Evaluation of Near-Surface Wind Speed Changes during 1979 to 2011 over China Based on Five Reanalysis Datasets

Abstract

Wind speed data derived from reanalysis datasets has been used in the plan and design of wind farms in China, but the quality of these kinds of data over China remains unknown. In this study, the performances of five sets of reanalysis data, including National Centers for Environmental Predictions (NCEP)-U.S. Department of Energy (DOE) Reanalysis 2 (NCEP-2), Modern-ERA Retrospective Analysis for Research and Applications (MERRA), Japanese 55-year Reanalysis Project (JRA-55), Interim ECMWF Re-Analysis product (ERA-Interim), and 20th Century Reanalysis (20CR) in reproducing the climatology, interannual variation, and long-term trend of near-surface (10 m above ground) wind speed, for the period of 1979–2011 over continental China are comprehensively evaluated. Compared to the gridded data compiled from meteorological stations, all five reanalysis datasets reasonably reproduce the spatial distribution of the climatology of near-surface wind speed, but underestimate the intensity of the near-surface wind speed in most regions except for Tibetan Plateau where the wind speed is overestimated. All five reanalysis datasets show large weaknesses in reproducing the annual cycle of near-surface wind speed averaged over the continental China. The near-surface wind speed derived from the observations exhibit significant decreasing trends over most parts of continental China during 1979 to 2011. Although the spatial patterns of the linear trends reproduced by reanalysis datasets are close to the observation, the magnitudes are weaker in annual, spring, summer and autumn season. The qualities of all reanalysis datasets are limited in winter. For the interannual variability, except for winter, all five reanalysis datasets reasonably reproduce the interannual standard deviation but with larger amplitude. Quantitative comparison indicates that among the five reanalysis datasets, the MERRA (JRA-55) shows the relatively highest (lowest) skill in terms of the climatology and linear trend. These results call for emergent needs for developing high quality reanalysis data that can be used in wind resource assessment and planning.