Scintillometer measurements inside two tree canopies

The knowledge about in-canopy turbulent exchange becomes increasingly important, because attempts are made to model dispersion processes within rough canopies like forests or cities.

Within any roughness sublayer, and especially in canopy sublayers, turbulent exchange of sensible heat and momentum are spatially highly variable and their measurement by eddy correlation instrumentation needs extensive arrays to retain statistically representative averages. Scintillometry on the other hand already provides area averaged fluxes, but its applicability in this environment is in doubt, because calculation algorithms rely on surface layer scaling.

To test the applicability of scintillometry within canopies, two experiments were conducted in 2002 and 2003 within Mediterranean canopies. In the trunk space of two plantations, a small aperture scintillometer (Scintec SL S20) was operated simultaneously with an array of 8 and 10 sonics.

The first canopy, an olive tree plantation in Southern Italy had a stand age of 100 years, a density of 132 trees per ha, and an average tree height of 5 m. Between the trees was bare soil and the canopy was relatively open. Fetch in the main wind directions (W and N) was 200 to 300 m. The experiment took place from July 27 to August 6, 2002. The second canopy, a cork oak plantation, is located 35 km SE of Lisbon, Portugal. The age of the trees was 100 years with a density of 76 trees per ha and an average canopy height of 10 m. The understorey was patchy and varied between dry grass and shrubs (cistus), the latter reached in average up to 0.4 m. Fetch conditions in the main wind direction (NW) were good. Measurements were carried out from July 1 to July 13, 2003.

The point measurements of sensible heat fluxes measured with the sonics show a large spatial variability. The average calculated from all sonics (Hso) shows a surprisingly good linear relationship with those derived from the scintillometer (Hsz), however the agreement is only qualitatively. For a 1:1 relationship Hsz has to be adjusted to Hso by the factors 1.35 (olives) and 0.8 (cork oaks). Possible reasons fort this under and overestimations will be discussed.

The momentum fluxes, presented as friction velocities, show a different picture. For the olive tree plantation the general agreement is good. The values from the scintillometer (u*sz) during unstable conditions tend to be smaller than the u*so-values. For the cork oaks the situation is very different for unstable situations: here the scintillometer seems to highly overestimate the friction velocities relative to the sonic values.