A simple hybrid statistical–dynamical downscaling method for emulating regional climate models over Western Europe. Evaluation, application, and role of added value?

A hybrid statistical dynamical downscaling method intended to emulate regional climate models is described and applied Western Europe. The based on a constructed analogues algorithm, already used for downscaling. For emulation, the relationship not derived from observations but projections at low high resolution. approach therefore does rely stationarity assumption inherent conventional Within perfect model framework, using large number of projections, shown reproduce change signals very well outperform also analogues. remains skillful even when resolution data. In practice, two emulation modes exist. GCM/RCM mode, built between RCM its forcing GCM. RCM/RCM same after aggregation results grid. large-scale signal downscaled GCM generally retained with /RCM mode. Additionally, choice pair learning leads differences in scale (i.e. resolution) These are explained by that exist projected changes current RCMs their GCMs. Whether these testimony real added value context, or whether they have other causes, crucial question.