Surface Wind Speed over Land: A Global View

Abstract

This paper provides a global analysis of surface wind speed (SWS) based on data from 1506 weather stations. It is found that the local variation in monthly normal SWS can generally be captured as a sine function of calendar month using three descriptors: the annual mean SWS, the annual amplitude of SWS, and the month of peak SWS. With station-specific descriptor values, the function accounts statistically for 97% of the variance in SWS contained in the total dataset; the corresponding root-mean-square error is 0.30 m s?1, or 8.2% of the SWS data mean. Global maps of the three SWS descriptors are produced by geographically interpolating the locally fit values. By reversing the above procedure, the maps may be used to provide first approximations to monthly SWS anywhere in the world.