Comment on essd-2024-93

作者：Dominique Gantois, Guillaume Payen, Sicard, Michaël, Valentin Duflot, Nicolas Marquestaut, Thierry Portafaix, Sophie Godin‐Beekmann, P. Rodríguez Hernández, Eric Golubic · 年份：2024 · DOI：10.5194/essd-2024-93-rc2 · 研究领域：Radiomics and Machine Learning in Medical Imaging

Abstract. Understanding optical and radiative properties of aerosols and clouds is critical to reduce uncertainties in climate models. For over 10 years, the Observatory of Atmospheric Physics of La Réunion (OPAR) has been operating three active lidar instruments (named Li1200, LiO3S and LiO3T) providing time-series of vertical profiles from 3 to 45 km of the aerosol extinction and backscatter coefficients at 355 and 532 nm, as well as the linear depolarization ratio at 532 nm. This work provides a full technical description of the three systems, details about the methods chosen for the signal preprocessing and processing, and an uncertainty analysis. About 1737 night-time averaged profiles were manually screened to provide cloud-free and artifact-free profiles. Data processing consisted in Klett inversion to retrieve aerosol optical products from preprocessed files. The measurement frequency was lower during the wet season and the holiday periods. There is a good correlation between the Li1200 and LiO3S in terms of stratospheric AOD at 355 nm (0.001–0.107; R = 0.92 ± 0.01), and with the LiO3T in terms of Angström exponent 355/532 (0.079–1.288; R = 0.90 ± 0.13). The lowest values of the averaged uncertainty of the aerosol backscatter coefficient for the three time-series are 64.4 ± 31.6 % for the LiO3S, 50.3 ± 29.0 % for the Li1200, and 69.1 ± 42.7 % for the LiO3T. These relative uncertainties are high for the three ...