Published works

Large-eddy simulation of a neutral atmospheric surface layer (ASL) flow is per- formed over a modelled tree canopy with heterogeneous leaf-area density. The canopy is arranged as a series of equally-sized stripes of different leaf-area density, emulating the study of Bou-Zeid et al. (E. Bou-Zeid, C. Meneveau, and M.B. Parlange. Large-eddy simulation of neutral atmospheric boundary layer flow over heterogeneous surfaces: Blending height and effective surface roughness. Water Resources Research, 40(2), 2004.) over heterogeneous rough surfaces. The simulation results are analysed to understand the qualitative similarities and differences between ASL flows over heterogeneous canopies and heterogeneous roughnesses. This will allow, in the future, the identification of the equivalent roughness length, displacement length, and blending height which parameterises the flow above the heterogeneous canopy. In the present work we restrict attention to the characterisation of the four canopy case and the blending height and β parameter, the ratio of shear stress to velocity at the canopy top. The general characteristics of the four-canopy case are representative of the other cases. Strong vertical velocities (ie up- and down-drafts) exist at the interface between the heterogeneous roughness stripes. However, for a canopy, vertical velocity couplets exist on the vertical interface between two canopies. This implies the presence of sub-canopy recirculation zones at canopy interfaces, which can be confirmed by visualisation of the fluid streamlines. Above the canopy internal boundary layers form over each canopy stripe and exhibit similar features to the characteristic upstream plumes of flow over a rough surface. The shear stress immediately above the canopy varies over the stripes but it varies more smoothly over a canopy than over a heterogeneous roughness. These simulations will allow the development of parameterisations for the near-surface layer and the sub-canopy winds. A better understanding of the effect of heterogeneous canopies on the sub-canopy winds will improve predictions of the wind reduction factor, and in turn, improve operational fire spread predictions.