An extended series of surface observations is used to compare observed
surface winds with winds computed using the geostrophic relationship.
These computations are done for both steady and unsteady wind regimes
. Large differences are found in the comparisons of observed to comput
ed winds. The differences exhibit pronounced seasonal and diurnal vari
ability that appear to reflect both boundary layer stability and small
-scale wind and pressure fields-for example, those attending land-sea
breezes and thunderstorms. The results of this study may be useful to
those engaged in studying global datasets and to modelers, who are con
tinually challenged to improve the treatment of parameterization of tu
rbulent processes. However, it is not obvious that any simple paramete
rization can be applied to obtain an accurate estimate of the surface
wind in central Florida, given only the large-scale pressure gradient
or a model-predicted wind above the surface as input. The use of the p
ressure field to estimate surface winds is an uncertain exercise at be
st.