Mapping carbon monoxide pollution from space down to city scales with daily global coverage

• Main Authors: T. Borsdorff, J. aan de Brugh, H. Hu, O. Hasekamp, R. Sussmann, M. Rettinger, F. Hase, J. Gross, M. Schneider, O. Garcia, W. Stremme, M. Grutter, D. G. Feist, S. G. Arnold, M. De Mazière, M. Kumar Sha, D. F. Pollard, M. Kiel, C. Roehl, P. O. Wennberg, G. C. Toon, J. Landgraf
• Format: Article
• Language: English
• Published: Copernicus Publications 2018-10-01
• Series: Atmospheric Measurement Techniques
• Online Access: https://www.atmos-meas-tech.net/11/5507/2018/amt-11-5507-2018.pdf

<p>On 13 October 2017, the European Space Agency (ESA) successfully launched the Sentinel-5 Precursor satellite with the Tropospheric Monitoring Instrument (TROPOMI) as its single payload. TROPOMI is the first of ESA's atmospheric composition Sentinel missions, which will provide complete long-term records of atmospheric trace gases for the coming 30 years as a contribution to the European Union's Earth Observing program Copernicus. One of TROPOMI's primary products is atmospheric carbon monoxide (CO). It is observed with daily global coverage and a high spatial resolution of 7×7&thinsp;km². The moderate atmospheric resistance time and the low background concentration leads to localized pollution hotspots of CO and allows the tracking of the atmospheric transport of pollution on regional to global scales. In this contribution, we demonstrate the groundbreaking performance of the TROPOMI CO product, sensing CO enhancements above cities and industrial areas and tracking, with daily coverage, the atmospheric transport of pollution from biomass burning regions. The CO data product is validated with two months of Fourier-transform spectroscopy (FTS) measurements at nine ground-based stations operated by the Total Carbon Column Observing Network (TCCON). We found a good agreement between both datasets with a mean bias of 6&thinsp;ppb (average of individual station biases) for both clear-sky and cloudy TROPOMI CO retrievals. Together with the corresponding standard deviation of the individual station biases of 3.8&thinsp;ppb for clear-sky and 4.0&thinsp;ppb for cloudy sky, it indicates that the CO data product is already well within the mission requirement.</p>