ABSTRACT: We present empirical downscaling of 5 state-of-the-art AOGCMs (Atmosphere-Ocean General Circulation Models) to investigate potential changes in wind speeds and energy density in northern Europe. The approach is based on downscaling the Weibull parameters of wind speed probability distributions from AOGCM-derived 500 hPa relative vorticity and sea-level pressure gradients, and is demonstrated to generate accurate depictions of the wind climate during the transfer function conditioning period. Bootstrapping is used to develop 100 realizations for each downscaling period and these are used to assess the uncertainty in the results due to stochastic effects in the AOGCM-derived downscaling predictors. Projected changes in the wind speed probability distribution vary with the AOGCMs from which the predictors are derived, but generally it is shown that mean wind speeds, 90th percentile wind speeds and energy density are slightly lower in the 20812100 climate projection period than during 19611990 at the majority of the 46 stations studied. Conversely it is found that there is no significant difference between conditions during 20462065 and 19611990 based on the ensemble of downscaling results. Equally, the winter time of 20462065 is largely indistinguishable from 19611990 for the majority of stations, while the winters of 20812100 appear to be associated with lower mean and 90th percentile wind speeds and energy density.
KEY WORDS: Wind speeds · AOGCM · Climate projections · Empirical downscaling
Full text in pdf format |
Next article |