A simple formulation of the boundary layer is developed for use in large-scale models and other situations where simplicity is required. The formulation is suited for use in models where some resolution is possible within the boundary layer, but where the resolution is insufficient for resolving the detailed boundary-layer structure and overlying capping inversion. Surface fluxes are represented in terms of similarity theory while turbulent diffusivities above the surface layer are formulated in terms of bulk similarity considerations and matching conditions at the top of the surface layer. The boundary-layer depth is expressed in terms of a bulk Richardson number which is modified to include the influence of thermals. Attention is devoted to the interrelationship between predicted boundary-layer growth, the turbulent diffusivity profile, 'countergradient' heat flux and truncation errors.The model predicts growth of the convectively mixed layer reasonably well and is well-behaved in cases of weak surface heat flux and transitions between stable and unstable cases. The evolution of the modelled boundary layer is studied for different ratios of surface evaporation to potential evaporation. Typical variations of surface evaporation result in a much greater variation in boundary-layer depth than that caused by the choice of the boundary-layer depth formulation.

In an effort to describe the basic vertical structure of the nocturnal boundary layer, observations from four experiments are analyzed. During the night, the depth of significant cooling appears to increase with time while the depth of the turbulence and height of the low level wind maximum tend to remain constant or decrease with time. Since the inversion layer extends above the low level wind maximum and shear is small in the region of the low level jet, the Richardson number reaches a maximum at the jet level and then decreases again with height. As a result, turbulence is observed to be a minimum at the height of the low level wind maximum and then increases again above this height.

We propose a simple gust definition based on the theory of excursions by Rice (1944 and1945). We discuss the relation to the distribution of extreme events and demonstrate theoretically and experimentally that the most probable extreme event is very close to being identical to the gust according to our definition. We demonstrate how it is possible to predict the gust on the basis of the measured mean wind and variance rather than rely on actually measured extreme excursions. Our gust definition also allows us to predict the average duration of a gust.

Four models of surface boundary-layer flow in complex terrain are compared with observations made at Blashaval Hill, North Uist, Scotland. The field experiment is described by Mason and King (1985). Three of the models are derived from the two-dimensional theory of Jackson and Hunt (1975) and are described in Mason and King (1985), Walmsley et al. (1986) and Troen and Petersen (1989). The fourth is a mass-consistent code based on Traci et al. (1979). The model results are in good agreement with each other and are generally within the observed range of variation ( -+ 16%) in normalized wind speed. For most wind direcions (7 of ll), model results of normalized wind speed at the summit were within 7% of the observed mean values. For some wind directions, calculations using the "Guidelines" of Walmsley et al. (1989) suggested that variations in surface roughness were important. This led us to apply one of our models incorporating nonuniform surface roughness. The lack of significant improvement for cases when water lay upstream of Blashaval Hill is attributed to compensating changes at summit and reference sites and to very local effects on the wind data. Sensitivity to topography lying to the west and northwest of Blashaval was also investigated. Results suggested an influence from those distant topographic features for some wind directions. When those features were incorporated, maximum errors in normalized wind speed at the summit were reduced from 18 to 13%.

A model has been derived for calculating the effects of stability and the finite height of the planetary boundary layer upon the long-term mean wind profile. A practical implementation of this probabilistic extended similarity-theory model is made, including its incorporation within the European Wind Atlas (EWA) methodology for site-to-site application. Theoretical and practical implications of the EWA methodology are also derived and described, including unprecedented documentation of the theoretical framework encompassing vertical extrapolation, as well as some improvement to the methodology. Results of the modelling are shown for a number of sites, with discussion of the models' efficacy and the relative improvement shown by the new model, for situations where a user lacks local heat flux information, as well as performance of the new model using measured flux statistics. Further, the uncertainty in vertical extrapolation is characterized for the EWA model contained in standard (i.e. WAsP) wind resource assessment, as well as for the new model.