This paper summarizes the results of a study of wind speeds observed at heights up to 150 m above ground level around Missouri. This is an amalgamation of four projects that allowed a total of eleven tall communication towers to be instrumented with wind observation equipment across the State of Missouri. This provided an assessment of the wind resource and the characteristics of the seasonal and diurnal cycles of wind in different areas of Missouri at the heights of utility scale wind turbines. Comparisons were also made to wind speeds predicted at these levels from a previously published wind map. The main finding was that the observed winds at each tower were smaller than those presented in the wind map. The discrepancy is most likely to be due to underestimation of the surface roughness and turbulence leading to an overestimation of near-surface wind shear. However, the wind shear, as expressed by the shear parameter was consistently greater than the 'standard' value of 1.4. The reconciliation of these two apparently contradictory findings is that the shear varies with the height at which it is measured. In wind resource assessment, wind shear is usually observed below 50 m and is tacitly assumed to be constant with height when used to extrapolate winds to higher levels. The author advocates the use of the friction velocity as a measure of shear in wind power applications in preference to the shear parameter that is usually used. This is because the shear parameter has a velocity bias that can also manifest as a bias with height or season. As wind power resource assessment is starting to use taller towers than the standard 50 m, intercomparison of site resources and extrapolation to turbine heights can be compromised if the shear parameter is used. (c) 2010 Elsevier Ltd. All rights reserved.