A statistical-dynamical downscaling procedure for global climate simulations is described. The procedure is based on the assumption that any regional climate is associated with a specific frequency distribution of classified large-scale weather situations. The frequency distributions are derived from multi-year episodes of low resolution global climate simulations. Highly resolved regional distributions of wind and temperature are calculated with a regional model for each class of large-scale weather situation. They are statistically evaluated by weighting them with the according climate-specific frequency. The procedure is exemplarily applied to the Alpine region for a global climate simulation of the present January climate. Δλ west-east mesh size in geographic coordinates Δϕ south-north mesh size in geographic coordinates N number of large-scale weather classes n number of regional-scale event classes p pressure P probability O large-scale event ϕ regional-scale event q v specific humidity θ potential temperature u west-east wind component v south-north wind component