Surface wind speed estimation important for a variety range of engineering applications such as wind energy generation, pollutant plume dispersion and tall buildings design. Like other climatic variables, wind speed is likely to change in the future and the direction of that change may have large implications for the safety and productivity of several civil engineering structures. Projected wind speed under various climate change scenarios by Global Circulation Models (GCMs) and Regional Climate Models (RCMs) are available from several climate centers. However, surface wind speed is required at much smaller scales than that resolved by GCMs and RCMs. Therefore, it is vital to develop methods to downscale the GCM data to finer scales. In this paper, a linear model and two polynomial models of degrees 2 and 3 were used to link the large scale NCEP reanalysis wind velocity to the station-level wind velocity at the Agadez station in Niger. One model was developed for each of the 365 days of the year. The 1950-1985 period was used for calibration, and the obtained model was verified on the 1986-1988 period. A good agreement was found between the simulated and observed wind velocity at Agadez (R2=0.53, 0.56 and 0.57 for the calibration period and 0.58, 0.58, 0.57 for the validation period). All models were afterward applied using simulated wind values from the KNMI-RACMO2.2 Regional Climate Model. The bias and changes in the standard deviations in the wind values simulated using the RCM data were afterward used to build downscaling relationships for future periods at the Agadez station.
展开▼
