Datasets used to train an observation operator for weather radar data assimilation. LISCA_RADAR_2019_2023_INTERPOLATED_160km.nc.zip The dataset contains quality-controlled weather radar reflectivity observations from Lisca radar in Slovenia (46.07N, 15.29E, 945 m a.s.l., 0.5deg elevation angle) interpolated onto the ALADIN model grid using a nearest-neighbour methodology, yielding a horizontal spatial resolution of 4.4 km. Reflectivity is provided as hourly-summed values. Outliers are filtered out using a histogram-based Mahalanobis distance approach. ALADIN_2019_2023_radar_time_aligned.nc.zip The dataset contains temperature (t), horizontal wind components (u and v), relative humidity (r), at four level pressure (975 hPa, 925 hPa, 850 hPa, and 800 hPa) and three surface variables, 2m temperature (t2m), 2m relative humidity (r2m), and mean sea level pressure (msl) from the ALADIN numerical weather prediction model used at ARSO. Data are available at a horizontal resolution of 4.4 km (137×137 grid points centred on the Lisca radar location) and at temporal resolutions of 1h and 6h lead time, with a 6h analysis cycle (00, 06, 12, 18 UTC). Finally, the dataset is filtered to retain only those timesteps for which corresponding radar observations were available.