[A-AS07] Atmospheric Chemistry

[AAS07-14] Quantifying thermal power plant CO₂ emissions from CO₂ total column measurements with portable FTIR spectrometers

*Hirofumi Ohyama¹, Kei Shiomi², Nobuhiro Kikuchi², Isamu Morino¹, Akihiro Hori¹, Tsuneo Matsunaga¹ (1.National Institute for Environmental Studies, 2.Japan Aerospace Exploration Agency)

Keywords:carbon dioxide, thermal power plant, emission amount, remote sensing

Emissions of carbon dioxide (CO₂) from large point sources such as thermal power plants and steel plants account for approximately 40% of the world’s energy-related CO₂ emissions. Nowadays, the CO₂ emissions from thermal power plants have been quantified based on CO₂ column abundances obtained by airborne and satellite-borne remote sensing. In the present study, we evaluated precisions in CO₂ emission estimates from simpler and more inexpensive ground-based CO₂ column measurements. To this end, we conducted two measurement campaigns for one month each during October 2018 and October−November 2019 to measure CO₂ column abundances using two portable Fourier transform infrared (FTIR) spectrometers (EM27/SUN) around a thermal power plant in Japan. In addition, wind profiles in the planetary boundary layer and meteorological parameters at the ground were observed by a Doppler lidar and a meteorological observation system, respectively. We assumed that the CO₂ plume rises by a certain height depending on heat emission rate and wind speed at the stack top, and that the horizontal distribution of CO₂ column enhancement relative to the background can be represented by a Gaussian plume model. Because the shape of the plume is disturbed by turbulence, the observed CO₂ column and meteorological data were averaged into 10 min bins to mitigate that effect. The wind speeds at an effective plume height and the plume widths necessary for CO₂ emission estimates were obtained or estimated from the observed values, and highly variable wind directions were retrieved simultaneously with the CO₂ emission. We compared the CO₂ emissions estimated from the observed data with those converted from hourly fuel consumptions, and it was found that the standard deviation (1σ) of the differences was ~20%.