A Weibull Distribution Based Technique for Downscaling of Climatic Wind Field
This study proposes a simple approach based on Weibull distribution parameters for downscaling climatic wind speed and direction. In this method, the Weibull parameters of a Global Climate Model (GCM) are modified using Weibull parameters of the reference data (ECMWF). To correct the wind direction, the downscaling technique was applied to the eastward and northward wind components. All the wind components were simply transformed to positive values in order to fit a Weibull distribution. The unbiased wind speed was calculated by integrating the corrected wind components. Moreover, other models were considered to directly modify the wind speed (not wind components) using the same methodology. Performance and ability of the proposed approach were evaluated against the existing statistical downscaling techniques such as Multiplicative Shift Method (MSM), quantile mapping and support vector regression. In the models, the 6-h GCM wind component/speed was the sole predictor and the ECMWF reanalysis wind data was considered as the predictand. It is demonstrated that direct application of the proposed method on the wind speed slightly gives better estimation of the predictand rather than its application on wind components. The results indicate the Weibull distribution based method outperforms the other techniques for wind direction and magnitude. The method provides sound predictions for a wide range of wind speed from low to high values. By using the proposed downscaling technique for wind components, wind direction can be adjusted accordingly.
KeywordsStatistical downscaling Weibull parameters Wind direction Wind components Quantile mapping
- He, Y., Monahan, A.H., Jones, C.G., Dai, A., Biner, S., Caya, D., Winger, K.: Probability distributions of land surface wind speeds over North America. J. Geophys. Res. Atmos. 115(D4), (2010)Google Scholar
- Li, H., Sheffield, J., Wood, E.F.: Bias correction of monthly precipitation and temperature fields from intergovernmental panel on climate change AR4 models using equidistant quantile matching. J. Geophys. Res. Atmos. 115(D10) (2010)Google Scholar
- Shirkhani, H., Seidou, O., Mohammadian, A.: Downscaling methods for wind data: case study of Agadez in Niger. In: 3rd Specialty Conf. on Disaster Prevention and Mitigation, Canadian Society for Civil Engineering (CSCE), Montreal, Canada (2013)Google Scholar
- Vapnik, V.: The nature of statistical learning theory. Springer science & business media, (2013)Google Scholar