Regional Mountain Conference

Globally, mountain regions remain sparsely covered by long-term meteorological observations, and the number of monitoring stations has been declining. Coarse-resolution global climate models (GCMs) and gridded observational datasets fail to adequately represent small-scale orographic and local climatic drivers and characteristics, which hampers reliable assessments of climate change impacts in high-altitude environments. This study assesses the empirical downscaling tool RASCAL (Reconstruction by AnalogS of ClimatologicAL time series), which reconstructs local temperature variability by linking station observations to large-scale predictors. Seventeen long-term stations along the Alpine ridge in Switzerland, Austria, Germany, Italy and Slovenia were selected for reconstruction. They span elevations from valleys to more than 3000 m a.s.l. and represent a broad topographic gradient.

The evaluation, based on reconstructions using geopotential height at 925 hPa as predictor, indicates that reconstruction skill varies substantially across sites. While RASCAL generally reproduces the mean temperature series and, in most cases, the trend, its ability to capture long-term climatic trend differences along the elevation gradient is inconsistent. Some reconstructions reproduce elevation-dependent warming rates (EDW), but their magnitude is highly variable and strongly dependent on local topography, elevation, and site-specific climatic factors. Further analyses will incorporate multiple predictor variables, including geopotential height at different pressure levels and radiation, to reconstruct minimum, maximum and mean temperature series, to assess whether this improves reconstruction accuracy.