CR 77:1-21 (2019)  -  DOI: https://doi.org/10.3354/cr01540

ABSTRACT: Monthly, seasonal and annual wind-speed series from 119 meteorological stations in the Czech Republic indicate significant decreasing trends in the period 1961-2015. Attribution analysis, applying multiple linear regression, was used to identify wind-speed components related to natural and anthropogenic climate forcings and internally induced climate variability. A significant link to wind speeds was detected for the North Atlantic Oscillation index, as well as for the closely-related Central European Zonal index, especially during the winter. An influence from the East Atlantic/Western Russia Pattern was found during autumn and winter, especially in the eastern part of the country. Changes in large-scale circulation did not seem to be primarily involved in long-term wind stilling, despite a formal correlation between the stilling and anthropogenic forcing. Distinct geographical variations in the regression-estimated links suggest profound influences from interactions between local features of the measuring sites and large-scale climate-forming factors. In total, 11 Euro-CORDEX regional climate model (RCM) simulations for representative concentration pathways RCP4.5 and RCP8.5 were used for projection of annual and seasonal mean daily wind speeds for the Czech Republic for 1951-2100. Despite correction of the model biases for individual RCMs, these simulations largely underestimated the magnitude of declining observational trends in 1981-2010, with only annual, winter and spring values sharing the same trend for both RCPs. Linear trends in wind speeds calculated for 1981-2100 for both RCPs show a significant negative trend in summer, while significant positive trends in winter and spring wind speeds were recorded for RCP8.5.

KEY WORDS: Wind speed · Climate forcings · Circulation indices · Attribution analysis · Windspeed projections · Regional climate models · Czech Republic