Evaluating the accuracy of CFSR reanalysis hourly wind speed forecasts for the UK, using in situ measurements and geographical information

Climate data can be used in simulations to estimate the output of wind turbines in locations where meteorological observations are not available. We perform the most comprehensive evaluation of the NCEP CFSR reanalysis model hourly wind speed hindcasts to date, and the first for the UK, by correlating the data against 264 onshore and 12 offshore synoptic weather stations, over a period of 30 years. The correlation of CFSR data to in situ measurements is similar to alternative approaches used in other studies both onshore and offshore. We investigate the impact of the topography, land use and mean wind speed on the onshore locations for the first time. The analysis of these spatial factors shows that CFSR represents the variety of terrain over UK well, and that the worst correlated sites are those at the highest elevations. •CFSR wind speed is compared to onshore and offshore in situ measurements.•The effect of elevation, land use and wind speed on correlation is described.•CFSR is as accurate as any other raw reanalysis dataset that has been evaluated.•CFSR wind speeds are less accurate at high elevation locations.•CFSR wind speed data represents the conditions over diverse land uses well.