AirCore Carbon Monoxide Profiles for MOPITT and TROPOMI Validation

Abstract:

AirCore is an atmospheric sampling system consisting of a long coiled stainless steel tube that can be accommodated in any airborne sampling platform (Karion et al., 2010; Membrive et al., 2017). The tube, initially filled with a known standard gas, empties during ascent and samples ambient air during descent. Mixing ratios of the gases in the tube are later analyzed in the laboratory. Balloon-borne AirCore profiles routinely reach high altitudes (~ 30 km) that are not often represented in aircraft-based in situ datasets.

Here we use balloon-borne AirCore CO profiles to validate satellite CO retrievals from MOPITT (the Measurements Of Pollution In The Troposphere instrument, on board NASA’s Terra) and TROPOMI (the TROPOspheric Monitoring Instrument, on board ESA’s S5P). The main goals are to characterize the effect of clouds on TROPOMI land retrievals and to augment the array of airborne profiles systematically utilized in MOPITT validation.

MOPITT has been measuring CO profiles and total columns since 2000, producing the longest global CO record to date. Global coverage is achieved every three days with a 22 x 22 km² footprint. MOPITT derives CO from cloud-free reflected- and thermal-infrared radiances. TROPOMI has been measuring total CO column since late 2017 with daily global coverage at about 7 x 7 km² spatial resolution or better. TROPOMI derives CO from reflected-infrared radiances over land (under clear and cloudy conditions) and water (cloudy only). Cloudy TROPOMI retrievals are achieved by estimating the cloud top altitude and approximating the partial CO column underneath with reference profiles (Landgraf et al., 2016). A previous comparison of clear MOPITT and TROPOMI CO retrievals showed good agreement between the two datasets (Martínez-Alonso et al., 2020). Cloudy TROPOMI CO water retrievals were validated respect to ATom (Atmospheric Tomography mission) airborne profiles in the same study; there it was shown that the use of reference profiles below cloud top results in very small errors. Since there are no major sources over water, CO values closer to the surface (most likely to be below cloud top) are well characterized by the reference profiles. Larger errors may occur in cloudy TROPOMI land retrievals, particularly near CO emission sources. We characterize these errors using AirCore profiles.