Turbulence Statistics Within and Above a Nearly Sparse Forest Canopy

The flow field in a sparse forest canopy of small size exhibits complex thermal and mechanical characteristics since the influence of individual trees or forest features on the turbulence statistics can be discernible [1]. Green et al. [2] identified significant spatial variations of the vertical distributions of horizontal wind speeds and shear stresses when comparing measurements between and within the rows of a widely spaced Sitka spruce. Depending on the tree density the statistical parameters could deviate between 20 % and 90 %. It is well known that a dense canopy produces flow with characteristics of a mixing layer [3, 4]. Flow characteristics over sparse canopies of small size may have attributes of both near surface rough-wall boundary and mixing layer. Further wind tunnel studies [5] found eddy attachment and turbulent diffusivity to vary with canopy density.

The herewith presented work expands the current state of knowledge of micrometeorological processes in a nearly sparse and short canopy. The objective of this field study is to analyze the air flow within and above a short Pinus radiata D. Don forest with a tall instrument mast mounted with five sonic anemometers. The mast covers the boundary layer flow between 0.3h and 2.3h and hence spans almost over the whole roughness layer. This preliminary data analysis focuses on two segments in time that resorts to data sets with highest reliability.

The experimental site was located in Kaingaroa forest, one of the largest plantations in the southern hemisphere. A ten meter mast with five Campbell Scientific Inc. CSAT3B sonic anemometers was erected between three year old Pinus Radiata trees. The average height of the trees that surrounded the mast was 4.3 m at the end of growing season. The tree spacing was roughly 3 m and the row spacing was nominally 4 m. All sensors were pointing at true north and recorded 20 Hz velocity data at heights of: 0.3h, 0.8h, 1.1h, 1.6h and 2.3h above ground. The heights were measured from the ground, but wind-rows (woody debris heaps) of 1.6 m height in every third row impede a precise determination of the tree height.

The trial commenced on 31 January 2017 and ended after 80 days. Several severe weather events occurred, these data were set aside for these analyses. For this preliminary analyses we used data collected over 2.25 days in mid-February 2017, which had a continuous data collection by all five sensors. These data are the basis for the following analysis and are an example of the boundary layer flow found in a sparse small in size canopy.

The daily cycle of the virtual air temperature as measured by the sonic anemometers is depicted in Figure 2. The temperature profile of these comparably widely spaced short trees does agree to some extent with generalized observations in high vegetation [6]. Although the wider spacing of the tree rows seems to increase the cooling effects near the surface due to radiation, as indicated by the lower temperatures in that part of the canopy layer during the second night. The highest temperatures appeared in the upper crown between 0.9h and 1.6h but also close to the forest floor due to an increased turbulent exchange. This indicates a weaker decoupling of the understory from the atmosphere by the tree crowns compared to high density forests systems.

Figure 3 is a summary of the vertical profiles of the canopy turbulence. The boxplots are based on 30 min averages that sum up to the 54 hours-time segment. The distinctive inflection point in the wind speeds agrees well with observations made by Raupach et al [3], Gao et al. [7] and Green et al. [2]. The roughness height z₀ was determined by computing a least-squares regression of the log-linear wind profile and by finding the interception of the regression line. The roughness height ranges between 0.9 m and 3.0 m with an average value of 1.4 m. This agrees well with preceding studies [8].

The higher order moments, represented by the velocity skewness in Figure 3, indicate that the distribution of the vertical velocity component strongly leans towards the left side of the expectation value close to the ground and above the canopy. This is the result of strong turbulences entering the forest canopy from above. The normalized shear stresses above the canopy are nearly constant but exhibit a strong decay with decreasing height caused by an elimination of the streamwise momentum through canopy drag. This observation is confirmed by the time trace depicted in Figure 4 that captures a strong single gust event (taken from a different part of the data). The occurring sweep is unusually intense and deeply penetrates the canopy.

This abstract summarizes an experimental study conducted in a sparse canopy with an instrumented mast that covers almost the complete height of the prevailing roughness layer. The presentation at the 23rd Symposium on Boundary Layers and Turbulence of the AMS will expand on the turbulence statistics and flow field characteristics by among others addressing the frequency spectra of this unique forest canopy.

References

1. Lee, X., Air motion within and above forest vegetation in non-ideal conditions. Forest Ecology and Management, 2000. 135(1-3): p. 3-18.
2. Green, S.R., J. Grace, and N.J. Hutchings, Observations of turbulent air flow in three stands of widely spaced Sitka spruce. Agricultural and Forest Meteorology, 1995. 74(3-4): p. 205-225.
3. Raupach, M.R., J. Finnigan, and Y. Brunet, Coherent Eddies and Turbulence in Vegetation Canopies: The Mixing-Layer Analogy. Garratt J.R., Taylor P.A. (eds) Boundary-Layer Meteorology 25th Anniversary Volume, 1970–1995, 1996.
4. Finnigan, J., Turbulence in Plant Canopies. Annual Review of Fluid Mechanics, 2000. 32(1): p. 519-571.
5. Poggi, D., et al., The Effect of Vegetation Density on Canopy Sub-Layer Turbulence. Boundary-Layer Meteorology, 2004. 111(3): p. 565-587.
6. Foken, T., Micrometeorology. 2008: Springer-Verlag Berlin Heidelberg.
7. Gao, W., R.H. Shaw, and K.T. Paw U, Observation of organized structure in turbulent flow within and above a forest canopy. Boundary-Layer Meteorology, 1989. 47(1-4): p. 349-377.
8. World Meteorological Organization: Guide to Meteorological Instruments and Methods of Observations. 2012.