Satellite–Ground Intercomparison between EarthCARE/ATLID and the China Aerosol Raman Lidar Network (CARLNET): Part I. Case Studies

Abstract. The Atmospheric Lidar (ATLID) aboard the Earth Cloud Aerosol and Radiation Explorer (EarthCARE) provides vertical aerosol profiles at 355 nm, which should be verified by other instruments. Here, we use ground-based Raman and polarization measurements from the China Aerosol Raman Lidar Network (CARLNET) as an independent ground-based reference for spaceborne products validation. We compare the particle backscatter coefficient, particle extinction coefficient, particle lidar ratio, and particle linear depolarization ratio from the ATLID aerosol product (A-AER) and the ATLID extinction, backscatter, and depolarization product (A-EBD). We also compare processing baselines for the same overpasses. For each case and product, we analyze both the ATLID profiles closest to each station and the quality-screened mean profiles of ATLID within 50 km of that station. Four cases at three ground stations were analyzed in detail. In the case of the Minfeng lidar station on 25 April, the layer-mean difference in the A-EBD particle extinction coefficient relative to the CARLNET lidar changes from -21.33% in baseline AE to +18.06% in baseline BA. The corresponding difference in the particle lidar ratio changes from -43.33% to +1.20%. In the Changdao-lidar case on 25 March and the Minfeng-lidar case on 31 March, A-EBD retains more within-layer peaks and troughs that correspond to those observed by CARLNET. A-AER yields smoother profiles with broader layers. In the Sheyang-lidar case on 30 May, cloud-classified bins affect the extremely high values in the low-altitude A-EBD profiles for baseline AF. Excluding these bins reduces the high values, but they remain above those in baseline BA and the CARLNET lidar observations. The Minfeng case on 31 March further shows that the particle extinction coefficient and particle lidar ratio are markedly low at low altitudes in baseline BA for both data products. Together, the four case studies demonstrate that CARLNET can provide independent reference values for ATLID aerosol optical data products over China. These four case studies also identify the products, parameters, and altitudes at which ATLID and CARLNET values differ. The effects of processing baselines vary by event and altitude. The baselines therefore cannot be ranked consistently by accuracy according to their release order.