Dependence of variational 2-m temperature analyses on synoptic situation and underlying terrain

A two dimensional variational (2DVar) 2-m temperature analysis is computed over Austria using a 1-km downscaled background grid from the operational ALADIN mesoscale model and surface observations from the Teilautomatisches Wetterstationsnetz (TAWES) network operated by the Central Institute for Meteorology and Geodynamics (ZAMG) of Austria. The dependence of the quality of the 2DVar analyses on synoptic situation and underlying terrain is assessed through a data denial methodology in which a single observation is withheld from an experimental analysis and compared to control analyses that use all observations. The 2DVar analyses are further compared to ZAMG's Integrated Nowcasting through Comprehensive Analysis (INCA) system, which computes high resolution surface analyses through spatial interpolation of observations using geometric distances in the horizontal and potential temperature differences in the vertical direction, as well as other terrain-dependent corrections.