Identification of processes governing within-canopy momentum and scalar transfer from vegetation canopies to urban landscapes

Turbulent transfer of momentum and scalars from urban environments, vegetation canopies or from bare soil is the principal means of connection between the planetary surface and the lower atmosphere and so plays an important role in many aspects of everyday life. Momentum transfer from vegetation canopies can be understood with the help of the mixing layer analogy (MLA), which provides a well-tested theory for the development of the coherent eddies responsible for most of the exchange. However, not all canopy flows follow this theory with urban landscapes providing the most important exceptions.

The MLA points out that plane mixing layers are a better model for canopy turbulence than classical rough wall boundary layers. Canopy flows share with mixing layers an inflexion point in their mean velocity profile. This inflexion point is inevitably seen when the canopy velocity field is averaged in time and over horizontal planes containing many roughness elements. In vegetation canopies this average is dynamically meaningful because the resulting inflexion-point instability is clearly linked to the dominant canopy eddies we observe (Finnigan et al., 2009). In canopies composed of sparser arrays of bluff bodies, however such as urban canopies, it is not obvious that the inflected mean velocity profile plays any dynamic role. Instead, the turbulent flow in the canopy may be dominated by the wakes of the individual roughness elements and displays no universal behaviour.

The ‘Black Forest' wind tunnel model canopy was constructed to investigate the effect of scalar source-sink distributions on turbulent heat transfer and displays features of both vegetation and bluff body canopies. It is an ideal environment in which to study the transition from universal MLA behaviour to wake-dominated turbulence. The effect of its turbulence regime on scalar transfer has been investigated by Böhm et al. (2008) using octant analysis that builds on the idea of quadrant analysis by introducing eight conditions linking scalar and velocity perturbations. For example, using heat as a scalar, octant analysis allows us to look at the behaviours of warm and cold sweeps relative to similar conditions on the other three quadrants. Octant analysis can give insights into the nature of the coupling and decoupling of momentum and scalar transport and is particularly useful in this transitional regime where aspects of both universal and wake dominated turbulence are present.

Böhm et al. (2008) divided the within-canopy flow in the Black Forest model into wake, non-wake and transitional regimes and showed that vertical scalar fluxes within the canopy were highest for scalars emitted at ground level into the non-wake region of the flow. Conversely, in the wake/transitional parts of the canopy, scalar concentrations were generally higher but vertical fluxes smaller than in the non-wake regions. The scalar field was also very intermittent in the wake compared with the non-wake region of the flow. These results were interpreted using octant analysis as evidence of scalar labelling of the turbulence, where scalar sources near the ground label the sweeps in non-wake regions and ejections in the wake regions of the flow whereas scalar sources higher in the canopy label sweeps.

In this paper we first integrate the ideas of Finnigan et al. 2009 with octant analysis to develop improved understanding of the physical processes driving scalar transfer in the highly turbulent flows within vegetation and urban canopies. We develop octant analysis as an interpretative tool to better understand the links between scalar and momentum transfer, including the influence of scalar source distribution, conditional scalar labelling of turbulence, and the instability mechanisms driving momentum transfer within and above roughness elements. We then apply octant analysis to investigate linked scalar and momentum transport in three canopies spanning the range: vegetation-transitional-urban using data from (1) Tumbarumba forest; (2) Black Forest wind tunnel experiment; and (3) an urban full-scale experiment (The Basel Urban Boundary Layer Experiment, BUBBLE, 2002).