Theory for a TKE based parameterization of wind gusts

Wind gusts are a manifestation of the stochastic nature of atmospheric turbulence. Wind gust predictions should reflect this fact and be essentially probabilistic. This article outlines a theory that provides a functional relationship between the value of a predicted gust and the factors that determine it: the windspeed U, the height above the terrain z, the surface roughness z 0 , the associated probability P, the length of time window T and the number of independent stations n to which the prediction applies, the defined gust averaging time t and the spectral characteristics of atmospheric turbulence under neutral and unstable conditions. This functional relationship can be used to formulate a parameterization of wind gusts based on the turbulent kinetic energy TKE from a NWP model. Other applications of this theory include the determination of the effective roughness at measurement sites from observation records, the derivation of observations under WMO reference conditions from actual observations and the reconstruction of wind records for gust related incidents. It is proposed that the present theory is a starting point for further development including recent insights in the structure of the atmospheric boundary layer and the nature of turbulence therein. Wind gusts play an important role in the determination of the wind loading of buildings and vehicles. Wind gusts are the extremes of the turbulent, stochastic wind field. The probability density function of wind gust magnitude can be derived from the spectrum of wind speed variations. This pdf can be used to determine risk profiles of buildings and vehicles and to determine warning levels for wind gust predictions. Similar to the behavior of buildings and vehicles the measurement of wind and wind gust can be considered to be the response of a (electromechanical) filter to the supplied wind field. In the translation of observed wind gusts to specific applications three components have to be considered: the characteristics of the wind field (spectrum or probability density), the characteristics of the observation (instrument properties, processing and reporting practice) and the characteristics of the application (length scales, environment, risk prole.) In the forecasting of wind gusts the widest range of application is achieved by modeling the complete pdf of wind speed variations. From a predicted pdf professional users, knowing their risk proles, may determine their own warning levels. Providing pdf’s however is not a practical proposition for issuing warnings to the general public. For public forecasting mimicking observational practice is a more reasonable approach. This implies the modeling of wind speed extremes as they are observed and reported, since the complete spectrum or pdf is not normally measured. This approach also provides the possibility of direct verification of the wind gust model with available observations.