J2.8
Roughness sublayer profiles over tall vegetation
Ricardo K. Sakai, University at Albany/SUNY, Albany, NY; and D. R. Fitzjarrald
The surface layer (SL) over vegetated surfaces can be divided into three sublayers: the inertial layer (IL), the roughness sublayer (RSL), and the canopy layer (CL).We seek similarity relationships using characteristic length and velocity scales to describe RSL profiles. In the CL, using the canopy area density (CAD, the vertical density distribution of the half plant area over a unit ground) for several forests, a new dimensionless height scale is defined (zc'). zc' is the normalized cumulative CAD, and it is defined by using assumptions analogous to those employed to describe the light transmission within the forest (Beer's law), in which plant elements attenuate the light beam. This parameter sorts the profile of the higher order moments among sparse and dense canopies in the CL. However, to find a more general description, information about the horizontal canopy distribution seems necessary.
The turbulent flow in the RSL may also depend on the spatial distribution of roughness elements. Anomalies of the dimensionless gradient in the RSL over a deciduous and a boreal forest will be investigated. We propose to use zc' instead of the normalized height (z/h, where h is the canopy height) to determine the displacement height (dc) as a better parameter for normalized horizontal wind profiles over several forests (from Boreas, HF, and LBA field experiments).
Spectral analysis of the turbulence indicates that the length scale within the forest is h-dc and that permits that the peaks of the spectral curves at several levels over and within a deciduous forest plotted against frequency normalized by the forest length scale have the same dimensionless frequency. The impact of the leaf presence on the spectrum and cospectrum in the RSL is assessed.
Recorded presentationJoint Session 2, Canopy Turbulence-ATMOSPHERE Interactions
Wednesday, 30 April 2008, 1:30 PM-4:00 PM, Floral Ballroom Jasmine
Previous paper