7A.3 Adjustment of turbulence to a forest canopy

Tuesday, 10 June 2014: 2:00 PM
Queens Ballroom (Queens Hotel)
Ian N. Harman, CSIRO Marine and Atmospheric Research, Canberra, Australia; and M. Bohm, J. J. Finnigan, and D. Hughes

Both the mean wind vector and turbulence adjusts as the atmospheric boundary layer passes from surface to surface through the creation of a series of internal boundary layers. The adjustment of the mean wind vector to a (forest) canopy is now well understood from numerical, theoretical and scaling perspectives. Phenomenological arguments for the dynamics of co-adjustment of the turbulence to a forest edge have also been advanced (Belcher et al. 2012). These indicate that the adjustment is characterised by the initiation and development of the characteristic inflexion in the mean velocity profile at the top of the canopy and 'mixing-layer type' turbulence evolving with distance downstream from the canopy edge. However, the observational evidence for this canonical picture consists of a few numerical studies augmented by low resolution field data and wind tunnel simulations.

To test the theoretical picture more stringently, we have investigated the adjustment of the turbulence to a forest canopy, using data taken in a wind tunnel study utilising laser-doppler velocimetry. Very detailed spatial sampling was used to resolve both the temporal and spatial adjustment of the mean flow and turbulence adequately (35 lateral positions across the edge spanning a range of -10 to 50 canopy heights, at 22 heights between 0 and 3 canopy heights in the vertical, and at approximately 2kHz). The adjustment of the mean wind vector follows the theoretical/canonical view, with adjustment largely completed within N Lc where Lc is the canopy adjustment length defined as 1/(drag coefficient X leaf area per unit volume). We will illustrate the co-adjustment of the turbulence through an analysis of the response of both of the second order moments and of the power and co-spectra. Specific consideration will be given to whether these data support the canopy-eddy generation picture of turbulence adjustment, the relative scales over which the adjustment processes occur and the presence of novel features in this highly resolved experiment.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner