The wind field in a cattle feedlot: measurements and simulations

Abstract

Cup and sonic anemometers were operated in and about an empty pen (60 m × 68 m) on the outer (south) edge of a large cattle feedlot in southern Alberta. Mean wind speed, measured at constant height above ground, varied by more than a factor of four across the pen, the spatial transects being distinct for different wind directions—implying (for instance) that efforts to quantify feedlot gas emissions by micrometeorological methods will be prone to error, unless the drastic lateral inhomogeneity of wind statistics is accounted for. A subset of the observations, selected for southerly winds and weak thermal stratification, were aggregated and compared with steady-state, three-dimensional numerical simulations using “ASL3D”, a Reynolds-averaged Navier–Stokes model with eddy viscosity closure that represents the influences both of feedlot windbreak fences and of topography (Wilson, 2018). Simulations confirm that wind drag on the tall (H ≈ 3 m), low porosity (25%) slatted wooden fences was by far the dominant aerodynamic disturbance at this site. Various options were tested for the placement of computational domain boundaries, and it was found that the influence of fences at the faraway edges of neighbouring pens is practically negligible in comparison with that of the fences lying immediately upwind—that is, the transect of relative mean wind speed within the instrumented pen was largely determined by the nearest upwind fence(s). It is also concluded that when the mean wind is obliquely incident on low porosity fences of this type, simulations are improved if the horizontal wind component tangential to the fence is forced to vanish (at the fence).