Regulations to limit the spread of the COVID-19 pandemic led to substantial changes in human life, industrial productivity, and mobility, which caused reductions in emissions from industry and ground and airborne transportation (Venter et al. 2020). Hence, the lockdown period offered the unique opportunity to directly measure the effects of reduced pollution emissions on atmospheric composition and thereby challenge our understanding of the anthropogenically perturbed chemical and physical environment (Kroll et al. 2020). The different starting times and types of regulations from the national governments as well as the different chemical and physical processing and hence lifetimes of the emissions caused regionally different evolutions of atmospheric concentrations of individual species. This leads to major uncertainties in the quantitative estimate of emission changes needed to establish emission inventories (Forster et al. 2020). For example, there have been efforts to derive trends in ground transportation in many countries from searches in web-based map platforms (Le Quéré et al. 2020). Guevara et al. (2021) estimated the reduction in primary emissions from different source sectors, such as energy and manufacturing industries and traffic sectors, based on publicly available data [with further data in Copernicus (2020) and Guevara et al. (2020)]. Extrapolating previous emissions to a 2020 business as usual scenario, they derive an average 33% emission reduction of nitrogen oxides (NOx) in Europe and similar but less reduction for other pollutants (Fig. 1). The transport and the industry sectors were affected differently from lockdown restrictions, with the highest per-sector emission reductions in aviation, whereas the highest total reduction was attributable to road transport (up to about 70% of all sectors, depending on pollutant). Le et al. (2020) compared satellite observations of particulate matter (PM2.5), nitrogen dioxide (NO2), sulfur dioxide (SO2), and ozone (O3) during the lockdown in China to prelockdown observations. While reductions in PM2.5, NO2, and SO2 agree with the trend expected from reduced emissions from transport and industry, enhancements in PM2.5 in the Beijing area could only be explained by taking the appropriate meteorology into account. The impact of meteorology, pollution, and other factors on the boundary layer composition and on PM2.5 is discussed in many studies (e.g., Chen et al. 2020; Dhaka et al. 2020; Hallar et al. 2021; Karle et al. 2021; Solimini et al. 2021). These studies emphasize the need for comprehensive atmospheric composition measurements from the boundary layer to the stratosphere in different parts of the world in order to determine and better understand atmospheric composition changes caused by human activities and distinguish the anthropogenic impact from natural factors.
Travel restrictions resulted in more than 80% reductions in air traffic worldwide during the early lockdown phase (Guevara et al. 2021), and aviation experienced significantly stronger reductions compared to other transport sectors. While air traffic generally recovered within a year in Asia and America, European air traffic lagged behind and showed a significant decrease throughout summer 2020 with a slight recovery toward the end of the year (see Fig. 2). One year after the initial lockdown, European air traffic was still reduced by about 30% compared to pre-COVID-19 levels.
Aircraft emit carbon dioxide (CO2), nitrogen oxides (NOx), water vapor (H2O), and aerosols. The aircraft emissions can modify cirrus clouds and lead to the formation of contrails in cold and humid areas at cruise altitudes (Lee et al. 2010). The recent, comprehensive assessment of air traffic effects on the atmosphere (Lee et al. 2021) shows that aviation up to 2018 contributed about 3.5% to the total anthropogenic effective radiative forcing, with about one-third coming from its CO2 emissions and two-thirds resulting from the non-CO2 effects. In fact, the major contributor to effective radiative forcing from aviation is caused by contrail cirrus (57%) (Burkhardt et al. 2018; Lee et al. 2021). The effects of aviation generated aerosol on natural clouds remains uncertain (Lee et al. 2021). Properties of contrail cirrus have been measured from aircraft (Heymsfield et al. 2010; Voigt et al. 2010, 2017, 2021; Schumann et al. 2017; Bräuer et al. 2021a,b) and satellites (Minnis et al. 2013; Vázquez-Navarro et al. 2015). These observations were used to evaluate the Contrail Cirrus Prediction model (COCIP) (Schumann 2012; Schumann et al. 2017), with which the radiative forcing from contrail cirrus can be calculated. Reduced contrail cirrus optical thickness and radiative forcing caused by diminished air traffic over Europe has been calculated for a 9-month period in 2020 (Schumann et al. 2021a). Significant reductions in seasonal and regional effective radiative forcing from contrail cirrus caused by reduced air traffic emissions during the lockdown 2020 were also found by Gettelman et al. (2021), while the annual mean effective radiative forcing was less affected. Also, the difficulty in separating the impact of meteorology from the reduced aircraft emissions impact on cloud properties induces considerable uncertainties in the calculations (Schumann et al. 2021b). Quaas et al. (2021) used satellite-based cloud retrievals in regions more and less affected by air traffic to derive changes in the aviation impact on cirrus clouds in 2020. Recent studies (Urbanek et al. 2018; Li and Groß 2021) also suggest changes in optical cirrus properties (extinction, depolarization) caused by aged air traffic emissions. Still, many research questions remain with respect to the derivation of changes in cirrus properties and radiative forcing for reduced air traffic in 2020.
Combined, these challenges motivated the BLUESKY mission (www.dlr.de/content/en/articles/news/2020/02/20200522_bluesky-examines-the-atmosphere-during-the-coronavirus-lockdown.html), with the objective to advance our understanding of the anthropogenic impact on atmospheric composition by acquiring a unique dataset on trace gas, aerosol, and cloud properties measured in the early lockdown phase over central Europe and the northern Atlantic flight corridor. Two aircraft were equipped for the BLUESKY campaign: the High-Altitude and Long-Range Research Aircraft (HALO), a Gulfstream 550 with a range of about 8,000 km and 14.5 km cruise altitude (Tadic et al. 2021; Voigt et al. 2017), and the DLR Falcon with 3,000 km range and up to 12 km cruise altitude (Voigt et al. 2011). For BLUESKY, the aircraft were equipped with instruments that measure long- and short-lived trace gases, aerosol, and cloud properties. The aircraft measurements were combined with satellite data of tropospheric NO2 column densities from the Global Ozone Monitoring Experiment 2 (GOME-2; Munro et al. 2016) and Sentinel-5P/Tropospheric Monitoring Instrument (TROPOMI; Veefkind et al. 2012), and with cloud data retrieved from the SEVIRI Imager on the Meteosat Second Generation (MSG) satellite, as well as the Cloud–Aerosol Lidar with Orthogonal Polarization (CALIOP) on the Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) (Winker et al. 2010). The measurements were also used to evaluate the global chemistry climate model EMAC (Jöckel et al. 2010) and the global–regional chemistry–climate model MECO(n) (Kerkweg and Jöckel 2012a,b; Mertens et al. 2016).
Due to travel restrictions, BLUESKY was conducted out of the home base in Oberpfaffenhofen (48°5′N, 11°17′E) in southern Germany. The observations were performed in May and June 2020 during the early COVID-19 lockdown phase in Europe. This paper describes the aircraft instrumentation and the campaign strategy supported by satellite and modeling activities, as well as first results and highlights from the BLUESKY mission.