CALIPSO (IIR–CALIOP) retrievals of cirrus cloud ice-particle concentrations

International audienceA new satellite remote sensing method is described whereby the sensitivity of thermal infrared wave resonance absorption to small ice crystals is exploited to estimate cirrus cloud ice particle number concentration N, effective diameter De, and ice water content IWC. This method uses co-located observations from the Infrared Imaging Radiometer (IIR) and from the CALIOP (Cloud and Aerosol Lidar with Orthogonal Polarization) lidar aboard the CALIPSO (Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation) polar orbiting satellite, employing IIR channels at 10.6 μm and 12.05 μm. Using particle size distributions measured over several flights of the TC4 (Tropical Composition, Cloud and Climate Coupling) and the mid-latitudes SPARTICUS (Small Particles in Cirrus) field campaigns, we show for the first time that N/IWC is tightly related to βeft; the ratio of effective absorption optical depths at 12.05 μm and 10.6 μm. Relationships developed from in situ aircraft measurements are applied to βeft derived from IIR measurements to retrieve N. This satellite remote sensing method is constrained by measurements of βeft from the IIR and is by essence sensitive to the smallest ice crystals. Retrieval uncertainties are discussed, including uncertainties related to in situ measurement of small ice crystals (D eft. The method is applied here to single-layered semi-transparent clouds having a visible optical depth between about 0.3 and 3, where cloud base temperature is ≤ 235 K. Two years of CALIPSO data have been analyzed for the years 2008 and 2013, with the dependence of cirrus cloud N and De on altitude, temperature, latitude, season (winter vs. summer) and topography (land vs. ocean) described. The results for the mid-latitudes show a considerable dependence on season. In the high latitudes, N tends to be highest and De smallest, whereas the opposite is true for the tropics. The frequency of occurrence of these relatively thick cirrus clouds exhibited a strong seasonal dependence in the high latitudes, with the occurrence frequency during Arctic winter being at least twice that of any other season. Processes that could potentially explain some of these micro- and macroscopic cloud phenomena are discussed