are often called “bottom-up” estimates. In contrast, in part driven by the routine measurements of atmospheric states and surface properties from space, satellite data is now increasingly used to constrain emissions; these so-called “top-down” estimates have several advantages when it comes to spatial coverage and timely updates.
In this talk, I will present the progress that my team (in collaboration with others) has made in top-down estimates of emissions from human activities, soils, and fires. I will show that multi-sensor data can provide coherent insights on the change in emissions and can be used as strong constraints to improve the forecasting of atmospheric composition. I will end the talk by highlighting recent developments of geostationary satellite constellation for atmospheric composition monitoring and the new opportunities to characterize the vertical distribution of aerosols from passive remote sensing techniques.