Development and evaluation of multisensor methods for EarthCare mission based on A-Train and airborne measurements

Kurzfassung

The impact of ice clouds on the water cycle and radiative budget is still uncertain due to the complexity of cloud processes that makes it difficult to acquire adequate observations of ice cloud properties and parameterize them into General Circulation Models. Passive and active remote sensing instruments, radiometers, radars and lidars, are commonly used to study ice clouds. Inferring cloud microphysical properties (extinction, ice water content, effective radius, ...) can be done from one instrument only, or from the synergy of several. The interest of using instrumental synergies to retrieve cloud properties is that it can reduce the uncertainties due to the shortcomings of the different instruments taken separately. The A-Train constellation of satellites has considerably improved our knowledge of clouds. Since 2006, the 532nm backscattering lidar CALIOP on board the satellite CALIPSO and the 94GHz cloud radar CPR on board the satellite CloudSat have acquired cloud vertical profiles globally and many lidar-radar synergetic methods have been adapted to CloudSat and CALIPSO data. In 2021 will be launched a new satellite, EarthCARE, boarding state of the art remote sensing instrumentation, in particular ATLID, a High Spectral Resolution Lidar (HSRL) at 355nm and a Doppler cloud radar at 94 GHz. The main mission of this satellite is to quantify interactions between clouds, aerosols and the Earth's radiation budget in order to improve weather prediction and climate models. Thanks to its advanced instrumentation mounted on a single platform, this new mission is expected to provide unprecedented observations of clouds from space. However, to do so, the synergistic algorithms that were developed for A-Train measurements have to be adapted to this new instrumental configuration. During my PhD, I focused on the Varcloud algorithm that was developed in 2007 by Delanoë and Hogan, based on a variational technique. The first part of the work consisted in adapting some parameters of the microphysical model of the algorithm to recent studies of a large dataset of in-situ measurements. In particular, the questions of a parameterization of the lidar extinction-to-backscatter ratio and the choice of the mass-size relationship for ice crystals were addressed. The second part of my work consisted in adapting the Varcloud retrieval algorithm to airborne platforms. Airborne platforms are ideal to prepare and validate space missions, allowing for direct underpasses of spaceborne instruments. Moreover, German and French aircraft, respectively HALO and French Falcon 20 have very complementary payloads and are perfectly designed for the preparation, the calibration and the validation of EarthCare. Both aircraft board a high spectral resolution lidar (355 nm on the French Falcon and 532 nm on the HALO) and a Doppler radar at 36 GHz (HALO) and 95 GHz (Falcon). In fall 2016 a field campaign related to the NAWDEX project took place in Iceland, Keflavik with both aircraft involved. The measurements collected during this campaign provide an interesting dataset to characterize cloud microphysics and dynamics in the North Atlantic, which are of high interest regarding the Cloudsat-CALIPSO and EarthCARE missions. In addition, a series of common legs with the same cloud scene observed by both platforms were performed, providing data to study the influence of the instrumental configuration on the retrieved ice cloud properties.