Wind Tunnel Analysis of 3D Fine Scale Momentum Adjustment Across and Downwind of a Forest Edge

Adjustment of the mean and turbulent characteristics of the atmospheric boundary-layer to changes in surface roughness remains a key topic within boundary layer meteorology with several important applications. Current understanding of the nature of turbulent transfer within and above vegetation canopies has been developed from many decades of investigation usually under ideal conditions of uniform ground cover and flat terrain (e.g. Finnigan 2000; Finnigan et al. 2009). However, most landscapes have some degree of complexity in terrain and are covered by fragmented and patchy landuse. Consequently there is considerable interest in understanding flow and flux adjustment across these kinds of landscapes. Data on internal boundary layer adjustment in complex terrain provide insights to the interpretation of flux tower measurements and into the parameterisation of surface exchange processes in such landscapes. They are also economically important. Windstorms can cause major damage to forests, for example Sweden sustained windthrow damage of 75 million m3 in January 2005, equivalent to a normal annual harvest. Windthrow is related to tree location relative to a forest edge and thus understanding wind flow adjustment, among other factors, offers valuable insights to wind damage processes.

The mixing layer analogy (MLA) provides an internally consistent theory for momentum transfer to vegetation canopies and the existence and structure of the observed coherent eddies which dominate the flow and transport canopy top. 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 observed (Finnigan et al., 2009). The development of coherent eddies behind a forest edge has been studied numerically (e.g. Dupont and Brunet 2009) but, to date, experimentation has been limited due to the requirement for observations which are highly resolved, in space and time (Rominger and Nepf 2011).

It is difficult and expensive to undertake momentum adjustment studies under field conditions. Consequently, we investigated flow adjustment using a well characterised wind tunnel model that mimics a tall vegetation canopy (Raupach et al., 1986). The ‘Black Tombstones' (BT) canopy has been constructed to extend our investigations into canopy-boundary layer exchange including studies of the impacts of atmospheric stability, complex terrain and source-sink distribution on the exchange of momentum and scalars. The BT canopy is an ideal environment in which to study the adjustment of momentum to a rapid roughness change, synonymous with a grassland-forest interface. Mean wind and turbulent characteristics were measured using 3D Laser Doppler Velocimetry at high spatial resolution from 1 canopy height (h) upwind of the interface from a short ‘grass' surface to a tall ‘forest' model canopy to 5h downwind of the roughness change. Beyond 5h, four spatially representative profiles were sampled at 7 locations to 40h downstream of the forest edge to allow estimates of spatial averages well downwind of the roughness change. At 50h, a detailed spatial analysis was undertaken using 84 vertical profiles to quantify the fully adjusted flow. In addition, profiles were measured at 5h and 10h upwind of the roughness change. High resolution vertical profiles extend from 0.1h to above 3h.

The mean wind speed begins to adjust to the roughness change slightly upwind of the model grassland-forest interface with rapid adjustment occurring immediately downwind of the transition. Above 1.5h turbulent mixing smooths out spatial variations in mean wind and turbulence profiles related to individual canopy elements. Within and immediately above the canopy, the profiles reflect their spatial location relative to these elements. In a wake-like zone behind each element, adjustment is rapid as momentum is efficiently absorbed by pressure forces across each element and wake patterns show little variation with distance downwind of the forest edge. In contrast, within the non-wake zones preliminary analysis shows that adjustment occurs at different X-locations depending on the flow characteristic. Mean streamwise windspeed is fully adjusted by 10h, whereas Reynolds stress and turbulent kinetic energy (TKE) appears to only reach adjustment downwind of 20h. There is no clear “enhanced gust zone” in the upper canopy as suggested by Dupont et al. (2011) although there is clear enhancement above the canopy. There is some enhancement close to the forest edge when compared with adjusted conditions (>30h ). Reynolds stress and TKE show increased turbulence close to the forest edge with the development of a highly turbulent layer downwind of 10h . The correlation between u and w shows clear development of coherent turbulent structures from h downstream of the forest edge extending above and into the upper part of the canopy.

In this paper we present a spatial analysis of the wind tunnel data, both in terms of 3D plots of individual profiles as well as 2D plots of the spatially averaged mean and turbulent wind field characteristics. We compare the BT results with other field, wind tunnel and modelling studies, and attempt to develop an improved understanding of the physical processes driving momentum adjustment over vegetation canopies following a change in roughness within the theoretical framework advanced by Finnigan et al. (2009). We will also illustrate how these ideas can be used to predict the adjustment of fluxes at changes in vegetation and surface roughness, in the interpretation of flux tower observations in heterogeneous terrain and in managing forests to reduce wind damage risk.

Dupont, S. and Brunet, Y., 2009. Coherent structures in a canopy edge flow: a large-eddy simulation study. J. Fluid Mech., 630: 93-128

Finnigan, J.J., 2000. Turbulence in plant canopies. Annu. Rev. Fluid Mech., 32: 519-571.

Finnigan, J.J., Shaw, R.H. and Patton, E.G., 2009. Turbulence structure above a vegetation canopy. J. Fluid Mech., 637: 387-424.

Raupach, M.R., Coppin, P.A. and Legg, B.J., 1986. Experiments on scalar dispersion within a model plant canopy. Part I: The turbulence structure. Boundary-Layer Meteorol., 35: 21-52

Rominger, J.T. and Nepf, H.M., 2011. Flow adjustement and interior flow at a finite porous obstruction. J. Fluid Mech. 680: 636-659