Characterization of Cirrus Clouds in the Mid-Latitude Tropopause Region

Cirrus clouds contribute to the general radiation budget of the Earth, playing an important role in climate projections. Of special interest are optically thin cirrus clouds close to the tropopause due to the fact that they are difficult to capture and thus their impact is not yet well understood. This study presents a characterization of cirrus clouds observed by the limb sounder GLORIA (Gimballed Limb Observer for Radiance Imaging of the Atmosphere) aboard the German research aircraft HALO during the WISE (Wave-driven ISentropic Exchange) campaign in September/October 2017. This campaign took place in Shannon, Ireland (52.70°N, 8.86°W). We developed an optimized cloud detection method and derived macro-physical characteristics of the detected cirrus clouds: cloud top height, cloud bottom height, vertical extent and cloud top position with respect to the tropopause. The fraction of cirrus clouds detected above the tropopause (> 0 km) is in the order of 13% to 27%, depending on the detection method and the definition of the tropopause. In general, good agreement with the clouds predicted by the ERA5 reanalysis dataset is obtained. However, cloud occurrence is ≈50% higher in the observations for the region close to and above the tropopause. Cloud bottom heights are also detected above the tropopause. Considering the uncertainties for the tropopause height, cloud top height and cloud bottom height determination we could not find unambiguous evidence for the formation of cirrus layers above the tropopause. In addition, for a better understanding of the tropopause cirrus properties and life conditions, two cirrus cases observed during two scientific flights were selected from the observations and compared with cirrus simulations performed with the 3D Lagrangian microphysical model CLaMS-Ice, which is based on the two-moment bulk cirrus model by Spichtinger and Gierens (2009) (doi: 10.5194/acp-9-685-2009). The model is fed by backward trajectories computed from high resolution ERA5 data (hourly, spatial grid 30 km). This contribution summarizes and extends on work described by Bartolome Garcia et al. (2020) (doi:10.5194/amt-2020-394).