Mean windspeed and turbulence were sampled in a square plot that was sheltered by a porous plastic fence. The side length of the plot was D=20m, while the height and the resistance coefficient of the fence were h=1.25m, k_r =2.4. The experiments took place in a large, uniform field of low, thin stubble (roughness length z₀ order 1 – 2 cm). Measurements, mostly with an array of 7 cup anemometers placed at z/h=0.5, cover all incident mean wind directions (b ) and a wide range in atmospheric stability, as determined from vertical profiles of the mean wind S₀(z) and temperature T₀(z) on a tower standing outside the shelter.

When the mean wind blew at nearly normal incidence (b=0) across a nearby fence to any particular anemometer in the shelter, a strong fractional wind reduction

F_WR={ S₀(z) – S(x,y,z) } / S₀(z)

was recorded, its numerical value being (roughly) that expected in the near lee of a long straight fence. F_WR was not very sensitive to stability, but very sensitive to mean wind direction: indeed when the wind blew over any corner of the plot in order to reach the anemometer (b=45^o), there was no reduction in mean speed!

This talk will also cover turbulence measurements from a sonic anemometer within the shelter, and speculate on the aerodynamics causing the dependency of F_WR on wind direction. A companion poster presentation will discuss tracer gas trials from an area source on ground inside this plot.