A new lidar design for operational atmospheric wind and cloud/aerosol survey from space
Résumé
Global wind profile measurement has for long been a first priority for numerical weather prediction. The demonstration from ground-based observations that a double-edge Fabry-Perot interferometer could be efficiently used for deriving wind profiles from the molecular scattered signal in a very large atmospheric vertical domain has led to the choice of the direct detection technique in space and the selection of the Atmospheric Dynamic Mission (ADM) Aeolus by ESA in 1999. ADM-Aeolus was successfully launched in 2018, after the technical issues raised for the lidar development have been solved, providing first global wind profiles from space in the whole troposphere. Simulated and real time assimilation of the projected horizontal wind information were able to confirm the expected improvements in forecast score, validating the concept of a wind profiler using a fixed line-of-sight lidar from space. The question is raised here about consolidating results gained from ADM-Aeolus mission with a potential operational follow-on instrument. Maintaining the configuration of the instrument as close as possible to the one achieved (UV emission lidar with a single slanted line-of-sight) we revisit the concept of the receiver by replacing the arrangement of the Fizeau and Fabry-Perot interferometers with a unique Quadri-channel Mach-Zehnder (QMZ) interferometer which relaxes the system operational constraints and extends the observation capabilities to recover the radiative properties of clouds. This ability is meeting first and second profiling priorities of the meteorological forecasting community on atmospheric dynamics and radiation. We discuss the optimization of the key parameters that may preside to the selection of a high performance system. The selected optical path difference (3.2 cm) of the QMZ leads to a very compact design allowing the realization of a high quality interferometer and offering a large field-angle acceptance. Performance simulation of horizontal wind speed measurements with different backscatter profiles shows results in agreement with the targeted ADM-Aeolus random errors, using an optimal 45° line-of-sight angle. The Doppler measurement is, from principle, unbiased by the atmospheric conditions (temperature, pressure, particle scattering) and only weakly affected by the instrument calibration errors. The study of the random systematic errors arising from the uncertainties in the instrumental calibration and in the modelled atmospheric parameters used for the backscatter analysis shows a limited impact under realistic conditions. The particle backscatter coefficients can be retrieved with uncertainties better than a few percent in the boundary layer and in semi-transparent clouds. Extinction coefficients and depolarization ratio can be derived accordingly.
Origine : Fichiers éditeurs autorisés sur une archive ouverte
