PIV Measurements of Atmospheric Turbulence above and within a Corn Canopy

Turbulence characteristics of the flow within and above a pollinating corn canopy were measured using Particle Image Velocimetry (PIV). The measurements were performed on the eastern shore of the Chesapeake Bay, in Maryland, during July 2003. Two PIV systems were used simultaneously, each with an overall sample area of 18x18 cm. The spacing between samples was about equal to the field of view. The PIV instrumentation on a rotating measurement platform, that could be extended up to 10m. The flow was seeded with smoke generated about 30m upstream of the sample areas. Data was recorded at 4Hz, providing a time series for the large scales. The local meteorological conditions were monitored by other sensors including sonic anemometers and temperature sensors.

Each processed PIV image provided an instantaneous velocity distribution containing 64x64 vectors with a vector spacing of ~3mm. The mean flow and turbulence characteristics including Reynolds stresses, velocity fluctuations and its higher moments closely followed those presented in the literature for different types of canopies. Triple correlations showed substantial vertical flux. Turbulent spectral densities calculated from PIV data displayed a clear inertial range. At the small scales (0.6–18 cm) the turbulence was isotropic, while at the large scale (1-1000m) the temporal spectral densities from both the sonic anemometers and the PIV data showed anisotropy. From the turbulence spectra the dissipation and production rate at different heights was estimated. Using this dissipation rate, the Kolmogorov length scale was about 0.5mm and the Taylor Microscale was 50mm resulting in a Reynolds number based on the Taylor microscale, Rel ~ 2000. The dissipation and production rates vary substantially, especially close (above and below) the canopy. Conditional sampling and subsequent quadrant analysis of the data show that the ratio of sweep and ejection durations is approximately unity. However the contribution of the stresses in the sweeps consistently exceeds those of the ejections. Data processing is currently underway to investigate the spatial structure of the turbulence for the sweep and ejection events.

This research is funded by the National Science Foundation (NSF).