Aeolus, ESA's wind mission, has provided almost five years' worth of global wind observations. The horizontal line-of-sight wind measurements from the world's first space-based Doppler wind lidar have provided positive impact by near-real-time assimilation in numerical weather prediction. To increase Aeolus's impact further, remaining errors should be identified. Currently, measurements are grouped by a scene classification algorithm to use the best wind retrieval algorithm. It is expected that misclassification may negatively influence measurement error and bias. Recently, a feature mask algorithm developed for EarthCARE has been added to the Aeolus data processing chain to improve scene classification and as a pre-launch EarthCARE algorithm test. In this work the current Aeolus clear/cloudy scene classification is compared to the new feature mask. The aim is to improve the classification and grouping of the Aeolus measurements to improve the retrieved winds, and to further test the feature mask algorithm for EarthCARE.
Hemminga et al. (Thu,) studied this question.