Comparison of different wind products and buoy wind data with seasonality and interannual climate variability in the southern Bay of Biscay (2000-2009)

Abstract

Ocean surface winds are essential factors in determining oceanographic and atmospheric processes that can affect ocean circulation and wave generation. Accurate surface wind datasets are needed, therefore, to enable the proper analysis of these processes. Wind data from six databases (National Centers for Environmental Prediction reanalysis (NCEP Reanalysis II), European Centre for Medium-Range Weather Forecasts (ECMWF) re-analysis (ERA-Interim), Modern-Era Retrospective-analysis for Research and Applications (MERRA), NCEP Climate Forecast System Reanalysis (CFSR), QuikSCAT and Cross-Calibrated Multi-Platform (CCMP)) were compared with wind measured in situ by four ocean buoys at the southern limit of the Bay of Biscay. The study covered the period 2000-2009 in such a way that the extent of the time series reduced the margin of error and allowed the disaggregation of the wind data using velocity bins and direction sectors. Statistical results confirmed that datasets with finer spatial resolution (lower than 0.5°×0.5°) gave better results, especially in near-shore areas. A more complete analysis was, therefore, carried out using the finer resolution datasets (QuikSCAT, CCMP and CFSR). This comparison showed that all the datasets were less accurate at low wind speeds (<4ms-1) and more accurate at moderate wind speeds. The calculated mean wind speed errors were similar for the three datasets, and the lowest value (1.67ms-1) was from the CCMP dataset. The lowest mean error for wind direction (~37°) was also observed in the CCMP data. The lowest mean wind speed (and direction) bias was obtained from the QuikSCAT data, and the next lowest from the CFSR data. The seasonality for north and east wind components was also determined for the last decade and the results were consistent with forcing for the continental slope current seasonality and winter temperatures or Navidad by wind stress. Correlations between NAO and north and east wind components were low showing that NAO could not be used as a proxy for local wind stress in the southern Bay of Biscay.