Effects of trees on mean wind, turbulence and momentum exchange within and above a real urban environment

Large-eddy simulations (LES) are used to gain insight into the effects of trees on turbulence, aerodynamic parameters, and momentum transfer rates characterizing the atmosphere within and above a real urban canopy. Several areas are considered that are part of a neighborhood in the city of Vancouver, BC, Canada where a small fraction of trees are taller than buildings. In this area, eight years of continuous wind and turbulence measurements are available from a 30 m meteorological tower. Data from airborne light detection and ranging (LiDAR) are used to represent both buildings and vegetation at the LES resolution. In the LES algorithm, buildings are accounted through an immersed boundary method, whereas vegetation is parameterized via a location-specific leaf area density. LES are performed including and excluding vegetation from the considered urban areas, varying wind direction and leaf area density. Surface roughness lengths (z(0)) from both LES and tower measurements are sensitive to the 0 <= LAI/lambda(b)(f) < 3 parameter, where LAI is the leaf area index and lambda(b)(f) is the frontal area fraction of buildings characterizing a given canopy. For instance, tower measurements predict a 19% seasonal increase in z(0), slightly lower than the 27% increase featured by LES for the most representative canopy (leaves-off LAI/lambda(b)(f) = 0.74, leaves-on LAI/lambda(b)(f)? = 2.24). Removing vegetation from such a canopy would cause a dramatic drop of approximately 50% in z(0) when compared to the reference summer value. The momentum displacement height (d) from LES also consistently increases as LAI/lambda(b)(f)increases, due in large part to the disproportionate amount of drag that the (few) relatively taller trees exert on the flow. LES and measurements both predict an increase in the ratio of turbulent to mean kinetic energy (TKE/MKE) at the tower sampling height going from winter to summer, and LES also show how including vegetation results in a more (positive) negatively skewed (horizontal) vertical velocity distribution - reflecting a more intermittent velocity field which favors sweep motions when compared to ejections. Within the urban canopy, the effects of trees are twofold: on one hand, they act as a direct momentum sink for the mean flow; on the other, they reduce downward turbulent transport of high-momentum fluid, significantly reducing the wind intensity at the heights where people live and buildings consume energy. (C) 2017 Elsevier Ltd. All rights reserved.