Abstract. Polar stratospheric clouds (PSCs) play a critical role in the stratospheric ozone depletion processes. The last 30 years have seen significant improvements in our understanding of the PSC processes but PSC parametrization in global models still remains a challenge due to the necessary trade-off between the complexity of PSC microphysics and model parametrization constraints. The French Antarctic station Dumont d'Urville (DDU, 66.6^∘ S, 140.0^∘ E) has one of the few high latitude ground-based lidars in the Southern Hemisphere that has been monitoring PSCs for decades. This study focuses on the PSC data record during the 2007–2020 period. First, the DDU lidar record is analysed through three established classification schemes that prove to be mutually consistent: the PSC population observed above DDU is estimated to be of 30 % supercooled ternary solutions, more than 60 % nitric acid trihydrate mixtures and less than 10 % of water–ice dominated PSC. The Cloud–Aerosol Lidar with Orthogonal Polarization PSC detection around the station are compared to DDU PSC datasets and show a good agreement despite more water–ice PSC detection. Detailed 2015 lidar measurements are presented to highlight interesting features of PSC fields above DDU. Then, combining a temperature proxy to lidar measurements, we build a trend of PSC days per year at DDU from ERA5 (the fifth generation of European ReAnalysis) and NCEP (National Centers for Environment Protection reanalysis) reanalyses fitted on lidar measurements operated at the station. This significant 14-year trend of −4.6 PSC days per decade is consistent with recent temperature satellite measurements at high latitudes. Specific DDU lidar measurements are presented to highlight fine PSC features that are often sub-scale to global models and spaceborne measurements.