Measurements of wind speed near the ground surface are strongly influenced by the topography surrounding the anemograph site. The influence of a change in ground level is relatively local in comparison with the influence of a change of surface roughness, which persists many kilometres downwind of the change. These exposure effects are of a similar magnitude to other effects significant in the process of engineering design, such as the shape and size of a building, and therefore need to be assessed in a similar way. In this paper, a method of exposure assessment is described which utilises contemporary engineering models to show the effect of changes of surface roughness and ground level on wind speed. Results of the method are compared with measurements of extreme wind speeds obtained from a directionally inhomogeneous anemograph site. The method is seen to account for 85% of the observed directional variance.