Day 4

Detailed paper information

Back to list

Paper title Analyzing local carbon dioxide and nitrogen oxide emissions from space using statistical methods: An application to the synthetic SMARTCARB dataset
Authors
  1. Janne Hakkarainen Finnish Meteorological Institute (FMI) Speaker
  2. Iolanda Ialongo Finnish Meteorological Institute
  3. Monika Szeląg Finnish Meteorological Institute
  4. Johanna Tamminen FMI
  5. Erik Koene Empa - Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland
  6. Gerrit Kuhlmann Empa - Swiss Federal Laboratories for Materials Science and Technology
  7. Dominik Brunner Empa - Swiss Federal Laboratories for Materials Science and Technology
Form of presentation Poster
Topics
  • A1. Atmosphere
    • A1.04 Greenhouse Gases
Abstract text Using satellite data for estimating carbon dioxide (CO2) emissions from anthropogenic sources has become increasingly important since the Paris Agreement was adopted in 2015, due to their global coverage. The very first study that estimated CO2 emissions from individual power plants using satellite data was published in 2017 (Nassar et al., 2017). In recent years, the literature has been rapidly expanding with several new approaches and case studies. Many of the proposed techniques for estimating CO2 emissions from local sources are based on single satellite overpasses (e.g., Varon et al., 2018). To estimate nitrogen oxide (NOx) emissions from averaged NO2 columns, statistical methods (i.e., based on multiple spatially co-located observations) are often applied. In Europe, one of the key activities to respond to Paris Agreement’s goal to monitor anthropogenic CO2, is the Copernicus Carbon Dioxide Monitoring mission (CO2M).

In this work, we discuss the use of statistical methods for estimating the CO2 emissions. The advantage of the statistical methods is that they do not require complex atmospheric modeling, and they generally provide more robust emission estimates compared to individual satellite overpasses. In addition, these methods have been successfully applied to instruments and locations, where the individual plumes are not detectable, but the emission signal becomes visible when multiple scenes are averaged. In particular, we use divergence method, developed originally for NO2 by Beirle et al. (2019), to estimate CO2 emissions, from the synthetic SMARTCARB dataset (Kuhlmann et al., 2020) that has been created in order to prepare for the upcoming CO2M mission. We analyze the effect of different denoising techniques to the CO2 emission estimates. In addition, we estimate source specific NOx-to-CO2 emission ratio and discuss converting the estimated NOx emissions to CO2 emissions.